JPH06189083A - Image direction recognition device and image direction recognition method - Google Patents

Abstract

(57) [Summary] [Purpose] For vertical writing and horizontal writing, even if vertical writing and horizontal writing are mixed, the vertical direction of the image can be accurately identified. A detecting step of detecting an image information size and an image information set direction, a line direction recognizing step of recognizing a line direction of image information, a character direction recognizing step of recognizing a character direction of image information, the size data and An image direction recognition step of recognizing the image direction of the image information based on the set direction data, the line direction data and the character direction data, and these steps are detected by the detection means, the line direction recognition means, the character direction recognition means, the image direction recognition. Each of them is executed by the means to accurately recognize the image direction.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image orientation recognizing device and an image orientation recognizing method in a copying machine, facsimile, printer, OCR and the like.

[0002]

2. Description of the Related Art Conventionally, various document image processing apparatuses have been proposed. Among these, for example, Japanese Patent Laid-Open No. 1-251818
In Japanese Patent No. 4 publication, a character image is cut out from image information of a document read by a scanner, a feature amount of the cut out character image is extracted, and the extracted feature amount is converted according to a rotation angle of the character to change. Recognize rotated characters by comparing and collating later features with a dictionary, and if the rotation angle of a character is not known in advance, character images that have been cut out are recognized as rotated characters with two or more different angles. Has been proposed to determine the rotation angle of a character.

On the other hand, in Japanese Patent Laid-Open No. 1-105266, an appropriate number of character images are vertically cut out from the image information of a document read by an image scanner (from the line head to the line end of the read document image). (The width is cut out so that the width from the end of the line to the beginning of the line is equal), and at this time, the number of black pixels included in the specified width from the beginning of the line to the end of the line is within the specified width from the end of the line to the beginning of the line. By using the property that the number of black pixels is larger than the number of black pixels included in, the top and bottom (vertical relation) of the document image is determined, and when the top and bottom of the document image are determined to be reversed, the image rotation means A technique has been proposed in which an image is rotated to make the top and bottom normal.

[0004]

Here, focusing on the general type of document (document image), there are a vertical document and a horizontal document, and further, a vertical document and a horizontal document respectively. Although there are portrait originals and landscape originals, in the former of the above-mentioned conventional techniques, it is possible to distinguish between vertical writing and horizontal writing, but image orientation determination including vertical and horizontal originals is not performed. . For example, in the case of the same vertical writing document and horizontal writing document, and when both vertical and horizontal documents are mixed,
When a document image is copied by a copying machine or the like, the top and bottom cannot be discriminated, and there is a risk that the copy will be created in the top and bottom unaligned state.

Further, in the latter character direction recognition method of the above-mentioned prior arts, since the top and bottom (vertical relation) of the document image are judged by the number of black pixels in the left and right end regions that are evenly cut out, the top and bottom judgment is made. There is a risk of error. More specifically, for example, in a document image (horizontal writing) whose top and bottom are correct, when a large number of characters such as kanji having a large number of pixels per character exist in the right end area (line end to line head) of the document image,
The edge area having a large number of pixels in the predetermined area is erroneously recognized as the left edge area (from the beginning of the line to the end of the line). In this case, it is determined that the top and bottom of the document image are reversed, and the document image is rotated. As a result, there was a big problem in recognition accuracy.

The present invention solves the above-mentioned drawbacks of the prior art and makes it possible to accurately identify the vertical direction of an image and the horizontal direction of an image even when both vertical and horizontal documents are mixed. An object of the present invention is to provide an image orientation recognition device and an image orientation recognition method with extremely high sex recognition accuracy.

[0007]

The above-mentioned object is to provide a detection means for detecting the image information size and the image information set direction,
A line direction recognition means for recognizing the line direction of the image information, a character direction recognition means for recognizing the character direction of the image information, and an image direction of the image information based on the size data and set direction data, line direction data and character direction data. Is achieved by the first means including the image direction recognition means for recognizing.

Further, the above-mentioned object includes a detection step of detecting the image information size and the image information set direction, a line direction recognition step of recognizing the line direction of the image information, and a character direction recognition step of recognizing the character direction of the image information. An image orientation recognition step of recognizing the image orientation of the image information based on the size data and the set orientation data, the line orientation data and the character orientation data, and the image orientation recognition is performed by executing each step. Achieved by means.

[0009]

According to the present invention, the image data size and the image data set direction are detected, the line direction of the image data is recognized, and the character direction of the image data is recognized. And the image direction of the image information is recognized based on the character direction data.

[0010]

Embodiments of the present invention will be described with reference to the drawings.
In addition, in order to make it easy to understand, the explanation is divided into each item, and the divided items are shown as a table of contents first.

1. Schematic configuration of a digital copying machine according to an embodiment 1.1 Overall configuration 1.2 Scanner section 1.3 Writing section 1.4 Photoconductor section 1.5 Developing section 1.6 Paper feeding section 1.7 Automatic document feeder ( ADF) 1.8 Sorter stapler (III) 1.9 Electrical equipment control section 1.9.1 Sequence control 1.9.2 Image data processing 1.9.3 Application unit 1.9.3.1 About APL1 1. 9.3.2 About APL2 1.9.3.3 About APL3 1.9.3.4 About APL4 1.9.3.5 About APL5 1.9.3.6 About display 1.9.4 Fax operation 1.9.5 Image processing unit 1.9.5.1 Shift, scaling, rotation, reverse scan, mirroring 1.9.5.2 Shift 1.9.5.3 Magnification 1.9.5.4 Rotation 1.9. 5.5 Reverse scan and mirroring 1.10 Human body detection sensor 2. Image orientation judgment 3. Image direction recognition detection based on character direction determination 3.1.1 Image direction recognition within output image information page based on character recognition 3.1.2 When a large number of documents are printed, reference output image information based on character recognition is replaced with other image information Matching of the image directionality (unification of image direction) 3.1.3 Method of distinguishing between vertical document and landscape document [Identification detection based on document size / direction data, character direction data, and line direction data (all types: (Vertical writing / horizontal writing can be supported)] 3.1.4 When multiple sheets / vertical / horizontal originals are mixed, image orientation matching check based on character direction recognition against predetermined reference image information 3.1.5 Multiple originals 3) Recover when image directionality NG is detected (mismatch in character direction) 3.1.6 Correspond when image directionality NG is detected when multiple sheets of originals (when image processing cannot be supported) 3.1.7 Blank sheet originals 3.1 . 8 Correspondence when identification is not possible 1.8.1 Image formation / warning display in predetermined predetermined direction 3.1.8.2 Image formation / warning display unified to identified predetermined reference image information direction 3.1.8. 3 Image interruption / warning display 3.2.1 Staple position determination by character recognition 3.2.2 Comparison of character string direction data and standard staple position 32.3 Predetermined standard when multiple sheets / horizontal and horizontal originals are mixed The image information direction is identified from the character direction recognition data for the image information, and the matching of the image direction is confirmed by comparison with the reference stapler position data. 3.2.4 Consistency between the character string direction data and the reference staple position Recover when NG 3 3.1.1 Selection control of image direction identification detection according to image forming mode 3.3.2 When manual stapling instruction is input after completion of sorting operation, staple position N based on character direction data
In case of G 3.3.2.1 Warning display 3.3.2.2 Staple prohibition 1. 1. Schematic Configuration of Digital Copier according to Embodiment 1.1 Overall Configuration FIG. 1 is an overall configuration diagram of the digital copier, and FIGS. 2 and 3 are a plan view and a side view of a writing unit in the digital copier.

First, a schematic structure of a digital copying machine will be described with reference to FIG. As shown in the figure, the digital copying machine includes a copying machine body (I) and an automatic document feeder [ADF].
II, a sorter with a stapler, a sorter stapler (III), and a double-sided reversing unit (IV). The copying machine body (I)
Includes a scanner unit, a writing unit, a photoconductor unit, a developing unit, a paper feeding unit, and the like. The configuration and operation of each unit will be described below.

1.2 Scanner Unit The scanner unit is equipped with the first mirror that is equipped with the reflecting mirror 1, the light source 3, and the first mirror 2 and moves at a constant speed, and the second mirror 4 and the third mirror 5. Then, it has a second scanner that moves following the first scanner at half the speed of the first scanner. An original document (not shown) on the contact glass 9 is optically scanned by the first scanner and the second scanner, and the reflected image is guided to the lens 7 through the color filter 6 and then on the one-dimensional solid-state image sensor 8. Image.

A fluorescent lamp, a halogen lamp or the like is used as the light source 3, and a fluorescent lamp is generally used because of its stable wavelength and long life. In this embodiment, the reflecting mirror 1 is attached to one light source 3, but two or more light sources 3 may be used. Since the solid-state image sensor 8 has a constant sampling clock, the fluorescent lamp will adversely affect the image unless it is lit at a higher frequency.

Generally, the solid-state image pickup device 8 is C
CD is used. Since the image signal read by the solid-state image sensor 8 is an analog value, it is analog / digital (A / D) converted, and various image processings (binarization, multi-value processing, gradation processing, Magnification processing, editing processing, etc.) is performed and converted into a digital signal as a set of spots. In order to obtain color image information, in the present embodiment, a color filter 6 that transmits only necessary color information is placed in and out of the optical path guided from the original to the solid-state image sensor 8. Color filter 6 according to the scanning of the document
It is possible to create a wide variety of copies by taking out and putting in and out, and each time, functions such as multiple transfer and double-sided copy are activated. In addition, R (red), G (green), B
In some cases, a color original is read using a 3-line CCD or the like in order to simultaneously obtain three (blue) information.

1.3 Writing Unit Image information after image processing is written on the photoconductor drum 40 in the form of a set of light spots by raster scanning of laser light in the optical writing unit. 2 and 3 are a plan view and a side view showing the writing unit. The laser light emitted from the semiconductor laser 20 is converted into a parallel light flux by the collimator lens 21, and is shaped into a constant light flux by the aperture 32. The shaped laser light enters the polygon mirror 24 in a form compressed by the cylinder lens 22 in the sub-scanning direction. The polygon mirror 24 has an accurate polygonal shape and is rotated by a polygon motor 25 in a fixed direction at a fixed speed. This rotation speed is the rotation speed of the photosensitive drum 40, the writing density, and the polygon mirror 24.
It is determined by the number of faces. The reflected light of the laser light incident on the polygon mirror 24 is deflected by the rotation of the polygon mirror 24. The deflected laser light sequentially enters the fθ lenses 26a and 26b. fθ lens 26
a and 26b are converted so that scanning light with a constant angle is scanned on the photosensitive drum 40 at a constant speed, and are imaged so as to be the minimum light spot on the photosensitive drum 40. Have

The laser light that has passed through the fθ lenses 26a and 26b is guided to the synchronization detection light input section (synchronization detection plate) 30 by the synchronization detection mirror 29 outside the image area, propagated to the sensor section by the optical fiber, and then in the main scanning direction. Sync detection is performed as a reference for cueing and a sync signal is output. Image data for one line is output after a fixed time from the output of the synchronization signal, and one image is formed by repeating this process.

In FIG. 2, 27 is a mirror and 31 is a mirror.
Is a lens holding unit.

1.4 Photosensitive Member A photosensitive layer is formed on the peripheral surface of the photosensitive drum 40.
As a photosensitive layer having sensitivity to a semiconductor laser (wavelength 780 nm), an organic photoconductor (OPC), α-Si, Se-T
e and the like are known, and the organic photoreceptor (OPC) is used in this embodiment. Generally, in the case of laser writing, there are two types, a negative / positive (N / P) process of shining light on the image part and a positive / positive (P / P) process of shining light on the background part. N / P process is adopted.

The charging charger 41 is of the scorotron type having a grid on the side of the photoconductor, and is a photoconductor drum 40.
The surface of (1) is uniformly (-) charged, and the image forming portion is irradiated with laser light to reduce the potential of that portion. Then, the background portion of the surface of the photosensitive drum 40 has a potential of about −750 to −800 V and the image portion has a potential of about −500 V, and an electrostatic latent image is formed on the surface of the photosensitive drum 40. This is the developing device 42a, 4
In 2b, a bias voltage of -500 to -600 V is applied to the developing roller, and toner charged in (-) is attached to visualize the electrostatic latent image.

1.5 Developing Unit The apparatus of the present embodiment includes two developing devices, a main developing device 42a and a sub developing device 42b. In the case of solid black, the sub-developing device 42b and the toner replenishing device 43b are removed. In this embodiment having two developing devices, the main developing device 4
By inserting black toner into the toner replenishing device 43a paired with 2a and color toner into the toner replenishing device 43b paired with the sub-developing device 42b, the main pole position of the developing device of the other color is developed during development of one color. Selective development is performed by changing.

By using such a developing device, color information can be read by switching the color filter 6 of the scanner, and the multi-transfer and double-sided copying functions of the paper carrying system can be combined to perform multifunctional color copying and color editing. Becomes Development of three or more colors can be achieved by a method of arranging three or more developing devices around the photosensitive drum 40, a revolver method of rotating and switching three or more developing devices, and the like.

The image visualized by the developing devices 42a and 42b is transferred onto the surface of the paper synchronously sent to the photoconductor drum 40 from the back surface of the paper by a (+) charge by the transfer charger 44. It The transferred paper is subjected to AC charge removal by a separation charger 45 which is held integrally with the transfer charger 44, and separated from the photosensitive drum 40. The toner remaining on the photosensitive drum 40 without being transferred to the paper is scraped off from the photosensitive drum 40 by the cleaning blade 47 and collected in the attached tank 48. Further, the potential pattern remaining on the photoconductor drum 40 is erased by irradiating it with light from a charge eliminating lamp.

