JPH06290267A - Image processor - Google Patents

Abstract

(57) [Abstract] [Purpose] When converting an image in which a halftone image and a non-halftone image are mixed into a binarized image, the image quality of the binarized image is improved. [Structure] The corrected digital image signal DVa is output to the white pixel determination unit 2
4 and the black pixel determination unit 25, and the character area is determined using the determination results. For the pixel determined as the character region, the determination result of the black pixel determination unit 25 is used as the binarized image signal BW.
And the other pixels are output as the halftone binarization unit 26.
Is output as a binarized image signal BW.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for binarizing multi-valued image data obtained by raster-scanning a document image to form a binarized image signal with good image quality.

[0002]

2. Description of the Related Art For example, in an image reading apparatus such as a facsimile apparatus, an image processing apparatus that binarizes multivalued image data obtained by raster scanning an original image to form a binarized image signal The dither binarization process and the halftone binarization process using the error diffusion method are applied so that the halftone image can be read and input with high image quality.

However, when a halftone image and a non-halftone image such as a character are mixed in a document whose image is to be read, halftone binarization processing is uniformly performed on the document image. Then, there arises a disadvantage that the reading result of the non-halftone image is deteriorated.

Therefore, conventionally, for example, the halftone binarization process and the non-halftone binarization process are executed in parallel, and the feature of the character image is detected to determine whether or not the pixel of interest is a character region. For the pixels that are determined to be in the character area by performing the image area determination processing, the result of the non-halftone binarization processing is selected and output as a binarized image signal, and it is determined that it is not the character area. By selecting the halftone binarization result for the selected pixels and outputting it as a binarized image signal, the original image in which halftone images and non-halftone images are mixed can be read with higher image quality. What has made it possible is proposed.

[0005]

However, such a conventional device has the following disadvantages.

That is, the processing amount of the image area determination processing is large,
There have been problems that the circuit configuration becomes complicated and the device cost becomes high, or the accuracy of the image area determination does not increase even though complicated processing is executed, and the expected effect cannot be obtained.

The present invention has been made in view of the above circumstances, and it is possible to obtain a binarized image of good quality, which has a relatively small amount of circuitry, can perform accurate image area determination. An object is to provide an image processing device.

[0008]

SUMMARY OF THE INVENTION According to the present invention, halftone binarization means for forming halftone binarization signals by subjecting multivalued image data obtained by raster scanning an original image to halftone binarization processing. And, the multi-valued image data is binarized with a plurality of threshold values,
A character area determination unit that determines whether or not the target pixel that is the processing target is included in the character area based on the processing result, and a target pixel that is determined that the character area determination unit is not included in the character area Regarding, regarding the halftone binarizing signal output by the halftone binarizing means and outputting as a binarized image signal, the character area determining means determines that the character area is included in the character area. Regarding the pixel, selection is made by selecting, from among the binarization processing results by the plurality of threshold values executed by the character area determination means, a signal representing the binarization processing result of determining the black pixel and outputting it as a binarized image signal. It is equipped with means.

The multi-valued image data obtained by raster-scanning the original image is subjected to a half-tone binarization process to form a half-tone binarized signal, and the multi-valued image data. A character area determining unit that performs binarization processing with a plurality of threshold values, and determines whether or not the target pixel to be processed is included in the character area based on the processing result, and the character area determining unit is a character area. For the pixel of interest determined not to be included in, the halftone binarization signal output by the halftone binarization means is selected and output as a binarized image signal, while the character area determination means For the pixel of interest determined to be included in the region, a signal representing the binarization processing result in which the black pixel is determined is selected from the binarization processing results by the plurality of thresholds executed by the character region determination unit. Output as a binary image signal A selection means that, in which an operation means for setting operating the plurality of thresholds used by the character region determining means.

Further, halftone binarization means for forming halftone binarization signals by subjecting multivalued image data obtained by raster scanning an original image to halftone binarization processing, and the multivalued image data. Black pixel binarization means for binarizing with a predetermined black pixel determination threshold, white pixel binarization means for binarizing the multi-valued image data with a predetermined white pixel determination threshold, and the black pixel binarization A character area determining means for determining whether or not the target pixel to be processed is included in the character area based on the respective binarization results of the binarizing means and the white pixel binarizing means; For the pixel of interest determined by the area determination means as not included in the character area, the halftone binarized signal output by the halftone binarized means is selected and output as a binarized image signal, while It is determined that the character area determination means is included in the character area. The eye pixels, in which a selection means for outputting a binarized image signal by selecting a signal representing a binarization processing result of the black pixel binarization means.

Further, the multi-valued image data obtained by raster-scanning the original image is subjected to a half-tone binarization process to form a half-tone binarized signal, and the multi-valued image data. Black pixel binarization means for binarizing with a predetermined black pixel determination threshold, white pixel binarization means for binarizing the multi-valued image data with a predetermined white pixel determination threshold, and the black pixel binarization A character area determining means for determining whether or not the target pixel to be processed is included in the character area based on the respective binarization results of the binarizing means and the white pixel binarizing means; For the pixel of interest determined by the area determination means as not included in the character area, the halftone binarized signal output by the halftone binarized means is selected and output as a binarized image signal, while It is determined that the character area determination means is included in the character area. For the eye pixel, a selection unit that selects a signal representing the binarization processing result of the black pixel binarization unit and outputs the signal as a binarized image signal, and a setting operation of the black pixel determination threshold value and the white pixel determination threshold value It is provided with an operating means for operating.

