JPH02230870A - Picture reader - Google Patents

Abstract

PURPOSE:To form a sharp picture with excellent contrast by providing a background eliminating means reducing a picture data based on a background data and a density correction calculation means applying the operation to increase an output from the background eliminating means based on a contrast coefficient data and outputting a correction picture data. CONSTITUTION:An original D is scanned by a linear image sensor 11 and read while being divided minutely into picture elements. A density discrimination means 112 calculates a density coefficient data to detect a maximum density and a background density in a prescribed area of the original D and to correct the picture data and the background data corresponding to the background density before the main scanning of the original D to output the picture signal. A background elimination means 501 decreases a picture data in the case of main scanning based on the background data. A density correction calculation means 502 applies the calculation to increase the output from the background eliminating means 501 based on the density coefficient data to output a correction picture data. Thus, a sharp picture with excellent contrast is formed regardless of the background density of the original.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention relates to an image reading device 1 that outputs an image signal for forming a hard copy image, and particularly relates to an automatic density correction for forming a clear image. Regarding. [Prior Art] Traditionally, digital copying machines and facsimiles have been used as image reading means to optically read originals consisting of still images using image sensors, perform various image processing on the obtained image data, and then An image reading device that outputs image signals is used. In order to form clear images in such image reading devices, density correction is performed to correct image data to an optimal value depending on the degree of shading of an image consisting of characters, figures, etc. in a document. For example, even if the image is pale, as if drawn with a pencil, the image data is corrected so that the hard copy image has an appropriate density. Conventional density correction is performed at the stage of analog-to-digital (A/D) conversion of a read signal from an image sensor to generate image data of a predetermined pit. This was done by changing the reference voltage for A/D conversion. Note that the density correction amount may be specified by the operator or automatically adjusted. In automatic density correction, the density of the image is detected in advance by performing a preliminary scan or the like. [ill the invention tries to solve! I) With conventional image reading devices, if the document has an image formed on a colored (other than white) base, such as a blueprint or newspaper, the density of the base is corrected along with the image. Ru. For this reason, especially in originals with dark background colors, the background
There was a problem in that an unclear image was formed with poor contrast between the image and the image, resulting in so-called background stains. In addition, since it involves analog control, not only is the accuracy unstable, but density correction changes the dynamic range of A/D conversion, deteriorating image processing characteristics such as gradation reproduction and image edge enhancement. Sometimes. SUMMARY OF THE INVENTION In view of the above-mentioned problems, it is an object of the present invention to provide an image reading device that can form a clear image with good contrast regardless of the background density of a document. [Means for Solving the Problems] In order to solve the above-mentioned problems, the present invention uses a one-dimensional image sensor that scans the original to subdivide and read the original into pixels, and quantizes the output of the one-dimensional image sensor. An image reading device that performs image processing on image data and outputs an image signal corresponding to each pixel. a density determination unit that detects a maximum density and a background density and calculates density coefficient data for correcting the image data and background data corresponding to the background density; and a density correction calculation means that performs a calculation to increase the output from the background removal means based on the density coefficient data and outputs corrected image data. It will be done. [
Operation] The original is scanned by a one-dimensional image sensor, divided into pixels, and read. The density determination means detects the maximum density and the background density in a predetermined area of the original before main scanning of the original is performed to output an image signal, and detects the density coefficient data and the background density for correcting the image data. Calculate the underlying data corresponding to . The background removal means reduces the image data based on the background data during the main scan. The density correction calculation means performs a calculation to increase the output from the background removal means based on the density coefficient data, and outputs corrected image data. [Examples] Examples of the present invention will be described below with reference to the drawings. Fig. 4 is a perspective view showing the optical system of the image reader IR incorporated in the digital copying machine, Fig. 5 is a plan view of the image sensor l1, and Fig. 6 is an enlarged view of the CCD sensor chips lla and llb in Fig. 5. It is. - A document D placed on a document table glass (not shown) is scanned in the sub-scanning direction by a slider 14 equipped with an image sensor l1, and an exposure lamp 17 such as a halogen lamp that emits light in the visible light band and a rod lens array. 15 and an image sensor 11, the image is subdivided into pixels and read. As shown in FIG. 5, the image sensor 11 includes five contact type CCD sensor chips lla to 1'le.
are arranged in a staggered manner so as to be continuous in the horizontal direction (main scanning direction 5) and alternately at a pitch of 4 pixels in the vertical direction (sub-scanning direction). Therefore, the rear CCD sensor chip in the sub-scanning direction
Although a delay occurs in the output signals from la, llc, and lie, this is corrected by setting the timing of the line shift pulse signal applied to each CCD sensor chip lla to lie.