A photo sensor 50 is provided at a position immediately after the development. The photo sensor 50 is composed of a pair of a light emitting element and a light receiving element, and detects the reflection density on the surface of the photosensitive drum 40. This is a fixed pattern in the optical writing part (eg black or halftone dot pattern)
Is written in a position corresponding to the photosensor reading position, and the image density is determined from the ratio of the reflectance of the pattern portion after development and the reflectance of the photosensitive drum 40 other than the pattern portion. Give a signal. In addition, it is possible to detect that the remaining amount of toner is insufficient by utilizing the fact that the density does not increase even after replenishment.

1.6 Paper Feeding Unit In this embodiment, a plurality of cassettes 60a, 60b, 60c are provided, and the paper once transferred can be passed through the re-feeding loop 72 to perform double-sided copying or re-feeding.

When one of the cassettes 60a, 60b and 60c is selected and then the start button is pressed, the paper feed rollers 61 (61a, 61b, 61c) near the selected cassette 60 are selected. ) Rotates and is fed until the leading edge of the paper hits the registration roller 62. Although the registration roller 62 is stopped at this time, the registration roller 62 starts rotating at the timing of the image position formed on the photoconductor drum 40, and feeds the paper to the peripheral surface of the photoconductor drum 40. After that, the toner image is transferred to the paper at the transfer unit, suctioned and conveyed by the separating and conveying unit 63, and the transferred toner image is transferred onto the paper surface by the fixing roller composed of a pair of the heat roller 64 and the pressure roller 65. Establish.

The paper thus transferred is guided by the switching claw 67 to the paper discharge port on the sorter (III) side during normal copying. On the other hand, during multiple copying, the switching claws 68, 69
The direction is changed by and the sheet passes through the lower sheet re-feeding loop 72 without being discharged to the sorter (III) side, and is guided to the registration roller 62 again.

The case of double-sided copying will be described. The paper is guided downward by the switching claw 67, and is guided by the next switching claw 69 to the tray 70 further below the refeed loop 72. Then, by reversing the roller 71, it is fed again in the opposite direction, and the switching pawl 6
By switching 9, the sheet is guided to the sheet re-feeding loop 72 and fed to the registration roller 62.

1.7 Automatic Document Feeder (ADF) The original placed on the original table 100 is called by the calling roller 104. The called originals are pulled out rollers 105 and 106 that are pressed against each other, and a separation belt 1 that is wound around the pullout rollers 105.
Double feed is prevented by the action of 07, and the single guide plate 10
Sent along 8. The original fed along the guide plate 108 is fed by the belt conveying device 125 to the contact glass 9
Is sent to a predetermined exposure position and stopped.

The belt conveying device 125 is a drive roller 10.
9 and the driven roller 110, the fixing roller 111 sets the insertion position of the document, and the pressure roller presses the document against the contact glass 9.
Have. Hereinafter, description of known operations will be omitted.

1.8 Sorter Stapler (III) Entrance guide plates 1101 and 1102 are provided at the receiving port A on the copy ejected from the copying machine, and for conveying the copy following the entrance guide plates 1101 and 1102. A switching claw 1103 is provided. The path above the switching claw 1103 has an inlet guide plate 1101 and a guide plate 111.
0, 1114, transport roller 1108, driven roller 1109,
The upper transfer unit 1100 is provided with a discharge roller 1111, a driven roller 1115, and a proof tray 1116. Further, the path below the switching claw 1103 is a slanting portion guide plate 1205, a slanting portion driven guide plate 1217, a lower conveyance portion guide plate 1308, and driven guide plates 1309 and 131.
0, oblique receiving roller 1201, oblique roller 1202,
Oblique portion discharge roller 1203, driven rollers 1214, 121
6, a sphere 1215, conveyance rollers 1301 and 1302, driven rollers 1305 and 1306, and an inclined portion 1200 that follows the path of the deflection portion B.

A deflecting claw 1312 and a deflecting portion discharging roller 130 are provided at positions corresponding to the respective bins 1350 of the deflecting portion B path.
4 are provided respectively, and the driven roller 1307 is in pressure contact with the deflecting portion discharge roller 1304 with the copy vertical conveyance path interposed therebetween. The conveying roller 1108 and the discharging roller 1111 are driven by a proof motor 1117, and the oblique receiving roller 1201, the oblique roller 1202, and the oblique discharging roller 1 are provided.
The drive motor 1313 drives 203, the conveyance rollers 1301 and 1302, and the deflection portion discharge roller 1304.

The description of the post-processing unit such as the stapler mechanism and the punch mechanism and the double-sided reversing unit will be omitted. Also, 46 is a separating claw, 80 is a main motor,
81 is a fan motor.

1.9 Electric Control Section FIGS. 4A and 4B are block diagrams showing the control unit of the copying machine, and FIGS. 5A and 5B are control block diagrams of the entire copying system. In FIG. 4, the control unit has two CPUs. The CPU (a) 200 controls the sequence and the CPU (b) 201 controls the operation. CPU (a) 200 and C
PU (b) 201 is a serial interface (R
S232C). 4, 202 is an image control circuit, 203 is a signal switching gate array, 204 is an operation unit, 205 is an editor, 206 is a scanner control circuit, 207 is a page memory, 208 is an image processing unit, and 209 is a calendar I.
C and 210 are application systems, and 211 is a laser beam scanner unit.

In FIG. 5, the same parts as those described above are designated by the same reference numerals, 220 is a main control plate, 221 is a paper feed control plate, 222 is a sorter control plate, 223 is a double-sided control plate, 22
Reference numeral 4 is an ADF control plate.

1.9.1 Sequence Control First, sequence control will be described. In this sequence, the conditions for paper conveyance timing and image forming conditions are set and output.The paper size sensor, paper conveyance sensors such as paper discharge detection and registration detection, duplex unit,
A driver such as a high-voltage power supply unit, a relay, a solenoid, a motor, a sorter unit, a laser unit, a scanner unit, etc. are connected.

Regarding the sensor, a paper size sensor that detects the size and direction of the paper loaded in the paper feed cassette and outputs an electric signal according to the detection result, a sensor related to paper conveyance such as registration detection and paper discharge detection, an oil end, There are inputs from sensors that detect the presence or absence of supplies such as toner end, and sensors that detect machine abnormalities such as door open and fuse blown.

In the double-sided unit, a motor for aligning the paper width, a paper feed clutch, a solenoid for changing the transport path, a paper presence / absence detection sensor, a side fence home position sensor for aligning the paper width, a paper There are sensors related to the conveyance of the.

The high-voltage power supply unit applies predetermined high-voltage power to the outputs of the charging charger, the transfer charger, the separation charger, and the developing bias electrode only by the duty obtained by the PWM control. The PWM control is controlled so that the feedback value of each high-voltage power output is converted into a digital value by A / D conversion and becomes equal to the target value. Drivers include a paper feed clutch, registration clutch, counter, motor, toner replenishment solenoid, power relay, and fixing heater. It is connected to the sorter unit via a serial interface, and the paper is conveyed at a predetermined timing by a signal from the sequence and discharged to each bin.

To the analog input, a fixing temperature, a photosensor input, a laser diode monitor input, a laser diode reference voltage, a feedback value of an output value from various high voltage power supplies, and the like are input. On / off control or phase control of the heater is performed by an input from a thermistor in the fixing unit so that the temperature of the fixing unit becomes constant. As a photo sensor input, a photo pattern formed at a predetermined timing is input by a photo transistor, and by detecting the density of the pattern, the toner replenishment clutch is turned on / off to control the toner density. The toner end is also detected based on this density.

The analog input of the A / D converter and the CPU is used as a mechanism for adjusting the power of the laser diode to be constant. This is controlled so that a preset reference voltage (this voltage is a monitor voltage when the laser diode is turned on in this embodiment) is matched.

Next, the operation-related control will be described. The main CPU (b) 201 controls a plurality of serial ports and the calendar IC 209. In addition to the sequence control CPU (a) 200, the operation unit unit 204, the scanner control circuit 206,
The application system 210, the editor 205, etc. are connected. The operation unit unit 204 has an indicator for displaying the operator's key input and the state of the copying machine,
Information of key input is notified to the main CPU (b) 201 by serial communication. Based on this information, the main CPU (b) 201 determines whether the indicator of the operation unit unit 204 is turned on, turned off, or blinks, and serially transmits it to the operation unit unit 204. The operation unit CPU turns on, off, and blinks the display according to the information from the main CPU (b) 201. Further, the CPU that performs the sequence control at the start of copying by determining the operating conditions of the machine from the obtained information.
(A) The information is transmitted to 200.

In the scanner section (scanner control circuit 206), information relating to scanner servo motor drive control, image processing, and image reading is serially transmitted to the main CPU (b) 201, and ADF (ADF control board 224).
And the interface processing of the main CPU (b) 201 is performed.

The application (application system 210) is an external device (fax, printer,
Etc.) and an interface between the main CPU (b) 201, and exchanges preset information contents. The editor 205 is a unit for inputting an editing function, and stores image editing data (masking, trimming, image shift, etc.) input by the operator in the main CPU (b).
Serial transmission to 201. Calendar IC209 is
It stores the date and time, and the main CPU (b) 201
Since it can be called at any time, it is possible to control the power on / off of the machine by a timer by displaying the current time on the operation unit display and setting the on / off time of the machine.

1.9.2 Image Data Processing Next, the flow of image data processing will be described. The gate array 203 receives image data (DATA0 to DATA0) in the following three directions in response to a select signal from the CPU (b) 201.
7) and a sync signal are output.

1) Scanner control circuit 206 → image control circuit 202 In this case, the image signal continuously transmitted by 8-bit data from the scanner (however, it can be 4 bits or 1 bit) is synchronized by the laser beam scanner unit 211. Signal PM
Output to the image control circuit in synchronization with SYNC.

2) Scanner control circuit 206 → application 210 In this case, an image signal continuously sent from the scanner as 8-bit data (however, it can be 4 bits or 1 bit) is output in parallel to the application 210. The application 210 outputs the input image data to an output device such as a printer connected to the outside.

3) Application 210 → image control circuit 202 In this case, the application 210 is continuously transmitted by 8-bit data (however, it can be 4 bits or 1 bit) from an externally connected input device (fax or the like). The image signal is synchronized with the synchronization signal PMSYNC from the laser beam scanner unit 211 and output to the image control circuit.
In this case, when the image signal from the outside is 1 bit or 4 bits, it is necessary to perform a process of converting the image signal into 8 bit data.

FIG. 6 is a block diagram of the image scanner section. An analog image signal output from the CCD image sensor 250 is an image preprocessor (IPP) 2
The signal processing circuit 252 inside 51 amplifies and corrects the amount of light, and the A / D converter 253 converts the signal into a digital multilevel signal. This signal is subjected to correction processing by the shading correction circuit 254, and the image processing unit (I
PU) 255.

Image process unit (IPU) 25
A schematic block diagram of No. 5 is shown in FIG. The image signal applied to the IPU 255 is high-frequency emphasized by the MTF correction circuit 270, electrically scaled by the scaling circuit 271, and the γ conversion circuit 272.
Applied to. The γ conversion circuit 272 optimizes the input characteristics according to the characteristics of the machine. γ conversion circuit 272
The image signal output from is converted into a predetermined quantization level by SW1 of the data depth switching mechanism. This switching mechanism switches to the three data types shown in FIG.
The 4-bit conversion circuit 273 outputs 4-bit data,
The binarization circuit 274 converts the input 8-bit multi-valued data into binary data with a preset fixed threshold value, and outputs 1-bit data. Dither circuit 275
Is 1-bit data and creates an area gradation. SW1 is 3
DATA0-DATA by selecting one of the two data types
Output as 7.

Returning to FIG. 6 again, the scanner control circuit 206
Controls a fluorescent lamp ballast (lamp control circuit) 256, a timing control circuit 257, an electric scaling circuit of the IPU 255, and a scanner drive motor 258 in accordance with an instruction from the main CPU (b) 201. Fluorescent ballast 256
According to the instruction from the scanner control circuit 206, the fluorescent lamp 1 is turned on / off and the light amount is controlled. Scanner drive motor 2
A rotary encoder 259 is connected to the drive shaft of 58, and the position sensor 260 detects the reference position of the sub-scanning drive mechanism. The electric scaling circuit is the scanner control circuit 20.
The electrical scaling processing is performed according to the magnification on the main scanning side set by 6.

The timing control circuit 257 outputs each signal according to the instruction from the scanner control circuit 206. That is, when reading is started, the CCD image sensor (C
The CD) 250 is supplied with a transfer signal for transferring data for one line to the shift register and a shift clock pulse for outputting the data in the shift register bit by bit. The pixel synchronization clock pulse CLK, the main scanning synchronization pulse LSYNC, and the main scanning effective period signal LGATE are output to the image reproduction system control unit.

This pixel synchronizing clock pulse CLK is C
The signal is almost the same as the shift clock pulse given to the CD image sensor 250. Also, the main scanning synchronization pulse L
SYNC is almost the same signal as the main scanning synchronization signal PMSYNC output by the beam sensor of the image writing unit, but is output in synchronization with the pixel synchronization clock pulse CLK. The main scanning effective period signal LGATE becomes the high level H at the timing when it is considered that the output data DATA0 to DATA7 are effective images.