Further, the multi-valued image data obtained by raster-scanning the original image is subjected to a half-tone binarization process to form a half-tone binarized signal, and the multi-valued image data. A binarization process is performed with a plurality of thresholds, and a character region determination unit that determines whether or not the pixel of interest to be processed is included in the character region based on the process result, and the multi-valued image data is set to a predetermined value. For the non-halftone binarization processing means for performing the binarization processing with the threshold value and the pixel of interest determined by the character area determination means not included in the character area, the halftone output by the halftone binarization means. While the binarized signal is selected and output as a binarized image signal, the target pixel determined by the character region determination means to be included in the character region is processed by the non-halftone binarization processing means. Binarized image signal by selecting the signal that represents the result Selection means for and outputting, in which an operation means for setting operating a predetermined threshold value used by the plurality of threshold values and the non-halftone binarization means used by the character region determining means.

Further, the multi-valued image data obtained by raster-scanning the original image is subjected to a half-tone binarization process to form a half-tone binarized signal, and the multi-valued image data. Black pixel binarization means for binarizing with a predetermined black pixel determination threshold, white pixel binarization means for binarizing the multi-valued image data with a predetermined white pixel determination threshold, and the black pixel binarization The binarization results of the binarizing unit and the white pixel binarizing unit are extracted for a predetermined pixel region including a plurality of pixels including the target pixel to be processed, and white pixels are continuously formed in the pixel region. When the black pixels are continuous, it is determined that the target pixel is included in the boundary area of the character image, and the character area determination means is not included in the character area. For the pixel of interest, see The halftone binarized signal output by the digitizing unit is selected and output as a binarized image signal, while the target pixel determined by the character region determination unit to be included in the character region is the black pixel binary signal. It is provided with a selecting means for selecting a signal representing the binarization processing result of the binarizing means and outputting it as a binarized image signal.

The multi-valued image data obtained by raster-scanning the original image is subjected to a half-tone binarization process to form a half-tone binarized signal, and the multi-valued image data Black pixel binarization means for binarizing with a predetermined black pixel determination threshold, white pixel binarization means for binarizing the multi-valued image data with a predetermined white pixel determination threshold, and the black pixel binarization The binarization results of the binarizing unit and the white pixel binarizing unit are extracted for a predetermined pixel region including a plurality of pixels including the target pixel to be processed, and white pixels are continuously formed in the pixel region. When the black pixels are continuous, it is determined that the target pixel is included in the boundary area of the character image, and the character area determination means is not included in the character area. For the pixel of interest, see The halftone binarized signal output by the digitizing unit is selected and output as a binarized image signal, while the target pixel determined by the character region determination unit to be included in the character region is the black pixel binary signal. Selecting means for selecting a signal representing the binarization result of the binarizing means and outputting it as a binarized image signal; and operating means for setting and inputting the size of the predetermined pixel area applied by the character area judging means. Be prepared.

Further, the multi-valued image data obtained by raster-scanning the original image is subjected to a half-tone binarization process to form a half-tone binarized signal, and the multi-valued image data. Black pixel binarization means for binarizing with a predetermined black pixel determination threshold, white pixel binarization means for binarizing the multi-valued image data with a predetermined white pixel determination threshold, and the black pixel binarization The binarization results of the binarizing unit and the white pixel binarizing unit are extracted for a predetermined pixel region including a plurality of pixels including the target pixel to be processed, and white pixels are continuously formed in the pixel region. When the black pixels are continuous, it is determined that the target pixel is included in the boundary area of the character image, and the character area determination means is not included in the character area. For the pixel of interest, see The halftone binarized signal output by the digitizing unit is selected and output as a binarized image signal, while the target pixel determined by the character region determination unit to be included in the character region is the black pixel binary signal. Selection means for selecting a signal representing the binarization processing result of the binarization means and outputting it as a binarized image signal; and the predetermined value applied by the character area determination means according to the reading resolution of the multivalued image data. And a setting means for setting the size of the pixel area.

[0016]

Therefore, the image area discrimination processing for checking whether or not the pixel of interest is located at the boundary portion of the character area is executed based on the binarization processing results with a plurality of threshold values, and based on the image area discrimination processing results. Since the non-halftone binary image signal or the halftone binary image signal is selected, the processing contents for image area discrimination are relatively simple and the characteristics of the original image are improved. Since it can be extracted, the accuracy is good. As a result, the image of the original in which the non-halftone image and the halftone image are mixed can be read with high quality, and the device configuration can be realized at a relatively low cost. Further, since the characteristics of the image area discrimination processing can be changed appropriately according to the reading resolution of the image to be processed, it is possible to obtain a higher quality image.

[0017]

Embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

FIG. 1 shows a group 3 facsimile apparatus according to an embodiment of the present invention.

In the figure, the control unit 1 performs control processing of each unit of the facsimile apparatus and facsimile transmission control procedure processing, and the system memory 2 is
The control unit 1 stores a control processing program executed by the control unit 1 and various data necessary for executing the processing program, and constitutes a work area of the control unit 1. The parameter memory 3 is a group memory. It is for storing various information unique to the facsimile machine.

The scanner 4 is for reading an original image at a predetermined resolution, the plotter 5 is for recording and outputting an image at a predetermined resolution, and the operation display unit 6 is a facsimile device. For operation,
It consists of various operation keys and various indicators. The operation display unit 6 also includes image reading conditions of the scanner 4 and operation keys for setting and inputting the type of image to be read.

The encoding / decoding unit 7 is for encoding and compressing the image signal and for decoding the encoded and compressed image information into the original image signal. The image storage device 8 is
This is for storing a large amount of image information in a coded and compressed state.