Each CCD sensor chip lla-lie has a size of approximately
A large number of 3.5 am (d=1/400 inch) square elements 12, l2, . FIG. 7 is a bronch diagram of the image reader TR.

The image sensor 11 has five COD sensor chips lla to 1 to increase the reading speed in the main scanning direction.
1e are driven at the same time, and effective read pixel signals for 960 pixels are serially output from each. The 5 systems of photoelectric conversion outputs serially output simultaneously (in parallel) from the 5 COD sensor chips lla to lie are sent to the line memory 111 and the CPU (central processing unit) 11.
2. Signal processing is performed by each of the following image processing circuits 101 to 110, which together with the ROM 113 constitute an image processing section. First, each of the five photoelectric conversion outputs is quantized into 8 bits (256 gradations) by a digitization processing circuit 101 having a sample and hold circuit and an A/D converter.
is converted into image data and sent to the 5-channel synthesis circuit 102.
is input to. Since the photoelectric conversion output is proportional to the intensity of reflected light from the original D, the value of the image data at this time is
It is r255J for the pixel in the lightest white background part, and "0" for the pixel in the darkest black part.

The 5-channel synthesis circuit 102 temporarily stores two lines of image data in a total of five first-in, first-out memories for each chip, and sequentially selects and reads out the image data from each chip at l-line intervals to arrange the pixels. Image data D1 as a serial image signal corresponding to (reading scan order)
7 ~ Output IO. Next, the image data D17 to D10 input to the shading correction circuit 104 are based on one line of reference image data read from the reference white plate 16 (see FIG. 4) in the main scanning direction of the exposure lamp 17. Light distribution (uneven light intensity)
A correction corresponding to the sensitivity difference between each element 12 is added. Further, in this shading correction circuit 104, the image data D17 to D1, which are data signals proportional to the reflected light intensity, are
0 is logarithmically converted in accordance with the visual characteristics in consideration of the reading range of the original D, and converted into image data 027 to 20 as a density data signal proportional to the density of each pixel.

In other words, the values of image data 27 to 20 increase as each pixel becomes darker, and are "0" for the pixel in the lightest white area of the original D, and r2 for the pixel in the darkest black area. 5 5J. The image data D27 to D20 are subjected to density correction processing, which will be described later, in order to form a clear image in the gamma correction circuit 106, and the gamma correction circuit 106 outputs image data D37 to D30 as corrected image data. Ru.