In this example, the CCD image sensor 2
50 outputs valid data of 4800 bits per line. The scanner control circuit 206 is the main CPU
(B) Upon receiving a reading start instruction from 201, the fluorescent lamp for illumination 1 is turned on, the scanner driving motor 258 is started, the timing control circuit 257 is controlled, and the reading of the CCD image sensor 250 is started. Further, the sub-scanning effective period signal FGATE is set to the high level H.
This signal FGATE becomes low level L after the time required to scan the maximum reading length (dimension in the A size longitudinal direction in this example) in the sub-scanning direction after being set to high level H.

FIG. 9 is a block diagram of the memory system of this apparatus. The image signal from the CCD 250 is output as 8-bit data through an image preprocessor (IPP) 251 having functions of shading correction, black level correction, and light amount correction. This data is a multiplexer (1)
Selected by (MUX1) 280, has spatial frequency high-frequency emphasis (MTF correction) function, speed conversion function (magnification), γ conversion function, data depth conversion function (8 bit / 4 bit / 1 bit conversion) Image process unit (IPU) 2
It is processed at 55 and output to the printer PR through the MUX (3) 282. Reference numeral 281 is a MUX (2), and 283 is a memory device (MEM).

In a system having a frame memory for image data, the image data from the IPU 255 is temporarily stored in the memory device (MEM) 283 as shown in FIG.
It is configured to take out from the memory device (MEM) 283 and output to the printer (PR) when necessary. Also,
Image data from IPU255 to printer (PR)
To the memory device (MEM) 283 while simultaneously outputting the same to the memory device (MEM) 283 and outputting the second and subsequent copies.
The method using image data from 83 was also common.

The present apparatus is configured so that the data flow shown in FIG. 11 is possible so that both the processed data and the raw data from the IPU 255 can be taken into the memory device 283. That is, the three multiplexers (MUX1, MUX2, MUX3) 280, 281, 2 of FIG.
The data flow can be changed by switching 82. For example, a single scan of the scanner may result in multiple I
When outputting a copy with the parameters of the PU 255 changed, this can be achieved by the following procedure. Set MUX (1) to A when scanning the scanner, and
(2) is set to B, MUX (3) is set to A, and the first sheet is output. At this time, the raw data enters the memory device (MEM) 283 through the MUX (2). For the second and subsequent sheets, the MUX (1) is set to B, the data from the memory device (MEM) 283 is put into the IPU 255, and is output to the printer (PR) through the MUX (3). At this time, the IPU parameter can be changed every time one copy is made.

When holding compact data such as 1-bit data, set MUX (2) to A and set I to I.
The output of the PU 255 is loaded into the memory device. In this case, the printer device switches to the binary data (1 bit) mode for copying. EXTIN and EXTOUT in FIG. 9 are an image data input signal from the outside and an output signal to the outside.

In FIG. 12, a compressor (COMP) 290 and an expander (EXP) 291 are arranged in a memory unit (Memory).
(Unit) 292 is inserted before and after so that compressed data other than the actual data can be stored. In this configuration, the compressor (COMP) 290 and the decompressor (EXP) 291 need to operate according to the speed of the scanner and the printer, respectively. Multiplexers MUX (4) 293 and MUX for storing actual data
(5) Set 294 to A and B to use compressed data. 295 is an error detector.

The memory unit 292 shown in FIG.
It has a configuration like 3. Data width converters 300 and 301 are provided at the input and output of a memory block (Memory Block) 302 in order to handle the three image data types of FIG. 14 and code data that is compressed data. Direct memory controller (DMC1, DM
C2) 303 and 304 write data to a predetermined address of the memory block 302 according to the number of packed data and the memory data width, and perform a read operation.

FIG. 14 shows the data type of image data as described above. Usually from a scanner,
Alternatively, the speed of image data to the printer is constant regardless of 8-bit data, 4-bit data, or 1-bit data. That is, the period of 1 pixel is fixed in the device. In this device, M of 8 data lines
It is defined as 1-bit data, 4-bit data, 8-bit data and MSB-filled from the SB side. The blocks that pack and unpack this data into the data width (16 bits) of the memory block are the input data width converter and the output data width converter. By packing, the memory can be used according to the data depth, and the memory device can be effectively used.

FIG. 15 shows a pixel process unit (PPU) 310 arranged outside the memory unit 292 instead of the compressor (COMP) 290 and the decompressor (EXP) 291. The function of the PPU 310 is to perform logical operations on image data (for example, AND, OR, EXOR, N
The unit realizing OT) can calculate the memory output data and the input data and output the same to the printer, and the memory output and the input data (for example, scan data) can be calculated and stored in the memory unit 292 again. Switching between output destination printer and memory unit 292 is MUX
(6) 311 and MUX (7) 312. This function is generally used for image synthesis, for example, the memory unit 2
It is used for putting overlay data on 92 and overlaying the scanner data.

FIG. 16 shows a configuration for storing image data using an external storage device. When saving image data to a floppy disk
A file controller (File Control) from XTOUT through an interface (I / F) 320.
output to the floppy disk controller (FDC) 322 controlled by the L. er) 321 and stored in the floppy disk on the floppy disk drive (FDD) 323. Under the control of the file controller 321, there are a hard disk controller (HDC) 324 and a hard disk drive (HDD) 325, which can be read and written on the storage medium of the hard disk. The hard disk drive (HDD) 325 stores format data and overlay data that are often used and can be used as needed.

FIG. 17 shows a configuration in which 100% recovery can be performed when the processing speeds of compression and expansion cannot be met in time. The memory unit 292 receives the compressed data and image data at the same time as the scanner scans. The incoming data is stored in different memory areas, but the compressed data is directly input to the decompressor (EXP) 291 and decompressed. By the time all the data of one page is stored in the memory unit 292, the compressor (CO
When the processing time of the MP) 290 and the decompressor (EXP) 291 ends normally in time, only the memory area of compressed data remains and the area of raw data is canceled. if,
An error detection circuit (Error Detect) 295 is a compressor (COMP) 290 or an expander (EXP) 291.
When an error signal from is detected, the compressed data area is immediately canceled and raw data is adopted.

The memory management unit (MMU) 330 is
This is a unit for controlling the memory so that two input data and one output data can be simultaneously input / output to / from the memory unit 292. By testing the compression and decompression in real time, high speed and reliability and effective use of the memory area became possible. In this embodiment, the memory management unit (MMU) 330 enables dynamic allocation of the memory area.
There may be two memory units for raw data and compressed data.

The configuration of FIG. 17 is most suitable for applications such as electronic sorting where a plurality of pages are stored and output to a printer in real time, in which both the number of stored pages and the printing speed must be compatible.

1.9.3 Application Unit A block diagram of the application unit will be described with reference to FIGS. 18 (a) and 18 (b). This example is APL1 (file unit), APL2 (FAX unit), APL3
(ON / OFF printer unit), APL4 (LA
N), APL5 (image direction recognition unit), display (T /
S, LCD). First, the base part will be described. Since the engine I / F 340 sends the image data serially, the image data is converted into parallel here. Further, the parallel data in the page memory 341 is converted into serial data by the engine I / F 340 and sent to EXTIN. The control signal is serial and the engine I / F3
40 to the system bus via SCI (serial communication interface) 342. In this example, the page memory 341 has a size of one page of A3 and converts it into a BIT image, and also arbitrates the data speed of EXTIN and EXTOUT and the processing speed of the CPU. The scaling circuit 343 is a page memory 34.
In order to perform high-speed processing for enlarging or reducing the data on 1 in this circuit, the DMAC 344 is used to perform high-speed processing without going through the CPU 352. Rotation control is, for example, FAX transmission for A4 portrait and A
In the case of 4 landscapes, the sending side automatically reduces the size by 71% and then sends it, which makes the receiving side difficult to see. In order to prevent this, when the size is the above, the transmission original is rotated by 90 degrees to be converted into A4 landscape and transmitted at the same size.

Another object is that, when receiving and outputting, when the receiving size is A4 horizontal and the cassette size is A4 vertical, the rotation control unit rotates the output image by 90 degrees and converts it to A4 vertical and outputs. This eliminates the need to distinguish between vertical and horizontal cassettes.

The CEP 345 is a circuit having functions of image data compression, expansion and through. Bus arbiter 3
Reference numeral 46 performs processing of sending data from the AGDC 385 to the image bus and sending to the system bus. Timer 348
Has a function of generating a predetermined clock. RTC34
Reference numeral 9 is a clock that generates the current time. The console is a control terminal, and in addition to reading and rewriting data inside the system, the console can also be used to develop software using a debug tool that is a function of the internal OS. The ROM 350 has basic functions such as an OS. The RAM 351 is mainly used for working. This unit performs basic control of this system.

1.9.3.1 About APL1 SCSI 360 is HDD (hard disk) 325, O
An I / F for a DD (optical disk) 361 and an FDD (floppy disk) 323. ROM362 is SCSI
HDD 325, ODD361, FDD3 via 360
It contains software for controlling 23 and filing system.

1.9.3.2 About APL2 This is a unit for FAX control and includes the following parts. G4
The FAX controller 370 is a unit that controls the protocol for G4, and this unit is a unit that supports class 1, class 2, and class 3 of G4. Needless to say, ISDN is also supported, and 2 in NET64.
B + 1D (64KBx2 + 16KB) line, so G4 / G4, G4 / G3, G3 / G3, G4 only, G3
Only one of them can be selected. G3FA
The X controller 371 is a unit that controls the protocol for G3, and in this part, the G3F by analog line is used.
It also has a AX protocol, a modem that converts digital signals to analog signals. An NCU (network control unit) 372 has a function of dialing when receiving a call from the other party when connecting to the other party using an exchange. SAF (Store and Forward) 373
Stores image data (including image data, code data, etc.) when transmitting and receiving FAX.

This unit is a semiconductor memory or HDD 3
25, ODD361, etc. are used. The ROM 374 contains a program for controlling the APL2. Further, the RAM 375 is used for those works and at the same time is made non-volatile by a battery, and the other party's telephone number, the other party's name, data for controlling the FAX function, etc. are stored therein, and the T / S of the display unit, It can be easily set using the LCD.

1.9.3.3 APL3 This is a control unit for an online printer and an offline printer. FDC (floppy disk controller)
Reference numeral 380 controls the floppy disk 381. The recent floppy supports SCSI,
Here, the SCSI and ST506 interfaces are supported. An SCI (Serial Communication Interface) 382 is used to connect to a HOST computer. The Centro I / F 383 is also similar to the SCI 382. The emulation card 384 performs the following functions. When looking at the printer from HOST, the current status is NEC
It is sold by many manufacturers such as EPSON, and the specifications of each are slightly different. If the functions of these printers are not the same as seen from the HOST, the HO
The software used in ST may stop running.
In order to eliminate such a problem, an emulation card 384 is attached so that the software contained therein can operate as a printer of each manufacturer when viewed from the HOST.

An AGDC (Advanced Graphics Display Controller) 385 stores the code data sent from the HOST in the CGROM 386 and the CG card 3.
FONT image in 87 page memory 341 at high speed
It will be deployed to. The ROM 388 contains software for controlling these. CGROM (Character Generator ROM) 389 is an FO corresponding to code data.
It contains NT data. The FONT format contains data such as outline FONT. The CG card 387 is an external CGFONT, and the contents are CG
It is similar to the ROM 386. Reference numeral 395 is a RAM.

1.9.3.4 About APL4 This is a unit for controlling the LAN. Here, the LAN controller 390 controls the currently operating LANs such as the Ethernet, omni, and star run. Naturally A
PL2 (FAX) and APL4 (LAN) work in the background even when other APLs are operating. Three
Reference numeral 91 is a CPU.

1.9.3.5 About APL5 This is a unit for identifying the image direction of the image read by the scanner. Reference numeral 400 is a page memory, 401 is an area memory, 402 is a database, 403 is a CPU, 404 is a ROM, and 405 is a RAM.

1.9.3.6 Display With this unit, the LCD 410 and the touch switch (T /
S) is controlled. The LCD 410 can display graphics and characters, and the CG 412 therein has ANK and kanji second-level codes built therein. The TSC 413 is a touch switch controller, which controls T / S. The T / S is divided by the grids of X and Y, and the switch size when used by the operator is TSC41.
It can be freely set by determining the number of grids for one key by 3. Further, the LCD 410 and the T / S have a two-layer structure, and the key size and the key frame of the LCD 410 can correspond to each other. Also, 414 is a CP
U, 415 is SCI, 416 is ROM, 417 is RA
M, 418 are LCDC, and 419 is DPRAM.

FIG. 19 shows an example of the display. As fixed keys, there are a ten key 430 for setting the number of copies, a start key 431 for starting copy, and user-settable function keys 432, 433, 434. This function key allows the user to freely set the mode. For example, the key 432 is assigned a sort mode, the key 433 is assigned a staple mode, the key 434 is assigned a double-sided mode, and the like. The display is the number of copies display 43
5 and the set number display 436 are fixed displays, and other displays are displayed on the LCD 410. Also, the LCD 410
Is a touch switch, and a mode can be selected by pressing an object displayed on the LCD 410.