The group 3 facsimile modem 9 is for realizing the modem function of the group 3 facsimile, and has a low-speed modem function (V.21 modem) for exchanging transmission procedure signals and mainly exchanging image information. High-speed modem function (V.33 modem, V.33
29 modem, V.29. 27ter modem etc.).

The network control device 10 is for connecting the facsimile device to a public telephone line network and has an automatic call originating / receiving function.

These control unit 1, system memory 2,
The parameter memory 3, scanner 4, plotter 5, operation display unit 6, encoding / decoding unit 7, image storage device 8, group 3 facsimile modem 9, and network control device 10
It is connected to the system bus 11, and data is exchanged between these respective elements mainly in the system bus 1.
It is done through 1.

Data is exchanged directly between the network control device 10 and the group 3 facsimile modem 9.

FIG. 2 shows an example of the signal processing system of the scanner 4. It should be noted that, in the same drawing, a portion not directly related to the present invention (for example, a portion for adjusting the timing between the processing elements) is omitted.

In FIG. 1, an image input unit 20 raster-scans a read original image and reads it at a predetermined resolution. An analog image signal AV obtained as a result of the reading is sent to an analog / digital converter 21. Has been added.

The analog / digital converter 21 is for converting the input analog image signal AV into a corresponding digital image signal DV having a predetermined number of bits. The digital image signal DV is subjected to γ (gamma) correction processing. Added to section 22.

In the γ correction processing unit 22, the relationship between the input data and the output data becomes the relationship shown by the curve CC1 in FIG.
The correction processing is executed, and its output data is added to the spatial image processing unit 23.

The spatial image processing unit 23 executes an MTF (spatial frequency) correction calculation process for eliminating the blur of the image and an edge emphasis calculation process for emphasizing the edge portion of the image,
The output data is for eliminating the deterioration of the image, and its output data is the corrected digital image signal DVa, which is the white pixel determination unit 24, the black pixel determination unit 25, and the halftone binarization unit 26.
Has been added to.

A white pixel threshold value THw output from the control unit 1 is added to the white pixel determination unit 24. The white pixel determination unit 24 compares the input corrected digital image signal DVa with the white pixel threshold THw, and when the value of the corrected digital image signal DVa is equal to or more than the white pixel threshold THw, the pixel at that time is set as the white pixel. And the output is set to the logical H level (= data “1”), and when the value of the corrected digital image signal DVa is smaller than the white pixel threshold THw, the output is set to the logical L level ( = Data “0”). The output of the white pixel determination unit 24 is added to the white pixel pattern matching processing unit 27 and the black pixel pattern matching processing unit 28 as white pixel determination data DW.

Here, the corrected digital image signal DVa takes a value according to the brightness of the corresponding pixel (see FIG. 3). Therefore, when the pixel is closer to black, the value of the corrected digital image signal DVa is smaller, and when the pixel is closer to white, the value of the corrected digital image signal DVa is larger.

A black pixel threshold THb output from the control unit 1 is added to the black pixel determination unit 25. The black pixel determination unit 25 compares the input corrected digital image signal DVa with the black pixel threshold THb, and when the value of the corrected digital image signal DVa is less than or equal to the black pixel threshold THb, the pixel at that time is set as a black pixel. And the output is set to data “1” and the corrected digital image signal DVa
Is larger than the black pixel threshold value THb, the output is set to data “0”. The output of the black pixel determination unit 25 is as black pixel determination data BWs,
It is added to the black pixel pattern matching processing unit 28 and one input end of the selector 29.

The halftone binarization unit 26 executes halftone binarization processing such as dither binarization processing and error diffusion processing based on the corrected digital image signal DVa. , The selector 2 as the halftone binarized image signal BWp
9 is added to the other input terminal.

The white pixel pattern matching processing section 27
The input white pixel determination data DW is arranged in a 3 × 3 size pixel matrix centered on the target pixel that is the processing target at that time, and the data pattern of the pixel matrix is as shown in FIGS. f) to check if it matches any one of the data patterns shown in
If they match, the output is set to data "1", and if they do not match, the output is set to data "0". The output data of the white pixel pattern matching processing unit 27 is added to the white active determination unit 30 as white pixel determination data DWa for determining that the target pixel at that time is a white pixel.

The black pixel pattern matching processing section 28
The input white pixel determination data DW and black pixel determination data BWs are arranged in a 3 × 3 size pixel matrix centered on the target pixel to be processed at that time, and the data pattern of the pixel matrix is 5 (a) ~
It is checked whether or not the condition that any one of the data patterns shown in (h) is satisfied and that none of the data patterns shown in FIGS. 6 (a) to 6 (f) is satisfied is satisfied. The output is set to data "1" when the condition is satisfied, and the output is set to data "0" when the condition is not satisfied. This black pixel pattern matching processing unit 28
Is output to the black active determination unit 31 as black pixel determination data DBa that determines that the pixel of interest at that time is a black pixel.

Here, FIGS. 4 (a) to 4 (f) and FIG. 5 (a)
6 (a) to 6 (f), ◯ represents white pixels, ● represents black pixels, ▲ represents one of which is a black pixel, and × represents. Indicates that the pixel is not inspected.

That is, the white pixel pattern matching processing unit 27 checks whether or not the target pixel is included in the continuous white region, and the black pixel pattern matching processing unit 28 determines that the target pixel is continuous in black. It is to check whether it is included in the area.

The white active determination section 30 arranges the white pixel determination data DWa on a 3 × 3 size pixel matrix centering on the target pixel to be processed, and one or more data is placed in the pixel matrix. When "1" is included, the output is set to data "1", and when no data "1" is included in the pixel matrix, the output is set to data "0". Yes, its output is applied to one input terminal of the AND circuit 32 as white active determination data DWc.