The scaling/editing processing circuit 108 determines the output timing of the image data D47 to D40 to be output in order to form a scaled image enlarged or reduced by a thinning method or an interpolation method, and an edited image such as movement or mirror inversion. , the output order, or the scanning speed in the sub-scanning direction. The MT-F correction circuit 109 performs smoothing to prevent the occurrence of moiré fringes and edge enhancement to eliminate edge loss. Image data D57 output from the MTF correction circuit 109
~50 is 2 based on the area gradation method in the gradation reproduction circuit 110.
After passing through the value conversion process, the image signal VI is sent to a laser printer unit (not shown) that forms a copy image on paper using an electrophotographic process.
Sent as DEO4~0. Note that the line memory 111 is used to temporarily store image data at a specific processing stage, and programs and various data are read from the ROM 113. Fig. 1 is a block diagram of the gamma correction circuit 106 according to the first embodiment, Fig. 2 is a diagram showing the relationship between input and output of the gamma correction circuit 106, and Fig. 13 shows the density of the background and image of various originals. This is a diagram showing In addition, in Figure 2, "g1"
is when no density correction processing is performed, that is, when the gamma correction circuit lO6 is in the through state, "g2" is when only background removal processing is performed, and "g3" is when density correction calculation is further performed. It shows the relationship between input and output. In FIG. 13, each image data D27 to
The white frames in the figure indicate the density range of the base of each document (base density range), and the shaded frames indicate the density range of the image (image density range). ing. For example, in the case of a normal printed matter, the background density range is a value of "0" to "16" near the minimum level, and the image density range is a value of r242J to r255J near the maximum level. In this case, if the difference between the background density and the image density is large, it is possible to form a clear image without performing density correction processing. On the other hand, in the case of newspapers, the base density range is larger than that of ordinary printed matter.
~ "32", and the image density range is rl44J~r
The value is in the range of l80J and is smaller than that for printed matter.
1. In other words, compared to regular printed matter, newspapers have a darker background color and a lighter image, so the contrast between the background and the image is low. Therefore, in order to form a clear copy image of such a document, density correction processing is required to enlarge the difference between the value corresponding to the background and the value corresponding to the image. In FIG. 1, the gamma correction circuit 106 includes a density correction calculation unit 502 that increases or decreases the value of image data based on density coefficient data GCD7-0, and background data UND7-0.
a background removal unit 50 that reduces the value of image data based on
It has 1. The density coefficient data GCD7-0 and the background data UND7-O are used as the image signal VtDE corresponding to the original D in the density determination process to be described later performed by CPUll2.
It is calculated according to the maximum density and the background density in a predetermined area of the document D detected during the preliminary scan performed before the scan (main scan) for outputting O4 to O0. Concentration coefficient data GCD7
~0 is treated as a positive decimal number in which the most significant bit and the lowest 7 bits of the 8 bits are assigned to the 1st integer place and the 1st to 7th places after the decimal point, respectively. The image data 027 to 20 inputted from the shading correction circuit 104 are first subjected to a background removal process in a background removal section 50l. For example, in the example shown in Figure 2, the underlying data UND7-0 is "3".
0", and when subjected to the background removal process, the input image data D27-20 are reduced by the value "30" of the background data UND7-O, and the second
``g2'' in the figure is translated by ``30'' to the right with respect to ``gl''. In Figure 1, the base data UND7
~0 is converted into negative data by the two's complement circuit 511 of the background removal unit 501 and added to the adder 512, and the adder 512 adds the positive image data D57-50 and the negative background data UND7-0. The calculation is performed. In the density correction calculation unit 502, data is increased or decreased by a combination of multiplication and addition in order to realize a wide range of multi-step correction by processing a limited number of bits (8 bits, the same as image data). In other words, concentration calculation section 502
is image data 0 input via the background removal unit 501
A multiplier 521 that multiplies 27 to 20 by density coefficient data GC01 to 0, a two's complement circuit 522 that converts the output of the multiplier 521 into negative data, and a grayscale control enable signal D.
The output of multiplier 521 or 2 according to EN - DOWN
gradation selector 523 that selects the output of the complement circuit 522
, the input image data D27 to D20 and the shade selector 5
The input image data D27 to D20 are corrected in the range of 0 to 3 times. When forming a dark image, enable signal DEN
- DOWN becomes rH, and at this time, the shade selector 5
23 selects the output of the multiplier 521. In addition, when the operator specifies the formation of a pale copy image, the enable signal DEN-DOWN is set to r.
LJ, and at this time the gradation selector 523 selects the output of the two's complement circuit 522. As a result, the adder 524 uses Do
An addition operation expressed as -DifyD+ is performed, and by changing the density coefficient data GC01 to 0(r), virtually stepless density correction can be performed. As is clear from the comparison between "g3" and "g1" in FIG. 2, input image data D27 to D20 are base data UND.
When it is smaller than 7~0, the output image data D37~
30 is rQJ, and when the input image data 027-20 exceeds the background data UND7-0, the output image data D37-30 increases with a slope greater than 1, thereby reducing the density difference between the background and the image. is expanded. next,
Base data UND7 to 0 given to the gamma correction circuit 106
and a CPU 11 that calculates concentration coefficient data GCD7 to GCD0.
Let us explain the second operation. FIG. 3 is a flowchart of the density determination process, FIG. 8 is a block diagram of the CPU peripheral section 20, and FIG.
The figure is a block diagram of the line mesori peripheral section 30. 8th
In the figure, the CPU 112 includes the above-mentioned ROM 113, RAM 114 used for temporary storage of calculation parameters, etc., via the CPU address bus and the CPU data path.
and interface 1 with parallel input/output ports
It is connected to 15. The decoder 116 decodes the address from the CPU 112 and generates chip select signals CSI to CS3 for accessing the ROMLL3, RAM114 and interface 115, respectively, and the line memory 111 and CPt.
It outputs the signal LMCS that controls the connection with the J data bus. Output ports PA7 to 0 of interface 115
, PB7-0 are connected to the gamma correction circuit 106 via latch circuits 117, 118, respectively, and base data UND7-0 and density coefficient data GC are transmitted from each boat.
D7~0 will be output. Also, output port P
A data clear signal DCLR is output from C7, and when this signal DCLR is "L", data of "0" is output from the launch circuit 117 as base data UND7-0.