Next, the operation of the APL will be described. 1.9.4 Fax Operation First, the fax operation will be described. This FAX is M
It has the functions of F, G2, G3, and G4, and the transmission density is 3.8.
5, 7.7, 15.4 lines / mm and 20 for G4
It supports 0, 240, 300, and 400 dpi, and can perform density conversion with each other using the scaling function.
Also, memory transmission / reception, relay, confidential reception, polling, etc. can be realized by using the SAF memory, and further, memory transmission, memory reception, reception output, etc. can be performed at the same time while storing transmission documents in the memory. Will be described. By setting the original and pressing the start key 431, the destination stored in the RAM 375 of the APL 2 is dialed and the destination is called. When it is known that the other party is a FAX, the document reading operation starts. If the start key 431 is pressed when there is no original, a display prompting the user to reset the original is displayed. By the document reading start operation, the scanner operates to read the document, and the data is output to the terminal of EXTOUT via the individual circuits in FIG. At this time, by selecting MUX (1) and MUX (3), I
You can choose whether to use PU255 or not.
Functions inside the PU 255 can be freely selected by a program. This signal enters the engine I / F 340 of FIG.
The data is stored in the page memory 341 in accordance with the bit size of the page memory 341 (EXTOUT is sent in a multi-value of 8 bits per pixel. On the other hand, page memory 3
41 corresponds to 16 bits, and the bit configuration is different, so match here).

The data from the scanner is the page memory 34.
When entering 1, SA of APL2 while compressing this data
Accumulates in F memory. In this way, by transmitting the data from the scanner while accumulating it in the SAF 373,
The following characteristics are obtained. Reading from the scanner is A4
One size can be read in about 2 seconds. In contrast,
It takes about 9 seconds to send an A4 size G3. Thus, the transmission time is about 4.5 times longer than the reading time. As in this embodiment, a copying machine, FA
In the device that can be used in combination with X, printer, etc.,
For example, if the next person wants to make a copy while sending a fax,
I want to finish the job of fax transmission quickly. However, the FAX transmission may be sent faster or slower depending on the performance of the partner machine. As in this embodiment, the read data is SAF373.
By transmitting while accumulating, it is possible to increase the apparent transmission speed. Further, since the transmitted document is stored in the SAF memory, it is possible to retransmit and recall the image to correctly send an image when an error occurs during transmission, the line is disconnected, or the like. In this way, the data stored in the SAF 373 is transferred to the G3 fax,
Or it can be accessed from the G4 fax unit.

FIG. 21 is a schematic block diagram of the IPU 255 shown in FIG. 20. In addition to the circuit shown in FIG.
40, marker editing circuit 441, outline circuit 44
2. The error spreading circuit 443 and the like are provided.

Next, reception of image information will be described. Figure 2
2 (a) and 2 (b), the received image data is the modem 4
At 50 it is converted to a digital signal. This is DCR45
The raw data is corrected via 1, further compressed and stored in the SAF memory 452. At this time, the reason why the DCR 451 restores the raw data and then compresses it again is that the normal received data includes an error on the line.
This is because if the data is stored in 2, it is impossible to distinguish between a hardware error and a data error. When recompressing, a memory efficient method is used. The data stored in the SAF memory 452 is printed out for each page (it is also possible to output after storing one file by setting the mode).

In order to output from the SAF memory 452, it is necessary that the page memory 341 shown in FIG. 18 is not used by another APL and that a copying machine is available. When these conditions are met, the data in the SAF memory 452 is set to CEP3.
The data is expanded to the page memory while returning to the raw data via 45. Select the optimum paper size after the development is completed. At this time, the data in the page memory 341 is A4 portrait, and when the optimum paper is A4 landscape, the page memory 3 is controlled by rotation control.
The data of 41 is rotated by 90 degrees and is output to the selected paper. With this function, until now A4 landscape paper has been
It is now possible to display vertical images and prevent margins from appearing. This function can rotate not only the received output but also the image information read by the other device by 90 degrees not only in the output mode but also in the transmission mode.
In the case of 4 portraits, it was sent at 71% reduction so far, but by incorporating 90 degree rotation it becomes possible to send at the same size and it becomes easier to see on the receiving side.

Here, instead of the SAF memory 452, H
When the DD 325 is used, it becomes possible by using the SAF memory 452 as a buffer and driving the HDD 325 via the SCSI interface of the APL1. The basic operation of the FAX has been described above.

1.9.5 Image Processing Unit 1.9.5.1 Shift, Magnification , Rotation, Reverse Scan, Mirroring FIG. 23 shows the image processing function of the base portion of FIG. As shown in FIG. 23, the image processing unit 208 can access the bitmap page memory 207. In order to leave the original document image, it is desirable that the bitmap page memory 207 has a memory for two sheets of the maximum document size that can be copied. That is, the original image data of the original is set in A of the bitmap page memory 207 in FIG. 23, and the image data processed by the image processing unit 208 is set in B. The input to the image processing unit 208 is the image data (video) bus 50.
0 and the image processing command 501.

The image data bus 500 is an 8-bit data bus, and has 8-bits for each pixel, that is, 256 density gradations. The image processing command is input through the system bus from a system controller [CPU (b) 201 in FIG. 4] that controls a system not shown in this figure. A code is determined for each function of image processing, and for example, in the case of image shift, a shift code, a shift direction, and a shift dimension are transmitted from the system controller. In addition, the scaling function, the scaling command,
The X-axis scaling factor and the Y-axis scaling factor are sequentially transmitted. Of course, in the case of normal scaling, the X-axis scaling factor code and the Y-axis scaling factor code have the same value. In addition, a rotation angle code is added to the rotation command.

When reverse scanning is performed by the scanner, image mirroring is required. In this case, the mirroring command or the reverse scan command and the reference axis information at the time of mirroring are transmitted.

FIG. 2 is an explanatory diagram of the bitmap memory 207.
4 shows. In the read image data, 1 byte is assigned as an address for each pixel. The 1-byte data is the above-mentioned image density data. The byte size is
297m in the main scanning direction when the maximum document size is A3
m, the sub-scanning direction is 420 mm and the resolution is 400 dpi, the main scanning direction is (297 ÷ 25.4) × 400 = 46
78 bytes (420 / 25.4) x 400 = 66 in the sub-scanning direction
15 bytes Therefore, in order to store the bitmap image for one A3 original, 4678 × 6615 = about 30 Mbytes are required.

Here, for easy understanding, an array is defined for the main scanning direction, the sub scanning direction, and the density gradation of each bitmap, and the main scanning direction is X, the sub scanning direction is Y, and the density gradation is defined. Z and each bit is DIM (X, Y, Z). The point A shown in FIG. 24 is (0, 0, 0), and the point B is (46
78,0,0), and point C is (6615,4678,0).

Next, the operation processing method of the image processing unit 208 in response to a command from the system controller will be described.

1.9.5.2 Shift First, when a shift command is received, the direction to shift and the mm size to shift are transmitted at the same time. The number of dots to be shifted is calculated from the mm size to be shifted. Based on the shift direction data, it is determined whether it is the forward direction in the main scanning X direction (direction in which the value of X in the array increases) or the backward direction, or the forward direction in the sub scanning Y direction or the backward direction. When the above judgment is completed, the image processing operation is started. For example, main scan X
If a shift command of 25.4 mm is received in the forward direction, the number of shift dots is 400 dots. Then, the bitmap data of the image is controlled by the DMA control in the image processing unit, and the bitmap page memory A
To B. When transferred, the following processing is performed. A (X, Y, Z) = B (X + 400, Y, Z) X is processed from 0 to 4678-400, Y is 0
To 6615 are processed. Z is a byte DMA
In this case, since 0 to bit 7 are transferred at the same time, it need not be considered.

The shift amount from the system controller is determined as follows. In FIG. 28, when the image start position is X1, the X coordinate of the punch or staple position is 20 mm.
If the punch hole diameter is 6 mm, and X1> (20 + 6/2) mm, the punch hole and the image do not overlap, so there is no need to shift. When X1 ≦ (20 + 6/2) mm 2, the punch hole and the image overlap. In this case, the image shift amount SFx is SFx = (20 + 6/2) −X1. This value is transmitted after the shift direction X-axis forward shift command. The same applies to staples.

On the other hand, when the image is spilled due to the SFx shift, that is, when the value of X2 in FIG. 28 is smaller than the shifted SFx. At this time, when the shift image is developed on the transfer paper, the image is cut off. The original image must be reduced in order not to cause image cutout.

1.9.5.3 Scaling When a scaling process is received, for example, in the case of 50% reduction, even bytes for both X and Y are thinned and transferred from A to B. In the case of 99%, 0 to 98 of X and Y of A are transferred to B, and then 100 to 198 of X and Y of A are transferred to B. By doing so, one pixel out of 100 pixels is thinned out, resulting in 99% reduction. Similarly 98%
At this time, one pixel out of 50 pixels may be thinned out.

When performing enlargement, if the bits are enlarged as they are, the pixels become rough and become a mosaic shape. Therefore, in order to make this smooth (smoothing), interpolation or a spatial low-pass filter (after two-dimensional Fourier transform) is performed. It is common to perform processing such as high frequency attenuation and inverse Fourier transform). When an image is generated at the above-mentioned punch position and an image is cut off when the image is shifted and the image must be reduced, the magnification is determined as follows. If the original length of the original in the X-axis direction is Lx, then M (magnification) = (Lx− (20 + 6/2) / Lx. To guarantee the image, only reduction in the X-axis direction is required. Normally, in order to keep the aspect ratio of the image constant, it is usual to change the magnification on both the X axis and the Y axis.

1.9.5.5.4 Rotation Rotation angle is transmitted in units of 90 degrees. In the case of 90 ° clockwise rotation, the pixel data at point A becomes the pixel at point B,
The pixel data of the image may be transferred from the memory A to the memory B so that the pixel data at the point B becomes the pixel data at the point (4678, 4678). However, in this state, the number of pixels differs in the main scanning direction and the sub-scanning direction. Therefore, in the case of 90-degree rotation, the pixels in the longitudinal direction (sub-scanning direction) overflow from the memory. However, since it also overflows from the transfer paper, this state may be left as it is when the reduction process is not performed. However, in the case of reduction, a memory for the sub-scanning direction × the sub-scanning direction is required for the B page memory. Is becoming.

When the rotation angle is 180 degrees, the A pixel corresponds to the C pixel. Therefore, the conversion is: A (X, Y, Z) = B (6615-X, 4678-Y,
Z).

1.9.5.5 Reverse Scan and Mirroring With mirroring, it is possible to obtain an image in which the image is mirror-finished with respect to a normal copy. This is useful in the design department as one application of image processing, but the following applications are also possible.

Normally, the scanner reads the document data from the document reference position and develops it in the memory. However, if the document size is known, the document is read from the opposite side of the document reference and developed in the memory. The image information expanded in the memory in the above-described normal scan is shown in FIG. 25 (a, b). The diagonal lines and arrows from the left to the right of the document indicate the main scanning direction. FIG. 26
In the figure, (a) and (b) are diagrams when a document is reverse-scanned and developed in a memory. Pixels scanned first together are (0, 0,
Expanded to 0). When the M line in FIG. 26 is assumed to be a mirror and mirroring is performed on the memory, the result is (c) in FIG. According to the mirror axis information received at the same time as the mirroring command or the reverse scan command is received, for example, if it is Y axis information, the memory of FIG. A (X, Y, Z) = B (4678-X, Y, Z) In the case of mirroring on the X axis, A (X, Y, Z) = B (X, 6615-Y, Z). Originally placed upside down, a document is reversely scanned and mirrored, which is shown in (a) to (c) of FIG.

In the present description, the data is temporarily loaded into the memory and mirrored, but it is also possible to output the M line as a mirror axis simultaneously with reading in real time.
Even in a low-cost device that does not have a page memory, the image can be rotated by 180 degrees by a simple process.

1.10 Human Body Detection Sensor As shown in the overall block diagram of FIG.
A human body detection sensor 225 is connected to the. This is to detect whether the operator of the copier is in front of the copier. The installation location is the operation section of the copier,
Or near the display unit, the document table, and the front side surface. The sensor 225 is a reflective sensor that is combined with an infrared light emitting diode and a phototransistor. This sensor 22
5 is used, for example, to turn on / off the control of preheating of the copying machine (which lowers the fixing temperature to save power when the machine is not in use), guidance explanation control, voice guidance on / off, and determination of machine operation control. Be seen.

The determination as to whether or not there is an operator is made as follows. The reflection type human body detection sensor 225 is configured such that the reflection object is turned on within about 1 m from the sensor. When the output of the sensor 225 is directly used as an operator presence signal, even when a person passes in front of the copying machine,
Since the operator instantaneously detects the presence of the operator, if the sensor output is continuously on for a certain period of time, it is determined that the operator is present. This certain time is generally 5
It is from 00 msec to 800 msec. Further, this time may be made as software by a timer by the CPU, or signal delay may be performed by hardware of the sensor 225.

2. Image direction determination The image direction determination method includes: -recognizing the image direction by detecting the end blank area-recognizing the image direction by the layout determination-recognizing the image direction by the character direction determination and character recognition means-punch There is a method of recognizing the image direction by detecting a hole or a staple hole and recognizing the image direction by detecting an end blank area.
The APL 5 and various functional hardware of the base unit are combined to function. Of these, the method of recognizing and detecting the image direction by the character direction determination will be described in detail.

3. Vertical determination that by the character direction determination image direction determination detects image data by the image vertical determiner,
This is done for the image memory. As the image memory, an image memory (page memory or arbitrary area memory) which is included in a single unit is used, but an image memory of another unit may be used. For example, the SAF memory described in the FAX unit or the page memory necessary for compression / expansion when reading the scanner may be used. here,
A method of determining the vertical direction of the image will be described. Generally, as a highly accurate judgment, character recognition is performed using OCR technology.
There is one that determines the top and bottom of the manuscript. In other words, cut out one character, one character in each manuscript, pattern that character,
This is a method of comparing the characteristics of the pattern with the character pattern information stored in the database to determine the character. Neuro and fuzzy techniques are often used to determine characters. If the character pattern cannot be determined without rotating the read and cut character pattern by 180 degrees, it is determined that the original is placed upside down. In this case, the top and bottom judgment is made more accurate by making not a single character but a few characters.