The black active determination section 31 arranges the black pixel determination data DBa on a 3 × 3 size pixel matrix centered on the target pixel to be processed, and the pixel matrix has two or more data. When "1" is included, the output is set to data "1", and when the pixel matrix does not include two or more data "1", the output is set to data "0". The output is the black active judgment data DB.
It is added to the other input terminal of the AND circuit 32 as c.

The AND circuit 32, when the white active judgment data DWc becomes the data "1" and the black active judgment data DBc becomes the data "1", that is,
If a continuous area of white pixels exists around the pixel of interest and a continuous area of black pixels exists, its output becomes data "1". In other cases, its output becomes data "0". become. The output of the AND circuit 32 is added to the character area determination unit 33 as the active determination data AC.

When the mode data MD added from the control unit 1 is set to data "1", the character area determination unit 33 puts it on the 5 × 4 size pixel matrix shown in FIG. 7A. In addition to arranging the active judgment data AC, when the mode data MD is set to data “0”, the active judgment data AC is arranged on the 3 × 3 size pixel matrix shown in FIG.
When the pixel matrix includes one or more data "1", it is determined that the target pixel H located in the center of the pixel matrix is a pixel in the character area, and its output is converted to data "1". In addition to setting, the output is set to data "0" in other cases, and the output is added to the selector 29 as the selection signal SL.

When the selection signal SL is data "1", the selector 29 determines the black pixel determination data BWs.
And output as a binarized image signal BW,
When the selection signal SL is data "0", the halftone binarized image signal BWp is selected and output as the binarized image signal BW. The binarized image signal BW is output to the next stage device.

FIG. 8 shows an example of the character area determination unit 33.

In the figure, the active judgment data AC
Is applied to the data input terminal of the D flip-flop 35, and the data ACa of the pixel A shown in FIG.
Is added to one input terminal of the NOR circuit 36.
The output data of the D flip-flop circuit 35 is shown in FIG.
The data ACb of the pixel B shown in (a) is added to the data input terminal of the D flip-flop circuit 37 and to one input terminal of the NOR circuit 36 and the NOR circuit 38.

The output data of the D flip-flop circuit 37 is D as the data ACc of the pixel C shown in FIG.
It is added to the data input terminal of the flip-flop circuit 39 and to one input terminal of the NOR circuit 36 and the NOR circuit 38. The output data of the D flip-flop circuit 39 is the data AC of the pixel D shown in FIG.
As d, the data input terminal of the D flip-flop circuit 40,
It is added to one input end of the NOR circuit 36 and the NOR circuit 38 and the input end of the line buffer 41. D
The output data of the flip-flop circuit 40 is shown in FIG.
1 of the NOR circuit 36 as the data ACe of the pixel E shown in
Two inputs have been added.

The line buffer 41 has a capacity capable of storing data for one raster (line),
The output data is added to the data input terminal of the D flip-flop circuit 42 and one input terminal of the NOR circuit 43 as the data ACf of the pixel F shown in FIG.

The output data of the D flip-flop circuit 42 is D as the data ACg of the pixel G shown in FIG.
It is added to the data input terminal of the flip-flop circuit 44 and also to one input terminal of the NOR circuit 43 and the NOR circuit 45. The output data of the D flip-flop circuit 44 is the data AC of the pixel H shown in FIG.
is added to the data input terminal of the D flip-flop circuit 46 as h, and the NOR circuit 43 and the NOR circuit 45
Has been added to one input end of. The output data of the D flip-flop circuit 46 is the data input end of the D flip-flop circuit 47, one input end of the NOR circuit 43 and the NOR circuit 45, and the line as the data ACi of the pixel I shown in FIG. It is added to the input end of the buffer 48. The output data of the D flip-flop circuit 47 is
The data ACj of the pixel J shown in FIG. 7A is added to one input terminal of the NOR circuit 43.

The line buffer 48 has a capacity capable of storing data for one line, and its output data is the D flip-flop circuit 49 as the data ACk of the pixel K shown in FIG. 7A. Of the NOR circuit 50 and one input terminal of the NOR circuit 50.

The output data of the D flip-flop circuit 49 is D as the data ACl of the pixel L shown in FIG.
It is added to the data input terminal of the flip-flop circuit 51 and also to one input terminal of the NOR circuit 50 and the NOR circuit 52. The output data of the D flip-flop circuit 51 is the data AC of the pixel M shown in FIG.
m is added to the data input terminal of the D flip-flop circuit 53, and the NOR circuit 50 and the NOR circuit 52 are also added.
Has been added to one input end of. The output data of the D flip-flop circuit 53 is the data input terminal of the D flip-flop circuit 54, one input terminal of the NOR circuit 50 and the NOR circuit 52, and the line as the data ACn of the pixel N shown in FIG. It is added to the input end of the buffer 55. The output data of the D flip-flop circuit 54 is
The data ACo of the pixel O shown in FIG. 7A is added to one input terminal of the NOR circuit 50.

The line buffer 55 has a capacity capable of storing data for one line, and its output data is the D flip-flop circuit 56 as the data ACp of the pixel P shown in FIG. 7A. Of the NOR circuit 57 and one input terminal of the NOR circuit 57.