Note that the CPU peripheral section 20 includes each image processing circuit 101 to
110, there is provided a clock generation circuit 119 that generates various clock signals for controlling each section, such as a clock signal SYNCK that serves as a reference for transmitting image data for each pixel. In FIG. 9, the line memory peripheral section 30 includes a line memory 111, this line memory 111
Selector 121 for selecting image data to be stored in
, an address counter 123 that increments in synchronization with the clock signal SYNCK, and a selector 1 that selects an address from the address counter 123 or an address from the CPU 112 to specify an address for the line memory 111.
22, and a bus gate circuit 124 controlled by the above-mentioned signal LMCS. In FIG. 3, the CPU 112 outputs an image signal VIDEO.
Before the scan (main scan) for outputting 4 to 0, a preliminary scan is performed in order to detect the degree of shading of the original D, that is, the density of the image and the density of the background. That is, in step #11, the slider 14 is moved to a predetermined position, for example,
Move it to near the center of document D. Next, in step #12, the exposure lamp 17 is turned on and one line of image data D27 to D20 output from the shading correction circuit 104 is stored in the line memory 111. That is, at this time, the write signal LMWR and switching signal MPX output from the output ports PC5 and PC4 of the interface 115 respectively become "L", the selector 122 selects the address from the address counter 123, and the selector 121 selects the image as input. Data 02
Select 7-20. In step #13, the CPU 112 sends the write signal LM
WR and signal LMCS are rH, and the line memory 11 is configured to sample image data of every 16 pixels from one line of image data using the CPU address bus.
1 and imports image data from the line memory 141 to the CPU 112 through the CPU data path. As a result, sampling is performed for one line area of the original D along the main scanning direction at intervals of approximately 1 mm. Subsequently, in step #l4, approximately 3 sampled
Divide the image data D27-20 for 00 pixels into 16 blocks, and select the individual maximum data with the largest value and the individual minimum data with the smallest value from the image data D27-20 in each block. do. Then, the average value (max) of the selected 16 individual maximum data is determined, and the obtained average value (max) is taken as the maximum data within one line area of document D, and the average value of the 16 individual minimum data is determined. (min) is determined, and the obtained average value (min) is set as the minimum data within one line area of document D. This maximum data corresponds to the density of the image of document D. Next, in step #15, a predetermined value "α" is added to the minimum data obtained in step #l4, and the run-dmm
"in+cr" and the obtained value "rund" is added to the manuscript D.
The background data UND7-0 corresponding to the density of the background are set to the output ports PA7-0 of the interface 115. The value "α" here is a correction constant determined experimentally to cope with fluctuations in the individual minimum data caused by background density 4A, etc. In step #16, the density coefficient data GCD is 7 to 0. Find the value "x" that satisfies the following equation (1), and obtain the value "
x” to the output ports P87 to 0 of the interface 115
Set to . (max-und)X (1+x)=255-(1)
Finally, in step #17, the above-mentioned enable signal DEN-DOWN and enable signal DCLR are set to rH, and the process ends. When the main scan is started, the gamma correction circuit 106 performs the background removal process and the density correction calculation as described above, and the image league part IR
The background is light (white), and the image is dark (black).
Image signal VIDEO4 for forming a clear copy image
~0 is output. FIG. 10 is a block diagram of a gamma correction circuit 106a according to the second embodiment, and FIG. 11 is a diagram showing the relationship between input and output of the gamma correction circuit 106a. In FIG. 10, the gamma correction circuit 106a includes a background removal unit 601 that reduces the value of image data based on the background data UND7-0, and a density correction operation that increases the value of the image data based on the density coefficient data GC07-0. It consists of a section 602. The background removal unit 601 extracts the image data D27 to D20 input from the shading correction circuit 104 and the background data UND7 to O.
and a selector 6 that selects and outputs image data D27-20 or white data WHD7-0 that always has a value of "0" according to the output of the comparator 611.
It consists of 12. In addition, the density correction calculation section 60
2 is a multiplier 621 that multiplies the image data 021 to 20 input via the background removal unit 601 and the density coefficient data GC07 to GC0, and the output of the multiplier 621 and the input image data D27 to 20. It consists of an adder 624 that performs an operation to increase the input image data D27 to D20 by a factor of 1 to 3. The image data D27 to D20 inputted from the shading correction circuit 104 are first subjected to a background removal process in a background removal section 60l. When the image data D27Y20 input from the shading correction circuit 104 is larger than the base data UND7~O, the output of the comparator 611 becomes rl,j, and the selector 612 selects the image data 027~20 as output. ．． Conversely, image data 027 to 20 are base data UND
7 to 0, the output of comparator 611 is r
H J, selector 612 selects white data WH07
Select ~0 as the output.