Next, the character direction recognition method will be described. Cut out the characters in an arbitrary character string from the original read by the scanner. At this time, the characters cut out in the vertically written original are rotated 90 degrees with respect to the horizontally written original. In addition, characters cut out depending on the direction in which the original is set, such as when the orientation of the original is uneven, are 0 degrees, 90 degrees, and 1
It may be rotating at various angles, such as 80 degrees and 270 degrees. In order to recognize rotated characters as described above, the method of extracting features while rotating the image of the input character at various angles and comparing and collating with the dictionary corresponding to the regular angle, or various rotation angles There is a method of preparing a plurality of dictionaries according to the above and comparing the features of the input image with the plurality of dictionaries.

An example of character direction recognition will be described with reference to FIGS. 29 to 33. FIG. 29 is a flowchart showing a schematic procedure of an embodiment of the character recognition method. A character image is input (step S601), and its characteristics are extracted (step S602). Here, the feature extraction is performed by extracting the contour of the input character image and adding the direction code shown in FIG. 30 to the contour portion. Next, for each direction code added to the contour portion of the input character image, a histogram for each direction code is generated as shown in FIG. (Step S603). This represents the characteristic amount H of the input character itself. Next, the histograms are rearranged according to the rotation angle of the input character to generate a histogram corresponding to the rotation of the input character (step S604). This is called a rotation histogram. This rotation histogram is collated with a dictionary (steps S605 and S60).
6) Output the recognition result (step S608). If they do not match in step S606, the value of the rotation control register is changed (step S607), the histogram of the cut-out character is rotated 90 degrees to generate a rotation histogram again (step S604), and the recognition dictionary collation (step S60) is performed.
Execute 5).

FIG. 32 shows an example of the hardware configuration for implementing the character recognition method. Reference numeral 206 denotes a scanner control circuit (scanner), and 255 denotes an image processing unit (IPU) 255 of the APL 5 including an image memory for storing an input image from the scanner 206. The unit recognizes the character direction.

The IPU 255 is a rotation information register 6 for storing how much the input unit 600 for handling image data with the scanner 206 and the original character image are rotated.
01, a recognition unit 602 for recognizing characters.

The recognition unit 602 cuts out a character image from the binary image information of the document read by the scanner 206 and performs pre-processing unit 603 for performing normalization such as noise removal. An extraction unit 604, a histogram generation unit 605 that generates a histogram of the extracted features, a rotation histogram generation unit 606 that rearranges the generated histograms according to the rotation angle of the rotation information register 601, and generates a rotation histogram. The rotation histogram is composed of a dictionary matching unit 607 that determines a candidate character by performing dictionary matching, a recognition result output result output unit 608, and a dictionary 609 that stores a standard histogram for each character without rotation. The rotation histogram generation unit 606 further includes a conversion unit 610 and a calculation unit 6.
It is composed of 11.

For example, FIG. 33 shows direction codes corresponding to respective angles of the character "sentence". When a character is read in the direction as shown in FIG. 6B, the feature extraction unit 604 converts the direction into a direction code, and the direction code is converted into a histogram by the histogram generation unit 605 as shown in FIG. To do. In this state, 0 is stored in the rotation information register 601.
Since the degree information is included, the output information of the histogram generation unit 605 and the information of the rotation histogram generation unit 606 match. This data is compared with the information in the standard histogram dictionary 609. However, the character "sentence" is 90
Since it is read after being rotated once, the judgment result of the dictionary matching unit 607 cannot be judged. In this case, the value of the rotation information register 601 is rewritten, the histogram generation unit data is again converted into the rotation histogram data with the value of the rotation information register 601, and compared with the dictionary 609. This operation is repeated until the dictionary collating unit that compares the data with the dictionary 609 determines a match. However, if there is no match even after rotating by 270 degrees, the determination is impossible and the same determination is performed for the cut out next character. By the above method, the character direction data is set in the rotation information register 601.

A method of generating rotation histogram data by using the value of the histogram generation unit 605 and the information of the rotation information register 601 will be described. In FIG. 33, for example, comparing the case of the rotation angle of 0 degrees shown in (a) and the case of the rotation angle of 90 degrees shown in (b), the direction code 1 at 90 degrees is the direction code 7, and the direction code 7 is the direction. It is possible to convert the code 2 to the same direction code as when the rotation angle is 0 degree by replacing the code 2 with the direction code 8 by adding 6 to it. However, when the result added to 6 exceeds 8, the direction code obtained by subtracting 8 from the result is replaced. This conversion is represented by the following calculation formula. D = MOD (d + c-1) +1 d: Direction code before conversion c: Constant depending on rotation angle (0 at 0 degree, 6 at 90 degree,
It is 4 at 180 degrees and 2 at 270 degrees. D: Direction code after conversion In addition to the processing up to character direction determination above, the following application is also possible by using the document size detecting means which is a known technique.

In Japan, horizontal writing (alphanumeric writing),
Vertical writing (Japanese writing) exists, and the paper orientation also has portrait orientation and landscape orientation. These combination types and the scanner 206
This is to determine the top-bottom direction when a document is placed on.
As shown in FIG. 34, the input means used for the determination is the document size, the direction angle of the line cut block output at the time of character determination, the character direction angle output by the character direction determination, and the phase difference between the angle of the character and the line. Is used. If the above combination is applied to the A4 original size, the scanner 206
In the document setting table of A, the A4 may be set in the vertical direction, that is, the setting direction of a to h in FIG. Further, as described above, vertical writing, horizontal writing, and upside down may be considered. 16 like this
According to the table of FIG. 34, the document states of the types are determined and the document states a to p are output.

3.1.1 Image Direction Recognition in Output Image Information Page Based on Character Recognition The image direction in the output image information page is detected based on the original character recognition judgment, the image direction matching is confirmed, and the image direction is confirmed. A unified copy of is made.

The contents will be described below. The following three methods are conceivable as means / methods for character recognition depending on the configuration of the copying system. (A) Digital copy machine having area memory At this time, a predetermined area or an area which is not the background portion of the image data is automatically recognized by prescan and the range is taken into the area memory to extract characters. Then, it recognizes characters. (B) Digital copier having a page memory (full memory) In this case, in the case of the page memory, a character is extracted from the page read during the reading scan and character recognition is performed. (C) Analog copier or case without sufficient memory capacity In this case, an automatic document feeder (ADF) having a reading means capable of reading only a predetermined range, an area memory, and a CPU for recognition is used. By doing so, by reading the image during the transportation of the original, the area where the predetermined area or the area of the image data that is not the background area is automatically recognized is taken into the area memory and the characters are extracted to perform the character recognition. .

A document whose image orientation has been determined follows a copy interruption and warning display output operation procedure shown in FIG. 36, which will be described later. Further, the warning display shown in FIG. 37 is displayed. Further, the present embodiment is of course also effective when applied to double-sided image formation (double-sided copying).

As described above, the image direction is detected based on the character recognition (from the document setting direction and the document size), and whether or not the copy image direction is different from the reference image direction is determined by each detected document direction. By making a judgment and confirming the matching of the image orientation, it is informed that there are originals with different image orientations, a copy with a uniform image orientation can be obtained, and missing pages can be prevented.

3.1.2 Confirmation of matching of image directionality with other image information from reference output image information based on character recognition when a large number of originals (unification of image direction) Character recognition of each original during a large number of originals The image direction is detected based on the detected document directions, and it is determined whether or not the image direction of the reference document in the copy group differs from the image direction of each detected document direction, and the matching of the image directionality is confirmed. When the image orientation mismatch is confirmed, the copying operation is interrupted and a warning display is displayed on the operation unit to obtain a copy with a unified image orientation.

The contents will be described below. When multiple originals are detected, the image orientation is detected based on the character recognition of each original,
Based on the detected document orientations, it is determined whether the copy image orientation is different from the reference image orientation, and the image orientation matching is confirmed. At that time, there are two possible methods: a) using the reference output image information as the first output image information (start page), and b) using any page as the output image information depending on the number of documents. What is optional means that the operator may set it by himself / herself, or may be written in the ROM so that the page of the reference document is automatically determined depending on the number of documents. The document whose image orientation has been determined follows a copy interruption and warning display output operation procedure shown in FIG. 36, which will be described later. Further, the warning display shown in FIG. 38 is displayed. Although not described above, the present invention is of course also effective when applied to double-sided image formation (double-sided copying).

As described above, the image orientation is detected based on the character recognition, and it is determined whether or not the copy image orientation is different from the reference image orientation based on the detected original orientation, and the image orientation matching is confirmed. By doing so, it was informed that there were originals with different image directions, and it was possible to make copies with uniform image directions and prevent missing pages.

3.1.3 Method of Detecting Discrimination between Vertical Original and Horizontal Original [Identification detection based on original size / direction data, character direction data, and line direction data (all types: vertical writing / horizontal writing can be supported)] Document size / direction data obtained by the document size detection means and the image direction and line direction in the output image information page based on the document character recognition judgment are detected to detect whether the document is portrait / landscape or portrait / landscape. Can be considered.

The contents will be described below. Figure 3
As shown in FIG. 4 and FIG. 35, even if the original size is the same, a.
It can be distinguished in 16 ways of p. However, since only the character recognition can distinguish the three parameters of the character direction, the line direction, and the phase difference between the character and the line in FIG. 34, for example, a and i, b and j, or c and k are distinguished. Is impossible. However, by adding the original size detection data to the reference data, that is, by adding two parameters, the original size and the original direction can be discriminated, and thus a and i, b and j, etc. can be distinguished. Becomes

As described above, by detecting the document size and direction in addition to the character direction and line direction by character recognition, it is possible to distinguish not only the image direction but also the vertical document or the horizontal document. , More accurate identification is possible. Therefore, it is possible to more accurately determine the consistency between individual documents and unify the image directions.

3.1.4 Confirmation of Matching of Image Directionality Based on Character Direction Recognition for Predetermined Reference Image Information When Mixed Multiple Sheets / Longitudinal / Language Originals As shown in FIG. Portrait, alphanumeric writing), landscape original landscape writing (paper orientation landscape orientation, alphanumeric writing), portrait original portrait writing (paper orientation portrait orientation, Japanese writing),
Horizontal manuscript There are four types of vertical writing (landscape in paper direction, writing in Japanese).

When many originals are stacked, it is possible that the above four kinds of originals are mixed. However, it is considered that the vertically written original and the horizontally written original are not mixed, so here is an example of mixed loading. Vertical writing and horizontal originals There are two types of vertical writing. The above-mentioned four types of originals can be identified including the top and bottom of the image by the recognition of the character direction by the character recognition means and the original size detection means. When the portrait original and the landscape original are mixed, the reference image information is determined and the image directions of the respective originals are unified with respect to the reference image information. The method of determining the reference image information is as described above. The reference stapling position (or reference punching position) of the aligned recording sheets is also as described above.

3.1.5 Image Directionality N for Multiple Originals
Recover when G is detected (mismatch in character direction) When a large number of originals are detected, the image direction is detected based on the character recognition of each original, and the image direction and the image direction of the first original in the copy group are detected based on the detected original directions. If the originals with different image orientations are detected in the image forming apparatus that determines whether or not different originals exist, a) the copying operation is interrupted and a warning display is displayed on the operation unit. B) Image rotation Perform processing (reversal on memory) c) Notify that there is a document with a different image direction by performing reverse scanning of the document, etc., or recover the image direction MG to obtain a copy with a uniform image direction. Is done. The operation procedure of the copy interruption and the warning display output will be described with reference to FIG. The flow shown in this figure is performed for each scan of one document. In the case where a command is previously written in the ROM of the copying machine body to interrupt the copy operation when the reference staple position and the detected image direction do not match in the staple mode (copy interrupt mode), as shown in step S201. Then, the copy interruption request flag is set. With this flag, the copy sequence control enters a sequence of stopping the copy operation by interrupting development of a new original image in the memory, feeding of new transfer paper, and the like. When the copy interruption command is not written, step S201 is ignored and the process proceeds to step S202.

In step S202, the operation unit unit 2
In order to display a warning on the LCD 410 of No. 04, a warning display code is transmitted if the operation unit is in the code reception READY state, and a warning is displayed to request confirmation from the operator.
FIG. 40 shows an example of a warning display on the operation unit.

An operation procedure for performing image rotation processing (inversion on memory) by image processing will be described with reference to FIG. When multiple originals are detected and the originals with different image orientations are detected by detecting the maximum edge margin area of each original,
In step S81, the image processing mode is entered, and after image processing, the process proceeds to step S82. In step S82, the operation unit receives the code READY in order to display a warning on the operation unit.
If it is in a state, a warning display code is sent and a warning is displayed to request confirmation from the operator.

FIG. 42 shows an example of a warning display on the operation unit.

The operation procedure when the reverse scanning of the original is performed will be described with reference to FIG. In the case of a large number of originals, when the originals having different image directions are detected by detecting the marginal areas of the respective originals, the original reverse scan mode is entered in step S91, and the image orientation is corrected by the reverse scan.
It proceeds to step S92. In step S92, in order to display a warning on the operation unit, if the operation unit is in the code reception READY state, a warning display code is sent out and a warning is displayed to request confirmation from the operator.