The output data of the D flip-flop circuit 56 is D as the data ACq of the pixel Q shown in FIG.
It is applied to the data input terminal of the flip-flop circuit 58 and also to one input terminal of the NOR circuit 57. The output data of the D flip-flop circuit 58 is shown in FIG.
The data ACr of the pixel R shown in (a) is added to the data input terminal of the D flip-flop circuit 59 and also to one input terminal of the NOR circuit 57. D
The output data of the flip-flop circuit 59 is shown in FIG.
The data ACs of the pixel S shown in (1) is added to the data input terminal of the D flip-flop circuit 60 and to one input terminal of the NOR circuit 57. The output data of the D flip-flop circuit 60 is added to one input terminal of the NOR circuit 57 as the data ACt of the pixel T shown in FIG.

The NOR circuit 36 outputs the data ACa, ACb,
The NOR operation is executed for ACc, ACd, and ACe, and the output data is applied to one input terminal of the NAND circuit 61. NOR circuit 43 receives data A
The NOR operation is executed for Cf, ACg, ACh, ACi, and ACj, and the output data is added to one input terminal of the NAND circuit 61. Noah circuit 5
0 is data ACk, ACl, ACm, ACn, ACo
With respect to, the output data is applied to one input terminal of the NAND circuit 61. The NOR circuit 57 uses the data ACp, ACq, ACr,
The NOR operation is executed for ACs and ACt, and the output data thereof is applied to one input terminal of the NAND circuit 61.

The NOR circuit 38 receives the data ACb, ACc,
ACd performs a NOR operation, and its output data is applied to one input terminal of the NAND circuit 62. The NOR circuit 45 outputs the data ACg, ACh, A.
A NOR operation is executed for Ci, and its output data is applied to one input terminal of the NAND circuit 62. The NOR circuit 52 uses the data AC1, ACm, AC.
The NOR operation is executed for n, and its output data is applied to one input terminal of the NAND circuit 62.

The output data of the NAND circuit 61 and the output data of the NAND circuit 62 are applied to the input terminals of the selector 63, respectively. The selector 63 operates the NAND circuit 61 when the value of the mode data MD is data “1”.
Output data is selected, and when the value of the mode data MD is data “0”, the NAND circuit 62
The output data of is selected, and the selected output is
It is output as the selection signal SL.

The pixel clock PC representing the output timing of one pixel is the D flip-flop circuit 35, 37, 39, 40, 42, 44, 46, 4 respectively.
7,49,51,53,54,56,58,59,60
Clock input terminal and line buffers 41 and 4
8, 55 are input to clock input terminals (not shown), and the reset signal RST is applied to the D flip-flop circuits 35, 37, 39, 40, 42, 44, respectively.
46, 47, 49, 51, 53, 54, 56, 58, 5
It is added to the reset input terminal of 9,60.

Therefore, each time the pixel clock PC is applied, the active determination data AC sequentially moves to the D flip-flop circuits 35, 37, 39, 40 and is accumulated in the line buffer 41. Further, the line buffer 41 outputs data one line before the active determination data AC, and this data is sequentially moved to the D flip-flop circuits 42, 44, 46, 47 every time the pixel clock PC is applied. At the same time, it is stored in the line buffer 48.

The line buffer 48 outputs the data two lines before the active determination data AC,
This data is stored every time the pixel clock PC is applied.
The data are sequentially moved to the D flip-flop circuits 49, 51, 53, 54 and are accumulated in the line buffer 55.

Further, the line buffer 55 outputs the data three lines before the active determination data AC,
This data is stored every time the pixel clock PC is applied.
The circuit sequentially moves to the D flip-flop circuits 56, 58, 59, 60.

In this way, each time the pixel clock PC is added, the data ACa, ACb, ACc, AC
d, ACe, ACf, ACg, ACh, ACi, AC
j, ACk, ACl, ACm, ACn, ACo, AC
The values of p, ACq, ACr, ACs, and ACt change sequentially.

Further, the output data of the NAND circuit 61 becomes data "0" when all the output data of the NOR circuits 36, 43, 50 and 57 are data "1", and otherwise the data. Since the data becomes "0", the NAND circuit 6
The output data of 1 is data ACa, ACb, ACc, A
Cd, ACe, ACf, ACg, ACh, ACi, AC
j, ACk, ACl, ACm, ACn, ACo, AC
When any one or more of p, ACq, ACr, ACs, and ACt becomes data “1”, the value becomes data “1”, and these data ACa, ACb, AC
c, ACd, ACe, ACf, ACg, ACh, AC
i, ACj, ACk, ACl, ACm, ACn, AC
When o, ACp, ACq, ACr, ACs, and ACt all become data “0”, the value becomes data “0”.

The output data of the NAND circuit 62 is data "0" when all the output data of the NOR circuits 38, 45, 52 are data "1", and is otherwise "0". Since it becomes "0", the NAND circuit 62
Output data is data ACb, ACc, ACd, AC
Of g, ACh, ACi, ACl, ACm, ACn,
When any one or more of them becomes data "1", the value becomes data "1", and these data ACb, ACc,
ACd, ACg, ACh, ACi, ACl, ACm, A
When all the Cn become data "0", the value becomes data "0".

Therefore, when the mode data MD added from the control unit 1 is set to data "1", one or more data "in the pixel matrix of 5 × 4 size shown in FIG. 1 ”active judgment data AC
, The selection signal SL is data “1”.
become.

Further, when the mode data MD added from the control unit 1 is set to the data "0", FIG.
When the 3 × 3 size pixel matrix shown in (b) includes the active determination data AC of one or more data “1”, the selection signal SL becomes the data “1”.

With the above configuration, in the state where the standard resolution is set from the operation display unit 6, the control unit 1 controls the mode data M
Data “0” is set in D and the white pixel threshold TH
The values of w and the black pixel threshold value THb are set to values corresponding to the standard resolution.