Image data D27 to D20 output from the selector 612
is then subjected to density correction calculation processing in the density correction calculation unit 602. Similar to the density correction calculation unit 502 of the first embodiment, the density correction calculation unit 602 increases data by a combination of multiplication and addition in order to achieve a wide range of correction with processing using a limited number of bits. That is, in the adder 624,
Output data of selector 612, concentration coefficient data GCD7
~0, the output image data D37-30 are Di, γ, respectively.
, DO, an addition operation expressed as Do=Di+yDi is performed. "g4" in FIG. 11 indicates the input/output of the gamma correction circuit 106a when the background data UND7-O and the density coefficient data GCD7-0 are both "O", that is, when the density correction processing is not substantially performed. "g5" in the same figure shows the characteristics of
2 shows the input/output characteristics when background data UND7-0 and density coefficient data GCD7-0 of predetermined values are given, and density correction processing consisting of both background removal processing and density correction calculation processing is performed. As is clear from the comparison between 'g4J and 'g5', the density correction process reduces the image data D27 to 2o corresponding to the background to '0', and increases the image data 027 to 20 corresponding to the image. In the output image data D37 to D30, the density difference between the background and the image is expanded more than in the input image data D27 to D20.

Next, the base data UN given to the gamma correction circuit 106a
C to calculate D7-O and concentration coefficient data GCD7-O
The operation of the PU 112 will be explained.

FIG. 12 is a flowchart of concentration determination processing according to the second embodiment. In FIG. 12, steps #11 to #l5, which are given the same reference numerals as those in FIG. 3, are the same as the density determination process according to the first embodiment, so the explanation thereof will be omitted. In step #21 following step #l5, a value "y" that satisfies the following equation (2), which becomes concentration coefficient data GCD7 to 0, is
is determined and the obtained value yJ is set in the output ports PB7 to PB0 of the interface 115.