FIG. 44 shows an example of a warning display on the operation unit.
Of course, these processes are also effective in the case of double-sided image formation (double-sided copying).

As described above, when it is determined that the image orientation detected by character recognition and the reference staple position do not match, warning display / copy operation is stopped, image rotation processing, document reverse scanning, etc. are performed. By notifying that there are originals with different image directions and recovering the image directionality NG, it is possible to make a copy with a uniform image direction and to prevent missing pages.

3.1.6 Image Directionality N for Multiple Originals
Correspondence when G is detected (when image processing is not supported) When a large number of originals are detected, the image orientation is detected based on the character recognition of each original, and the image orientation of the first original in the copy group is determined from the detected original orientations. In an image forming apparatus that determines whether or not a document having a different image direction is present, when a document having a different image direction is detected, the image directionality NG cannot be recovered, and no corrective action is taken by the operator. When it is impossible to obtain a copy having a unified image direction, a) -1 According to the number of sort copies, when the number of copies exceeds a predetermined number, the copy operation is interrupted and a warning display is displayed on the operation unit. Furthermore, if the operator does not take any measures after the lapse of a predetermined time, the copy operation is restarted. a) -2 When the number of copies is equal to or smaller than a predetermined number, the copy operation is continued and a warning display is displayed on the operation unit. b) -1 When the human body detection sensor is ON, the copy operation is interrupted and a warning display is displayed on the operation unit. Furthermore, if the operator does not take any measures after the lapse of a predetermined time, the copy operation is restarted. b) -2 When the human body detection sensor is OFF, the copy operation is continued and a warning display is displayed on the operation unit. In order to reduce the dead time of the machine, notify the operator that there are documents with different image orientations, request the operator to take measures to unify the image orientations, and perform prescribed actions if there is no response. Is done.

In this case, the operation procedure corresponding to the number of sorts will be described with reference to FIG. The procedure shown in this figure is 1
It is performed every first scan of the original. If there is a document determined to have the image directionality NG with respect to the reference document in the copy document, as shown in step S231, a predetermined predetermined number of copies N (arbitrary) and the number of sorts set by the operator are compared by the comparison means. In comparison, when the number of sets is larger than the predetermined number N, a command has been written in advance in the ROM of the copying machine main body to interrupt the copy operation (copy interrupt mode), and the copy interrupt request flag is set. With this flag, the copy sequence control enters a sequence of stopping the copy operation by interrupting development of a new original image in the memory, feeding of new transfer paper, and the like. Then, in step S232, a warning display as shown in FIG. 37 is displayed to request confirmation from the operator. On the other hand, when the operator takes some action and presses the start key 431 again, the copy operation is restarted.

On the other hand, when the operator is not in front of the copying machine and cannot respond to the warning, a timer in the main body for monitoring the elapsed time (TIME) after interruption of copying. Is defined by P
If no action is taken even after waiting the SET time, the copy operation is restarted. If the number of sort copies set by the operator is equal to or smaller than the predetermined number N in step S231, step S232 is ignored and the copy operation is continued.

After the completion of copying, in step S233, a warning display copy is made if the operation unit is in the copy reception READY state in order to display a warning on the operation unit to inform the operator that there are originals with different image directions. It is sent out and a warning is displayed to request confirmation from the operator. FIG. 38 shows an example of a warning display on the operation unit. A warning display as shown in FIG. 46 is also displayed.

The operation procedure of the copying machine equipped with the human body detection sensor 225 will be described with reference to FIG. The procedure shown in this figure is performed every first scan of one document. If there is a document determined to have the image directionality NG with respect to the reference document in the copy document, and if the human body detection sensor 225 is ON as shown in step S241, the copy operation is performed in advance in the ROM of the copying machine body. A command has been written to suspend (copy suspend mode), the copy suspend request flag is expanded to the memory, the feeding of new transfer paper is suspended, and the copy operation stop sequence is entered. Then, in step S242, a warning display prompting the operator to confirm (similar to FIG. 37 and FIG. 46) is displayed.
On the other hand, when the operator takes some action and presses the start key 431 again, the copy operation is restarted.

On the other hand, if the warning cannot be dealt with, such as when the human body detection sensor 225 is operating on an object other than the operator, the elapsed time (TIME) after the interruption of copying is monitored. If the timer in the main body does not deal with waiting for a predetermined PSET time, the copy operation is restarted. Also, step S
In 241, if the operator is away from the copying machine (when the human body detection sensor is OFF), step S242 is ignored and the copying operation is continued.

After the copying is completed, in step S243, if the operation unit is in the code reception READY state, a warning display copy is made on the operation unit so as to display a warning to inform the operator that a document having a different image direction exists. Is sent and a warning is displayed to request confirmation from the operator. The warning display of the operation unit is shown in FIG.

In the copying operation using the sorter, when it is recognized that there is a document in the image direction NG, if the number of copies that can be easily corrected later is small according to the number of sorts, it is not interrupted. By making a copy as it is and displaying a warning on the copy end operation unit, it is possible to prevent missing pages. On the other hand, when there are many sorts that are difficult to correct later, the copy operation is interrupted and a warning is displayed on the operation unit to prevent missing pages. Furthermore, after the copy operation is interrupted due to a large number of sort copies and a warning is displayed on the operation unit, if no action is taken even after a lapse of a predetermined time, the copy operation is automatically restarted,
Machine downtime and current jobs can save wasted time. At the same time, by displaying a warning on the operation unit after completion of copying, when the operator performs post-binding processing such as punching and stapling on the copy group after image formation, image direction NG originals (copy) are mixed. Automatically, the operator is requested to confirm the image direction of the copy by himself, and it is possible to prevent missing pages.

In the copying operation of the copying machine equipped with the human body detection sensor 225, when it is recognized that a document in the image direction NG is present, the output of the human body detection sensor 225 is output in accordance with the output signal of the human body detection sensor 225. When it is OFF, that is, when the operator is not in front of the copier, copying is performed without interruption and a warning is displayed on the operation unit after the copying is completed, so that the operator punches holes in the copy group after image formation, When performing post-binding processing such as stapling, it is automatically notified that the image direction NG original (copy) is mixed, and the operator is required to check the image direction of the copy by himself or herself to prevent missing pages. Made it possible. On the other hand, the human body detection sensor 22
When the 5th output is ON, the copy operation is interrupted and a warning is displayed on the operation unit, so that it is possible to prevent missing pages. Further, when the human body detection sensor output 25 is ON and the copying operation is interrupted and a warning is displayed on the operation unit, if no action is taken even after a predetermined time has elapsed, the copying operation is automatically restarted. , Wasted machine downtime, and current jobs save wasted occupation time. At the same time, by displaying a warning on the operation unit after completion of copying, when the operator performs post-binding processing such as punching and stapling on the copy group after image formation, image direction NG originals (copy) are mixed. Automatically, the operator is requested to confirm the image direction of the copy by himself, and it is possible to prevent missing pages.

3.1.7 Supporting blank originals When a large number of originals are used, the image direction of each original is detected from the recognition of the character direction by the character recognizing means, and the originals are copied according to the detected original directions. In the image forming apparatus that determines whether there is a document whose image direction is different from the image direction of the first document, when recognizing that the document is a blank sheet, after reversing the document by the document reversing unit,
It is possible to read the image again and detect the image direction again for the read image.

The above processing will be described according to the procedure shown in FIGS. In step S181, the document is set on the document table 100, and in step S182, the document number counter CNT is initialized to zero. In step S183, the blank document determination flag is set to 0. Step S184
In step S185, only one original is conveyed and set at a predetermined position on the contact glass 9, and then the original is read in step S185. An edge margin area of the document is detected in the edge margin area detection step S186 with respect to the read image information. Step S187 is a blank sheet detection step, in which it is determined whether or not the read document is a blank document, and if it is not a blank document, the directionality is confirmed by the directionality confirmation routine of step S188. If the directionality is correct, a normal copy operation is performed in step S189, the value of the document number counter CNT is incremented by 1 in step S190, and the document is ejected from the document feeding device in step S191. In step S192, the presence / absence of the remaining document is checked. If the document remains, the process returns to step S183. If no document remains, the job is ended and the standby mode is entered.

When it is determined in step S187 that the original is a blank sheet, the blank sheet original determination flag is determined in step S193. If the determination flag is 0, the process proceeds to step S196, the document is inverted, and the process proceeds to step S196.
In 197, the determination flag is set to 1 and step S185
Return to and scan the original again. Step S193
If the determination flag FLG is 1, it is determined that both sides are blank documents, so that it is regarded as a slip sheet, and step S
The process proceeds from step 194 to step S195, the image data is canceled, the process proceeds to step S189, the normal copy operation is performed, and then the document is ejected at step S191.
If it is determined in step S192 that there is a remaining document, the process proceeds to step S183 to start processing the next document. If the directionality checking routine of step S188 determines that the directionality is not correct, the process proceeds to the recovery process routine of step S198 to perform a predetermined recovery process, and then proceeds to the copy operation of step S189. However, in FIG. 49, "document reading / character direction recognition" may be performed in steps S185 and S186.

The warning display is shown in FIG.

Further, the contents are the same as those shown in FIGS.
This is similar to the flowchart of FIG. 3 and the warning display examples of FIGS. 42 and 44.

3.1.8 Correspondence when Unrecognizable 3.1.8.1 Image Formation / Warning Display in Predetermined Predetermined Direction When a large number of documents are used, each document is recognized from the character direction recognition by the character recognition means. In the image forming apparatus, which detects the image direction and determines whether or not there is a document whose image direction is different from the image direction of the first document in the document group copied according to each detected document direction, When it is recognized that the orientation cannot be identified, an image is formed in a predetermined direction and a warning display is displayed on the operation unit, so that the original and the copy of the original whose image orientation cannot be identified are the original group and the original. Notification is made that it exists in the copy group.

The contents will be described below. The operation procedure of warning display output will be described with reference to FIGS. 51 and 52. When copying is started by the copy start button of the main body (step S611), the value of the counter e is initialized to 0 (step S612). The value of the value of the counter e will be described later. When the copying is started, the original direction is read and the image direction is detected by the character direction recognition (step S613). When the image direction is detected by the character direction recognition, it is determined whether the character direction can be identified (step S614).

When the character direction cannot be identified in step S614, counter e = counter e + 1 is set (step S615), and 1 is added to the value of counter e. Counter e
The value of is the number of originals whose character direction cannot be identified, and the value of the counter e is set to 0 at the start of copying in step S612, and the value of the counter e is set in step S615 every time an original whose character direction cannot be identified is found. By adding 1, the number of originals whose character direction cannot be identified after completion of copying can be known from the value of the counter e. If the character direction can be identified in step S614, the matching of the image directions is confirmed. If the image directions are matched, the image is copied as it is. If the image directions are not matched, the image data is rotated. Copy (step S619).

After the copy is completed (step S616), the value of the counter e is determined in step S617. When the value of the counter e is 0, the READY mode is set and the next copy start is awaited. When the value of the counter e is not 0, a warning display code is sent out (step S618) and a warning display is issued to request confirmation from the operator if the operation portion is in the code reception READY state in order to display a warning on the operation portion. Has been done. The contents displayed on the operation unit are shown in FIG.
As described above, it is possible to prevent missing pages by displaying a warning to the operator that the image direction cannot be identified.

3.1.8.2 Image formation / warning display by unifying with the identified predetermined reference image information direction When a large number of documents are used, the image orientation of each document is detected from the recognition of the character direction by the character recognition means, In the image forming apparatus that uses the information as the reference image direction information, determines whether the image direction of the document is different from the reference image direction, and aligns the image directions,
When the image direction of the reference page is recognized as unidentifiable, the next page is determined to be the reference image direction page, and the image directions are aligned based on the information of the page. By displaying a warning display of the above items on the operation unit, the operator is informed that there is a document whose image direction cannot be identified and a copy of the document.

The contents will be described below. Equipped with an automatic document feeder that sets the document side facing down and feeds the document from the first sheet (original equivalent to the cover) of the bottom sheet,
When the image orientation of the first page, which is the reference image information, cannot be determined by the character orientation confirmation by the character recognizing means, using an image forming apparatus that aligns the image orientation with the image information of the document corresponding to the cover as the reference image information. The operation procedure will be described with reference to FIGS. 54, 55 and 56.

Copy start button (start key 43
When the copy is started according to 1) (step S62)
1), the value of the page counter m is initialized to 1 (step S622), and the value of the counter a is initialized to 0 (step S623). The value of the page counter m and the value of the counter a will be described later. When copying is started, the image direction is detected by reading the document and recognizing the character direction (step S624). The value of the counter a represents whether or not the reference image information is determined. When the value of the counter a is 0, the reference image information is not determined, and the value of the counter a is 0. If not, the reference image information has been determined. The value of the counter a is initially 0 (step S623), and when the reference image information is determined in step S633, the value of the counter a is incremented by one. In step S625, it is determined whether the value of the counter a is 0, and if the value of the counter a is 0, step S62.
6. If the value of the counter a is not 0, go to step S63.
Go to 1.

In step S626, it is determined whether the character direction cannot be identified. In step S626, when the character direction cannot be identified, the page counter m = page counter m + 1 is set (step S627), 1 is added to the value of the page counter, and the copy is performed as it is. The value of the page counter m is a value indicating how many pages of the document the character direction cannot be identified. In step S622, the value of the page counter m at the start of copying is initialized to 1, and the document whose character direction cannot be identified. By adding 1 to the value of the page counter m every time when is found, it can be understood that the character direction cannot be identified up to m originals.