As a result, the character area determination section 33 is brought into a state of selecting the above-mentioned 3 × 3 size matrix, and the white pixel determination section 24 and the black pixel determination section 25 are
The operating state corresponds to the standard resolution.

In this state, for example, when the user inputs a copy operation command, the control unit 1 causes the scanner 4
The reading operation is performed, whereby the binarized image signal BW output from the scanner 4 is sequentially transferred to the plotter 5, and a copy image is recorded and output from the plotter 5.

At this time, as described above, at the boundary portion of the character area of the image and its vicinity, the character area determination unit 33
The value of the selection signal SL output from the data becomes data "1", whereby the black pixel determination data BWs is selected by the selector 29 in this portion, and the binarized image signal B is selected.
It is output as W.

In the other portions, that is, in the halftone image portion, the black solid portion and the white solid portion of the character, the value of the selection signal SL output from the character area determination unit 33 becomes data "0", As a result, in this portion, the selector 29 selects the halftone binarized signal BWp and outputs it as the binarized image signal BW.

Further, when the fine resolution is set from the operation display unit 6, the control unit 1 sets the data "1" in the mode data MD and sets the values of the white pixel threshold value THw and the black pixel threshold value THb. Set to a value that corresponds to the fine resolution.

As a result, the character area determination unit 33 is in a state of selecting the above-mentioned 5 × 4 size matrix, and the white pixel determination unit 24 and the black pixel determination unit 25 are
The operating state corresponds to the fine resolution.

In this way, in the present embodiment, basically, the image is binarized in halftone, and only the boundary portion of the character whose image quality is deteriorated when halftoning is fixed is fixed. Binarization processing is performed with a threshold value. Therefore, as for image area determination, since only the boundary portion of characters is targeted, the image area determination processing can be made relatively simple. Was able to reduce the complexity of.

Also, since the image area determination process is relatively simple and the probability of erroneous determination can be reduced, the types of applicable images are expanded, and good binary images can be obtained for a wide range of images. It was

By the way, for example, when the value of the white pixel threshold value THw is set to a smaller value, the ratio of the pixels determined as the white pixels by the white pixel determination unit 24 increases.
The area of pixels determined to be white active by the white active determination section 30 is expanded, and thus the pixel area determined to be a character area by the character area determination section 33 is increased.

Similarly, if the value of the black pixel threshold value THb is set to a larger value, the ratio of pixels determined by the black pixel determination unit 25 as black pixels increases, so the black active determination unit 31 determines black active. The area of pixels to be expanded is expanded, and thereby the pixel area determined as the character area by the character area determination unit 33 is increased.

Therefore, depending on the content of the image read by the scanner 4, for example, the image having many characters or the image having many photographs, a mode in which an operation input can be made is set, and the mode can be set according to the set mode. White pixel threshold T
By allowing the values of Hw and the black pixel threshold value THb to be changed, a binarized image with better image quality can be obtained. At this time, a suitable set of values of the white pixel threshold value THw and the black pixel threshold value THb is set in advance for a plurality of types of images, and when the image type is set,
White pixel threshold THw and black pixel threshold TH corresponding thereto
It is advisable to select a set of values of b.

The setting of the values of the white pixel threshold THw and the black pixel threshold THb is related to the extraction ability of the character area and the erroneous determination of the character area. Therefore, the values may be determined by, for example, an experiment. preferable.

Further, in the present embodiment, since the black pixel judging section 25 used for judging the image area is also used as the non-halftone binarizing means, the circuit scale can be reduced. Further, in the present embodiment, since the processing capability of the black pixel pattern matching processing unit 28 greatly contributes to the determination capability of the character area, the setting range of the value of the black pixel threshold THb is secured to be relatively large, and It is preferable to be able to adjust the pixel determination in a larger range.

Further, the number of pixels obtained in the same area is determined by the image reading resolution, and naturally, the larger the value of the resolution, the larger the number of pixels obtained. In the above-described embodiment, the pixel matrix size applied by the character area determination unit 33 is 5 × when the resolution is fine.
The reason why 4 is set and 3 × 3 is set in the case of the standard resolution is that the determination operation of the character region can be executed for the same area in the case of the fine resolution and the case of the standard resolution.

Therefore, when different resolutions are used, it is preferable to use the pixel matrix size applied by the character area determination unit 33 that corresponds to the resolution.

Now, the γ correction processing is for realizing effective gradation expression with a small number of bits in consideration of human visual characteristics, and it is an image of a portion of high density where small changes cannot be recognized. Of the image is reduced, and instead, the resolution of the image of the portion where the change is more noticeable and the density is small is improved (see FIG. 3).

Therefore, when the image data after the γ correction processing is subjected to the non-halftone binarization processing, the image quality of the black portion of the image generally tends to deteriorate. In the apparatus shown in FIG. 2, since the image data after the γ correction processing is subjected to the non-halftone binarization processing, such an inconvenience may occur.

FIG. 9 shows a signal processing system of the scanner 4 according to another embodiment of the present invention, which can eliminate such inconvenience.
In the figure, the same parts as those in FIG. 2 and corresponding parts are designated by the same reference numerals.

In the figure, the digital image signal DV output from the analog / digital converter 21 is added to the γ correction processing unit 22 and the spatial image processing unit 23.

The output of the γ correction processing unit 22 is added to the spatial image processing unit 23a. The spatial image processing unit 23a
Similar to the spatial image processing unit 23, it is for eliminating image deterioration, and its output data is added to the halftone binarization unit 26 as a corrected digital image signal DVa ′.