(max)X (1+y)=255... (2) Then, in step #22, the enable signal DCLR is set to rH, and the process ends.

When the main scan starts, the gamma correction circuit 106a performs
Output boat PA7~O, PB7~O set in base data UND7~0, density coefficient data GC01~0
Based on this, the gamma correction circuit 106a performs background removal processing and density correction calculation as described above, and the image league part IR indicates that the background is light (white) and the image is dark (white).
(black) Image signal VI that enables the formation of clear copied images
DE04-O are output.

In the above-mentioned embodiment, the slider 14 is stopped at a predetermined position during preliminary scanning, and pixels within one line area of the document D are sampled to detect the density of the document D. The area of the document D to be sampled for this purpose can be set arbitrarily. That is, the background data UND7-0 and the density coefficient data GCD7-0 may be calculated from image data for pixels in a plurality of line areas or the entire area of the document D. Further, the background data UND7-0 and the density coefficient data GCD7-0 may be reset as appropriate during the main scan according to the density of each specific area of the document D. According to this, in a document D in which the background or image density changes greatly within one document D, for example, a dark character image is formed by printing in the first half region, and a light character image handwritten with a pencil is formed in the second half region. Even in the case of the document D in which the image is formed, the density is adjusted for each area as appropriate, and a clear copy image can be obtained over the entire area.

In the above-mentioned embodiment, it was explained that a preliminary scan is performed to detect the density of the document D before the main scan.
When performing a preliminary scan to detect the size of the document D, the density as well as the size of the document D may be detected during the preliminary scan. In the above embodiment, the density of the document D is detected by separately performing a preliminary scan before the main scan. In addition to performing the detection, density correction processing may be performed on subsequent image data based on the detected density. [Effects of the Invention] According to the present invention, it is possible to form a clear image with good contrast regardless of the density of the background.

[Brief explanation of the drawing]

The drawings show embodiments of the present invention, and FIG. 1 is a block diagram of a gamma correction circuit according to the first embodiment, FIG. 2 is a diagram showing the relationship between human power and output of the gamma correction circuit of FIG. 1, and FIG. The figure is the first
A flowchart of the density determination process according to the embodiment, FIG. 4 is a perspective view showing the optical system of Image League incorporated in a digital copying machine, FIG. 5 is a plan view of the image sensor, and FIG. 6 is the CCD of FIG. 5. An enlarged view of the sensor chip, Fig. 7 is a block diagram of the image league, Fig. 8 is a block diagram of the CPU peripheral part, Fig. 9 is a block diagram of the line memory peripheral part, and Fig. 10 is a gamma diagram according to the second embodiment. A block diagram of the correction circuit, FIG. 11 is a diagram showing the relationship between the input and output of the gamma correction circuit of FIG. 10, FIG. 12 is a flowchart of the density determination process according to the second embodiment, and FIG. 2 is a diagram showing the density range of the background and image. 1-1... Image sensor, 112... CPU
(density determination means), 501, 601...base removal section (
(background removal means), 502, 602...Density correction calculation section (density correction calculation means), D...Original, D27-20.
...Image data, 037-30...Image data (corrected image data), GCD7-0...Density coefficient data, I
R... Image reading device, UND7-0... Base data, VIDEO4-O... Image signal. applicant
Minolta Camera Co., Ltd. Fig.

Claims (1)

[Claims] (1) A one-dimensional image sensor that scans a document divides the document into pixels and reads the document, performs image processing on the image data obtained by quantizing the output of the one-dimensional image sensor, and outputs an image signal corresponding to each pixel. In an image reading device, a density coefficient is configured to correct the image data by detecting a maximum density and a background density in a predetermined area of the original before main scanning of the original is performed to output the image signal. density determination means for calculating data and background data corresponding to the background density; background removal means for reducing the image data based on the background data during the main scanning; and background removal based on the density coefficient data. An image reading device comprising: density correction calculation means for performing a calculation to increase the output from the means and outputting corrected image data.