If the character direction can be identified in step S626, the image information of the m-th original is used as the reference image information, the image direction consistency is confirmed, and if the image directions are aligned, the image is copied as it is. If the image directions are not matched, the image data is rotated and copied (step S632). Then, in step S633, the counter a = a
It is set to +1 and 1 is added to the value of the counter a.

After the copy is completed (step S628), it is determined in step S629 whether or not the value of the page counter m is 1, and when the page counter m = 1, the READY mode is set and the next copy start is waited. When the page counter m = 1, the original image information corresponding to the front cover is the reference image information. If the page counter m is not 1, a warning display code is sent out if the operation unit is in the code reception READY state to display a warning (step S630), and a warning is displayed to request confirmation from the operator. FIG. 57 shows the contents displayed on the operation unit. By doing so, the operator is warned that the image direction cannot be identified, and it is possible to prevent missing pages.

3.1.8.3 Image Discontinuation / Warning Display In the case of a large number of originals, the image direction of each original is detected from the recognition of the character direction by the character recognizing means, and a copy is made according to each detected original direction. In the image forming apparatus that determines whether or not there is a document whose image direction is different from the image direction of the first document in the document group, if the image direction cannot be identified, the copying operation is interrupted and a warning message is displayed. By displaying it on the operation unit, it is shown that a document whose image direction cannot be identified and a copy of the document exist in the document group and the copy group, and the operator is informed to change the document setting direction.

The contents will be described below. The operation procedure of image forming interruption / warning display output will be described with reference to FIGS. 58 and 59. When copying is started by the copy start button (step S641), the document direction is detected and the image direction is detected by character direction recognition (step S64).
2) In step S643, it is determined whether the character direction cannot be identified. If the character direction can be recognized, the consistency of the image direction is confirmed. If the image direction is matched, the image is copied as it is, and the image direction is not matched. At this time, the image data is rotated and copied (step S648). Step S
If the character direction cannot be identified at 643, the copy operation is stopped (step S644), and a warning display code is sent out to the operator for confirmation in order to display a warning on the operation unit if the operation unit is in the code reception READY state. A warning is displayed to request (step S645). The contents displayed on the operation unit are the same as in FIG.

When the copy start button is turned on after the warning is displayed, the original document whose stopped character direction cannot be identified on the contact glass 9 is copied as it is (step S646), and the warning is displayed on the operation unit. In order to perform the above, if the operation unit is in the code reception READY state, a warning display code is sent and a warning is displayed to request confirmation from the operator. After the copy is completed (step S647), RE
The ADY mode is set, and the next copy start is awaited. After the warning is displayed in step S645, when the ADF is opened / closed, it is determined that the document stopped on the contact glass 9 has been removed, and the READY mode is set.
Wait for the next copy start. 60 and 61 show examples of warning display.

As described above, the operator detects that the originals are mixed and the orientations of the originals are not recognizable, and the operator stops the copying operation and displays a warning. A uniform copy can be obtained in the image direction of the eye. Especially when performing post-binding processing such as punching after copying and stapling, usually, only the front cover is used to perform punching and stapling, and if there are copies with different image orientations in the copy group, pages will be missing pages. Turns over and looks bad. In order to prevent such a missing page, the present invention automatically notifies the operator that originals having different rear image directions are mixed during copying, and the operator himself checks the image direction of the originals and sets the originals. Demanding that the direction be changed, it is possible to prevent missing pages.

3.2.1 Staple Position Determination by Character Recognition Character recognition is used to recognize the direction of characters and character strings on the paper of the document, thereby detecting the image direction of the document,
The optimum post-processing (sorter / stapler, finisher, etc.) position is determined from the relationship between the paper and the character string.

The contents will be described below. As described above, when the direction of characters and character strings with respect to the paper of the original is recognized using the character recognition method, the relationship between the paper and the character string viewed from the normal character direction (0 °) is shown in the table of FIG. 39, the number is limited to 4 as shown in FIG. At this time, depending on what kind of post-processing the user desires (sorter / stapler, finisher, etc.), the following limitation is usually imposed. (I) In the case of stapling in one place As shown in FIG. 62, the upper left corner is (a, b) in horizontal writing, and the upper right corner is (c, d) in vertical writing.
Determine the staple position.

(Ii) In the case of stapling or punching holes at two places In this case, as shown in FIG. 63, the left side band portion is (a, b) in horizontal writing, and the right side band portion is (c, d) Determine the staple position or punch hole punching position (in the case of staples).

The following three methods are conceivable as means / methods for character recognition depending on the configuration of the copying system. (A) Digital copier having an area memory A predetermined area or an area of the image data which is not a background area is automatically recognized by prescanning, and the range is fetched into the area memory to extract the characters. What recognizes.

(B) Digital copying machine having page memory (full memory) In the case of page memory, a character is extracted from the page read at the time of reading scan and character recognition is performed.

(C) An analog copying machine or an automatic document feeder (ADF) having a reading means capable of reading only a predetermined range, an area memory, and a CPU for recognition are used in the case of not having a sufficient memory capacity. As a result, by reading the image data while the document is being conveyed, a predetermined area or a range in which the area of the image data that is not the background portion is automatically recognized is fetched into the area memory and characters are extracted,
Character recognition.

The flow of the actual copy operation is as shown in FIG. When the operator presses the copy start key, the control CPU identifies whether or not it is in the staple post-processing mode (step S651), and when it is not in the staple mode, a normal copy operation is performed (step S652). If it is in the staple mode, it is identified whether or not there is one designation (step S6).
53) and in the next step, the processing means determines the processing position as shown in FIGS. 62 and 63 according to each case (steps S654 and S655).

As described above, since the optimum post-processing position is automatically determined from the direction of the character or character string of the document to be read by the character recognition, the operator's work and confusion for the post-processing can be avoided. Is possible. Further, there is an advantage that the operator does not need to specify the position depending on the case, the operation is easy, and the work is performed speedily and the time is saved.

3.2.2 Comparison of Character String Direction Data and Reference Staple Position The character image recognition method is used to detect the image direction of the original from the direction of the character and the character string on the original paper, and the relationship between the paper and the character string is detected. Optimal post-processing (staple, punch) determined from
The consistency between the position and the post-processing position constrained by the hardware is checked.

As described above, in the sorter stapler, the operation position is limited to one position at a fixed corner position, and in the case of the finisher, the operation position is limited to a straight line on one end side. Therefore, in the related art, if the user's instruction input and the document setting direction are not in the correct positional relationship to save hardware, there is a possibility that post-processing will be performed at an incorrect position. Therefore, in the present embodiment, in addition to the image direction recognizing means using the character recognizing method as described above, a position capable of post-processing by hardware and an optimum post-processing position determined based on the character string direction data are compared. And a step of confirming the consistency of the positional relationship between the two. S / S
In this case, the sheet is stapled in the shaded portion with respect to the sheet direction as shown in FIG. At this time, the original is
If the vertical document is horizontally written as in (a), it is aligned with the reference staple position. However, in the case of vertical document vertical writing as shown in FIG. The following is an example of the determination made on the actual CPU. As shown in FIG. 67, numbers 1 to 8 are assigned to the absolute staple positions when they are output on the tray of the sorter. When the staple position obtained as a result of character recognition is i, i = 4 in the case of FIG. 66 (a). If the staple position on the hard side is k, the staple position on the hard side is limited to 4 or 8 depending on the sheet direction due to the S / S restriction. In the case of FIG. 65, k = 4. Only when i = k is compared on the CPU, it is determined that the two match. In the case of FIG. 66 (b), i =
Since 3 is 3 and i ≠ k, it is determined that they do not match. On the copy operation, the procedure is as shown in FIG. This procedure is the same as that in FIG. 64 from step S651 to S655, and the hardware constraint condition (step S
656) and the consistency confirmation process (steps S657 and S65).
8) is provided.

As described above, since the optimum post-processing position is automatically determined from the direction of the character / character string of the document to be read by the character recognition, the operator's work and confusion for the post-processing can be avoided. It becomes possible. Further, there is an advantage that the operator does not need to specify the position depending on the case, the operation is easy, and the work is performed speedily and the time is saved. Further, in this embodiment, it is possible to confirm whether or not the determination result conforms to the constraint condition by hardware, so that it is possible to prevent post-processing at an erroneous position.

3.2.3 When a large number of sheets and vertical and horizontal originals are mixed, the image information direction is identified from the character direction recognition data with respect to the predetermined reference image information, and the alignment confirmation of the image direction is made by comparison with the reference stapler position data . 2.4 Consistency between character string direction data and reference stapling position NG recovery a) Image interruption / warning display Character recognition is used to detect the image direction of the document from the direction of the characters and character string on the paper of the document, Optimal post-processing (staple, punch) determined from the relationship between paper and character strings
The consistency between the position and the post-processing position constrained by the hardware is checked, and if it is determined that they do not match, the copying operation is interrupted and a warning display is displayed on the operation unit, so that the optimum stapling position and the copying machine The post-processing device indicates that the possible post-processing positions do not match, and informs the operator to change the document setting direction or the paper direction.

The contents will be described below. Figure 6
9. As shown in the flowchart of FIG. 70, as a result of comparing the position that can be post-processed by hardware using character recognition with the post-processing position determined based on the character string direction data, it is found that the positional relationship between them is consistent. If it is determined that there is no image formation, the image formation is interrupted (step S659), and a warning such as "Cannot staple to the correct position. Check the orientation of the paper or document" is displayed on the display unit (step S659).
660). Note that steps S651 to S658 are shown in FIG.
4, the same as FIG. 68. When using a DF or the like, if the image formation is interrupted and a warning is displayed, the DF is also stopped. In this case, the DF process will be described separately.

As described above, since the optimum post-processing position is automatically determined from the direction of the character / character string of the document to be read by the character recognition, the operator's work and confusion for the post-processing can be avoided. It becomes possible. Further, there is an advantage that the operator does not need to specify the position depending on the case, the operation is easy, and the work is performed speedily and the time is saved. Further, in this embodiment, when the judgment result does not meet the constraint condition by the hardware, the copying work is stopped, the operator is notified that the direction of the post-processing is not suitable, the original or paper setting direction is confirmed, and the setting direction is changed. Since it is required to do so, it becomes possible to prevent post-processing to an erroneous position.

B) Image rotation processing The character orientation is used to detect the image direction of the document from the direction of the characters and character strings on the paper of the document, and the optimum post-processing (staple, stapling, etc.) determined from the relationship between the paper and the character string is detected. punch)
The consistency between the position and the post-processing position constrained by the hardware is checked, and if it is determined that the post-processing position does not match, the optimum staple position and the post-processing device of the copying machine are rotated by rotating the image data captured in the frame memory. It is conceivable to match the post-processable position and perform the optimum post-processing.

The contents will be described below. When the data copying machine has a page memory, as shown in FIG. 69, since the image data can be taken into the frame memory prior to the image direction recognition, the position where the post-processing by hardware can be performed by using the character recognition. And the post-processing positions determined based on the character string direction data are compared, and when it is determined that the positional relationship between the two is inconsistent, as shown in the procedure of FIG. The required rotation angle is calculated by using the orientation data and the orientation data of the post-processing device to rotate the image data on the memory (step S662). After that, copy execution and post-processing are executed (step S663).

As described above, since the optimum post-processing position is automatically determined from the direction of the character or character string of the original to be read by the character recognition, the operator's work and confusion for the post-processing can be avoided. It becomes possible. Further, since the operator does not need to specify the position depending on the case, there is an advantage that the operation is facilitated and the work is speedy and the time is saved. Further, in the present embodiment, when the determination result does not match the constraint condition by the hardware, the image data is rotated so as to match the direction of the post-processing, so that it is optimal without an operation such as changing the original (paper) set by the operator. It is possible to staple at a post-processing position.

C) Reverse scanning of original document Using character recognition, the image direction of the original document is detected from the direction of characters and character strings on the original document sheet, and the optimum post-processing (staple, stapling, etc.) determined from the relationship between the sheet and the character string is detected. punch)
If the consistency between the position and the post-processing position constrained by the hardware is confirmed, and if it is judged that the position does not match, the optimal staple position and post-processing of the copying machine will be performed under limited conditions even without a frame memory. If the device automatically matches the post-processable position and performs post-processing, and if it is determined that the device cannot be restored, the copying operation is interrupted and a warning is displayed on the operation unit. , Indicating that the optimum position for stapling does not match the post-processing position that can be processed by the post-processing device of the copying machine, and notifies the operator to change the document setting direction or the paper direction.

The contents will be described below. FIG. 69
As shown in, when the post-processing position determined by the hardware and the post-processing position determined by the character string direction data are compared using character recognition, and it is determined that the positional relationship between the two is inconsistent. Even if the digital copying machine does not have a page memory, it usually has a line buffer for several lines, so that the paper orientation is set correctly as shown in the procedure of FIG. 71 (at step S664). y), the reference staple position and the optimum post-processing position are 180 °
If it is deviated (y in step S665), the reading operation is not performed during the normal scanning, and the image is inverted in the main scanning direction while returning at the reading speed when returning (step S).
By performing the image forming process, it is possible to perform the post-processing at the correct position (step S667). In other cases, the image formation is interrupted (step S668) and a warning such as "Cannot staple to the correct position. Please check the direction of the paper or the document" is displayed on the display unit (step S669). When using a DF or the like, if the image formation is interrupted and a warning is issued, the DF is also stopped. In this case, the DF process will be described separately.