The corrected digital image signal DVa output from the spatial image processing unit 23 is added to the white pixel determination unit 24 and the black pixel determination unit 25.

In this way, in this embodiment, since the signal path for non-halftone binarization does not include the γ correction processing section 22, the inconvenience as in the above-mentioned embodiment can be solved. .

FIG. 10 shows a signal processing system of the scanner 4 according to still another embodiment of the present invention. In the figure, the same parts as those in FIG. 2 and corresponding parts are designated by the same reference numerals.

In the figure, the corrected digital image signal DVa output from the spatial image processing unit 23 is the white pixel determination unit 2
4, black pixel determination unit 25, halftone binarization unit 26, and
It is added to the binarization unit 65.

A threshold THc output from the control unit 1 is added to the binarization unit 65. The binarization unit 65 compares the input corrected digital image signal DVa with the threshold THc, and when the value of the corrected digital image signal DVa is equal to or less than the threshold THc, determines the pixel at that time as a black pixel, The output is set to data "1", and when the value of the corrected digital image signal DVa is larger than the threshold value THc, the output is set to data "0". The output of the binarization unit 65 is applied to one input terminal of the selector 29 as a non-halftone binarized image signal BWd.

The selector 29 selects the non-halftone binarized signal BWd to select the binarized image signal BW when the selection signal SL output from the character area determination section 33 is data "1". In addition, when the selection signal SL is data “0”, the halftone binarized image signal BWp is selected and output as the binarized image signal BW.

Therefore, in this embodiment, since the non-halftone binarization process is executed by the independent binarization unit 65, the threshold value is set to a value that can eliminate the influence of the γ correction process executed in the preceding stage. By setting THc, it is possible to realize an appropriate non-halftone binarization process and eliminate the image quality deterioration of the binarized image of the character portion.

In the above-described embodiment, the present invention is applied to the scanner of the group 3 facsimile machine, but the present invention can be applied to other image reading apparatuses in the same manner.

Further, in the above-mentioned embodiment, two threshold values T
Although the image is discriminated using Hw and THb, more threshold values can be used.

[0095]

As described above, according to the present invention,
Based on the result of binarization processing with multiple thresholds, image area discrimination processing is performed to check whether the pixel of interest is located at the boundary of the character area, and based on the result of the image area discrimination processing, non-halftone Since the binarized image signal or the halftone binarized image signal is selected, the processing contents for image area discrimination are relatively simple and the characteristics of the original image can be extracted well. Therefore, the accuracy is good. As a result, the image of the original in which the non-halftone image and the halftone image are mixed can be read with high quality, and the device configuration can be realized at a relatively low cost. Further, since the characteristics of the image area discrimination processing can be changed appropriately according to the reading resolution of the image to be processed, it is possible to obtain an effect that a higher quality image can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing a group 3 facsimile apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a signal processing system of a scanner according to an embodiment of the present invention.

FIG. 3 is a graph for explaining characteristics of γ correction processing.

FIG. 4 is a schematic diagram showing an example of a match pattern of a white pixel pattern matching processing section.

FIG. 5 is a schematic diagram showing an example of a match pattern of a black pixel pattern matching processing section.

FIG. 6 is a schematic diagram showing an example of an unmatched pattern of a black pixel pattern matching processing unit.

FIG. 7 is a schematic diagram showing a pixel matrix applied in a character area determination unit.

FIG. 8 is a block diagram showing an example of a character area determination unit.

FIG. 9 is a block diagram showing a signal processing system of a scanner according to another embodiment of the present invention.

FIG. 10 is a block diagram showing a signal processing system of a scanner according to still another embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 control part 4 scanner 22 γ correction processing part 23, 23a spatial image processing part 24 white pixel determination part 25 black pixel determination part 26 halftone binarization part 27 white pixel pattern matching processing part 28 black pixel pattern matching processing part 29, 63 Selector 30 White active determination unit 31 Black active determination unit 32 AND circuit 33 Character area determination unit 65 Binarization unit

Claims (8)