As described above, since the optimum post-processing position is automatically determined from the direction of the character / character string of the document to be read by the character recognition, the operator's work and confusion for the post-processing can be avoided. It becomes possible. Further, there is an advantage that the operator does not need to specify the position depending on the case, the operation is easy, and the work is performed speedily and the time is saved. Further, in this embodiment, if the determination result does not meet the constraint condition by hardware, it is considered that the erroneous setting is most likely 1
Since the image data is apparently rotated so as to match the post-processing direction with respect to the setting error of 80 °, the staples can be placed at the optimum post-processing position without the need for the operator to change the original (paper) set. It will be possible. In the case of any other wrong setting, the copying operation is stopped, the operator is notified that the post-processing direction is not correct, and it is requested to confirm the original or paper setting direction and change the setting direction. It is possible to prevent post-processing of the position.

3.3.1 Selection Control of Image Direction Identification Detection According to Image Forming Mode The image direction is detected by recognizing the character direction by using the character recognition means, and the image direction is detected by each detected image data. In order to identify and detect, it is necessary to perform work such as image processing. Then, the work has to be performed every time each original is scanned, which takes a very long time, and is inefficient in the case of a normal copy work. Therefore, in the case of a normal copy work, the copy work is performed without passing through the image direction recognition detection unit, and it is determined in step S351 of the flowchart shown in FIG. 72 whether or not to detect the image direction identification. When performing post-processing such as stapling and punching, the direction identification detection is performed unconditionally, and the direction identification flag is set. When the direction identification flag is set, the process proceeds to the image direction identification detection mode in step S352. After the copying is completed, the process proceeds to the stapling mode (post-processing mode) in step S354, and after the processing, the process proceeds to the standby mode in step S355.
Wait for key input instruction from the operator. When the direction identification flag is off, the process proceeds to the normal copy mode of step S353, and after the copying is completed, the process proceeds to the standby mode of step S355 to wait for a key input instruction from the operator.

As described above, in the image forming apparatus for detecting the image direction by detecting the marginal area of the document, the image direction is identified in the normal image forming mode (not including the post-processing) in the image forming apparatus. By deactivating the detection, it is possible to prevent a reduction in the efficiency of the copying work.

3.3.2 When the manual stapling instruction is input after the completion of the sorting operation and the staple position is NG based on the character direction data 3.3.2.1 Warning display The warning display output operation procedure is shown in FIGS. It will be described based on. When the stapling key is turned on and the copy is started (step S681), the image direction is judged from the recognition of the character direction by the character recognition means for each document, and the image direction consistency is confirmed, and the image direction is confirmed. If the consistency of is not satisfied, the image data is rotated and copied, and the copy is ended (step S682). After copying,
The stapler is activated (step S683), READY
Enter the mode. When the stapling key is not turned on, when copying is started (step S684), the image direction is judged from the recognition of the character direction by the character recognition means for each original, and the consistency of the image directions is confirmed. While copying, end the copy (step S68).
5). After the copying is completed, if the staple key is not turned on, the stapler does not operate and goes into the READY mode. After the copying is completed, even if the staple key is turned on, if the recording paper does not remain in the bin, the stapler does not operate and the READY mode is set. This is performed regardless of whether the image directions are consistent (step S686) or not (step S687). After the copying is completed, when the staple key is turned on (manual stapling), the alignment in the image direction is maintained, and when the recording paper remains in the bin, the stapler is activated (step S688), and REA is set.
Enter DY mode.

During manual stapling, when the image orientation is not consistent, the stapler does not operate even if the recording paper remains in the bin, and a warning is displayed on the operation panel (step S689).

3.3.2.2 Staple Prohibition A flow of staple prohibition operation will be described with reference to FIG. During manual stapling, if the image orientation is not consistent, the stapler operation is prohibited even if the recording paper remains in the bin, and a warning is displayed on the operation panel (step S691). Further, FIG. 76 shows an example of the warning display.

[0185]

As described above, the detecting means for detecting the image information size and the image information set direction, the line direction recognizing means for recognizing the line direction of the image information, and the character direction for recognizing the character direction of the image information. According to the invention of claim 1, the size data and the set are provided with a recognition means and an image direction recognition means for recognizing the image direction of the image information based on the size data and the set direction data, the line direction data and the character direction data. Since the image direction is recognized based on the direction data, line direction data, and character direction data, the vertical orientation of the image can be accurately identified even in the case of vertical and horizontal originals. It is possible to provide an image recognition device capable of performing image direction recognition with extremely high recognition accuracy.

Further, a detection step of detecting the image information size and the image information set direction, a line direction recognition step of recognizing the line direction of the image information, a character direction recognition step of recognizing the character direction of the image information, and the size An image direction recognition step of recognizing the image direction of the image information based on the data and the set direction data, the line direction data and the character direction data. According to the invention of claim 2, the same reason as that of the invention of claim 1 is provided. This makes it possible to accurately identify the vertical direction of the image even when the vertical document and the horizontal document are mixed, and to perform the image direction recognition with extremely high recognition accuracy.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an entire digital copying machine according to an embodiment of the present invention.

FIG. 2 is a plan view of an optical writing unit of the digital copying machine.

FIG. 3 is a side view of an optical writing unit of the digital copying machine.

FIG. 4 is a block diagram showing a control unit of the digital copying machine.

FIG. 5 is a block diagram of the entire electrical equipment control of the digital copying machine.

FIG. 6 is a block diagram of an image scanner unit.

FIG. 7 is a schematic block diagram of an image processing unit.

FIG. 8 is an explanatory diagram showing data switched by a data switching mechanism.

FIG. 9 is a block diagram of a memory system.

FIG. 10 is a block diagram of a memory system.

FIG. 11 is a block diagram of a memory system.

FIG. 12 is a block diagram of a memory device.

FIG. 13 is an internal block diagram of a memory unit of the memory device.

FIG. 14 is an explanatory diagram showing three image data types.

FIG. 15 is an internal block diagram of a memory device.

FIG. 16 is a block diagram of a memory system using an external storage device.

FIG. 17 is an internal block diagram of a memory device.

FIG. 18 is a block diagram of an application unit.

FIG. 19 is a plan view of an operation display unit.

FIG. 20 is a block diagram of a memory system.

FIG. 21 is a schematic block diagram of an image processing unit.

FIG. 22 is a reception block diagram of image information.

FIG. 23 is a block diagram of an image processing function.

FIG. 24 is a schematic diagram of a bitmap page memory.

FIG. 25 is an explanatory diagram of image information data expanded in a memory at the time of normal scanning.

FIG. 26 is an explanatory diagram of image information data expanded in a memory at the time of reverse scanning.

FIG. 27 is an explanatory diagram of image information data mirrored on a memory.

FIG. 28 is an explanatory diagram showing margins in the X and Y directions of a document.

FIG. 29 is a flowchart showing an example of a character recognition method.

FIG. 30 is an explanatory diagram showing direction codes of an outline portion of an input character image.

FIG. 31 is an explanatory diagram of a histogram for each direction code.

FIG. 32 is a character recognition block diagram.

FIG. 33 is an explanatory diagram showing direction codes corresponding to respective angles of the character “sentence”.

FIG. 34 is an explanatory diagram showing a combination table of document size and writing method.

FIG. 35 is an explanatory diagram showing the positions of various document placements.

FIG. 36 is a flowchart of copying interruption and warning display output when a blank original is determined.

FIG. 37 is a plan view of the operation display unit.

FIG. 38 is a plan view of the operation display unit.

FIG. 39 is an explanatory diagram showing types of horizontal writing and vertical writing of a document.

FIG. 40 is a plan view of the operation display unit.

FIG. 41 is a flowchart when image rotation processing is performed by image processing.

FIG. 42 is a plan view of the operation display unit.

FIG. 43 is a flowchart when reverse scanning of a document is performed.

FIG. 44 is a plan view of the operation display unit.

FIG. 45 is a flow chart of a copy operation for coping with the number of sort copies when the image directions are not aligned.

FIG. 46 is a plan view of the operation display unit.

FIG. 47 is a flowchart of a coping operation of a copying machine equipped with a human body detection sensor when image directions are not aligned.

FIG. 48 is a plan view of the operation display unit.

FIG. 49 is a flowchart of blank document determination processing.

FIG. 50 is a flowchart of blank document determination processing.

FIG. 51 is a flowchart of an operation of performing image formation and warning display in a predetermined direction determined in advance.

FIG. 52 is a flowchart of an operation of forming an image and displaying a warning in a predetermined direction determined in advance.

FIG. 53 is a plan view of the operation display unit.

FIG. 54 is a flowchart of the operation of forming an image and displaying a warning by unifying the identified predetermined reference image information directions.

FIG. 55 is a flowchart of an operation of forming an image and displaying a warning by unifying the identified predetermined reference image information directions.

FIG. 56 is a flowchart of an operation of forming an image and displaying a warning by unifying the identified predetermined reference image information directions.

FIG. 57 is a plan view of the operation display unit.

[Fig. 58] Fig. 58 is a flowchart of image formation interruption and warning display operation when the image direction cannot be identified.

[Fig. 59] Fig. 59 is a flowchart of image formation interruption and warning display operation when the image direction cannot be identified.

FIG. 60 is a plan view of the operation display unit.

FIG. 61 is a plan view of the operation display unit.

FIG. 62 is an explanatory diagram showing each pattern of one staple stopping.

FIG. 63 is an explanatory diagram showing each pattern of stapling at two staples.

FIG. 64 is a flowchart for stapling position determination.

FIG. 65 is an explanatory diagram showing the relationship between the paper orientation and the suitability of the staple position.

FIG. 66 is an explanatory diagram showing the relationship between the paper orientation and the suitability of the staple position.

FIG. 67 is an explanatory diagram showing the relationship between the paper orientation and the suitability of the staple position.

FIG. 68 is a flowchart for stapling position determination.

FIG. 69 is a flowchart of a warning display process when the consistency between the character string direction data and the reference staple position is NG.

FIG. 70 is a flowchart of image rotation processing when the matching between the character string direction data and the reference staple position is NG.

FIG. 71 is a flowchart of document reverse scanning processing when the matching between the character string direction data and the reference staple position is NG.

FIG. 72 is a flowchart of selection control of image direction identification detection according to an image forming mode.

FIG. 73 is a flowchart of a warning display process when the staple position is NG when a manual stapling instruction is input.

FIG. 74 is a flowchart of a warning display process when a manual stapling instruction is input and the stapling position is NG.

FIG. 75 is a flowchart of the stapling prohibition operation when the manual stapling instruction is input and the stapling position is NG.

FIG. 76 is a plan view of the operation display unit.

[Explanation of symbols]

 200, 201 CPU 202 Image control circuit 203 Signal switching gate array 204 Operation unit unit 207 Page memory 208 Image processing unit 210 Application unit 220 Main control board

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] October 25, 1993

[Procedure Amendment 1]

[Document name to be corrected] Drawing

[Correction target item name] All drawings

[Correction method] Change

[Correction content]

[Figure 3]

[Figure 8]

FIG. 30

[Figure 1]

[Fig. 2]

[Figure 9]

[Figure 10]

FIG. 11

FIG. 25

FIG. 26

[Figure 4]

[Figure 5]

[Figure 6]

FIG. 14

FIG. 27

[Figure 7]

[Fig. 12]

FIG. 15

FIG. 31

FIG. 66

[Fig. 13]

FIG. 16

FIG. 17

FIG. 33

FIG. 67

FIG. 18

FIG. 19

FIG. 20

FIG. 21

FIG. 22

FIG. 23

FIG. 37

FIG. 24

FIG. 28

FIG. 35

FIG. 65

FIG. 29

FIG. 38

FIG. 32

FIG. 39

FIG. 62

FIG. 34

FIG. 36

FIG. 40

FIG. 41

FIG. 42

FIG. 43

FIG. 44

FIG. 45

FIG. 46

FIG. 48

FIG. 53

FIG. 47

FIG. 49

FIG. 50

FIG. 52

FIG. 57

FIG. 51

FIG. 54

FIG. 60

FIG. 55

FIG. 76

FIG. 56

FIG. 58

FIG. 61

FIG. 59

FIG. 63

FIG. 70

FIG. 71

FIG. 64

FIG. 68

FIG. 69

FIG. 72

FIG. 73

FIG. 74

FIG. 75

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Akihisa Itabashi 1-3-6 Nakamagome, Ota-ku, Tokyo In Ricoh Co., Ltd. (72) Hiroshi Fukano 1-3-6 Nakamagome, Ota-ku, Tokyo Shares Within Ricoh Company (72) Inventor Fumio Kurizumi 1-3-6 Nakamagome, Ota-ku, Tokyo Within Ricoh Company, Ltd.

Claims (2)

[Claims] 1. A detection means for detecting the image information size and the image information set direction, a line direction recognition means for recognizing the line direction of the image information, a character direction recognition means for recognizing the character direction of the image information, and the size. An image orientation recognition device comprising: an image orientation recognition means for recognizing an image orientation of image information based on data and set orientation data, line orientation data and character orientation data. 2. A detecting step of detecting an image information size and an image information set direction, a line direction recognizing step of recognizing a line direction of the image information, a character direction recognizing step of recognizing a character direction of the image information, and the size. An image orientation recognition step of recognizing the image orientation of the image information based on the data and the set orientation data, the row orientation data and the character orientation data,
An image orientation recognition method characterized by performing image orientation recognition by performing each step.