[Claims] 1. Halftone binarization means for forming a halftone binarization signal by subjecting multivalued image data obtained by raster scanning an original image to halftone binarization processing, and the multivalued image data. A binarization process using a plurality of thresholds, and a character region determination unit that determines whether or not a pixel of interest to be processed is included in the character region based on the processing result; For the pixel of interest determined not to be included in, the halftone binarization signal output by the halftone binarization means is selected and output as a binarized image signal, while the character area determination means For the pixel of interest determined to be included in the region, a signal representing the binarization processing result in which the black pixel is determined is selected from the binarization processing results by the plurality of thresholds executed by the character region determination unit. And output as a binary image signal The image processing apparatus characterized by comprising a-option means. 2. Halftone binarization means for forming halftone binarization signals by subjecting multivalued image data obtained by raster-scanning an original image to halftone binarization processing, and the multivalued image data. A binarization process using a plurality of thresholds, and a character region determination unit that determines whether or not a pixel of interest to be processed is included in the character region based on the processing result; For the pixel of interest determined not to be included in, the halftone binarization signal output by the halftone binarization means is selected and output as a binarized image signal, while the character area determination means For the pixel of interest determined to be included in the region, a signal representing the binarization processing result in which the black pixel is determined is selected from the binarization processing results by the plurality of thresholds executed by the character region determination unit. And output as a binary image signal And-option unit, an image processing apparatus characterized by comprising an operation means for setting operating the plurality of thresholds used by the character region determining means. 3. Halftone binarization means for forming a halftone binarization signal by subjecting multivalued image data obtained by raster-scanning an original image to halftone binarization processing, and the multivalued image data. Black pixel binarizing means for binarizing with a predetermined black pixel determination threshold value; white pixel binarizing means for binarizing the multi-valued image data with a predetermined white pixel determination threshold value; Character area determining means for determining whether or not the target pixel to be processed is included in the character area based on the respective binarization results of the binarizing means and the white pixel binarizing means; For the pixel of interest determined by the area determination means as not included in the character area, the halftone binarized signal output by the halftone binarized means is selected and output as a binarized image signal, while Note that the character area determination means has determined that it is included in the character area. For pixels, the image processing apparatus characterized by comprising a selection means for outputting a binarized image signal by selecting a signal representing a binarization processing result of the black pixel binarization means. 4. Halftone binarization means for forming halftone binarization signals by subjecting multivalued image data obtained by raster-scanning an original image to halftone binarization processing, and the multivalued image data. Black pixel binarizing means for binarizing with a predetermined black pixel determination threshold value; white pixel binarizing means for binarizing the multi-valued image data with a predetermined white pixel determination threshold value; Character area determining means for determining whether or not the target pixel to be processed is included in the character area based on the respective binarization results of the binarizing means and the white pixel binarizing means; For the pixel of interest determined by the area determination means as not included in the character area, the halftone binarized signal output by the halftone binarized means is selected and output as a binarized image signal, while Note that the character area determination means has determined that it is included in the character area. As for the pixel, a selecting means for selecting a signal representing the binarization processing result of the black pixel binarizing means and outputting it as a binarized image signal, and setting operation of the black pixel determination threshold value and the white pixel determination threshold value An image processing apparatus comprising operating means. 5. A halftone binarizing means for halftone binarizing the multivalued image data obtained by raster scanning an original image to form a halftone binary signal, and the multivalued image data. A binarization process is performed with a plurality of threshold values, and a character region determination unit that determines whether or not the target pixel to be processed is included in the character region based on the process result, For the non-halftone binarization processing means for performing binarization processing with a threshold value and the pixel of interest determined by the character area determination means not included in the character area, the halftone output by the halftone binarization means is performed. While the binarized signal is selected and output as a binarized image signal, the target pixel determined by the character region determination means to be included in the character region is processed by the non-halftone binarization processing means. Select the signal that represents the result and use it as the binarized image signal. An image processing apparatus comprising: a selection unit configured to output a plurality of threshold values used by the character region determination unit; and an operation unit configured to set a predetermined threshold value used by the non-halftone binarization processing unit. . 6. Halftone binarization means for forming a halftone binarization signal by subjecting multivalued image data obtained by raster-scanning an original image to halftone binarization, and the multivalued image data. Black pixel binarizing means for binarizing with a predetermined black pixel determination threshold value; white pixel binarizing means for binarizing the multi-valued image data with a predetermined white pixel determination threshold value; The binarization results of the binarizing unit and the white pixel binarizing unit are extracted for a predetermined pixel region including a plurality of pixels including the target pixel to be processed, and white pixels are continuously formed in the pixel region. When the black pixels are continuous, it is determined that the target pixel is included in the boundary area of the character image, and the character area determination means is not included in the character area. For the pixel of interest, the above halftone binary The halftone binarized signal output by the means is selected and output as a binarized image signal, while the target pixel determined by the character area determination means to be included in the character area is the black pixel binary value. An image processing apparatus comprising: a selection unit that selects a signal representing a binarization result of the binarization unit and outputs the signal as a binarized image signal. 7. Halftone binarization means for forming halftone binarization signals by subjecting multivalued image data obtained by raster-scanning an original image to halftone binarization processing, and the multivalued image data. Black pixel binarizing means for binarizing with a predetermined black pixel determination threshold value; white pixel binarizing means for binarizing the multi-valued image data with a predetermined white pixel determination threshold value; The binarization results of the binarizing unit and the white pixel binarizing unit are extracted for a predetermined pixel region including a plurality of pixels including the target pixel to be processed, and white pixels are continuously formed in the pixel region. When the black pixels are continuous, it is determined that the target pixel is included in the boundary area of the character image, and the character area determination means is not included in the character area. For the pixel of interest, the above halftone binary The halftone binarized signal output by the means is selected and output as a binarized image signal, while the target pixel determined by the character area determination means to be included in the character area is the black pixel binary value. And a selecting means for selecting a signal representing the binarization result of the digitizing means and outputting it as a binarized image signal; and an operating means for setting and inputting the size of the predetermined pixel area applied by the character area determining means. An image processing device characterized by the above. 8. A halftone binarizing means for halftone binarizing the multivalued image data obtained by raster scanning an original image to form a halftone binary signal, and the multivalued image data. Black pixel binarizing means for binarizing with a predetermined black pixel determination threshold value; white pixel binarizing means for binarizing the multi-valued image data with a predetermined white pixel determination threshold value; The binarization results of the binarizing unit and the white pixel binarizing unit are extracted for a predetermined pixel region including a plurality of pixels including the target pixel to be processed, and white pixels are continuously formed in the pixel region. When the black pixels are continuous, it is determined that the target pixel is included in the boundary area of the character image, and the character area determination means is not included in the character area. For the pixel of interest, the above halftone binary The halftone binarized signal output by the means is selected and output as a binarized image signal, while the target pixel determined by the character area determination means to be included in the character area is the black pixel binary value. Selecting means for selecting a signal representing the binarization processing result of the digitizing means and outputting it as a binarized image signal; and the predetermined area applied by the character area determining means according to the reading resolution of the multivalued image data. An image processing apparatus comprising a setting means for setting a size of a pixel area.