JP2010187179A - Image processing apparatus, image forming apparatus, image processing method, computer program, and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of reducing the amount of data after compression for image data, and to provide an image forming apparatus, an image processing method, a computer program, and a recording medium. <P>SOLUTION: A color image processing apparatus 2 includes a compression processing apparatus 3. The compression processing apparatus extracts pixels of a character in an image and generates a foreground layer where a color of extracted pixels is expressed with an identifier (index). Then, the compression processor 3 generates a background layer where characters are excluded from the image, generates a binary image by identifier, and compresses the binary image and the background layer, respectively. When the foreground layer is generated, the compression processor 3 associates an identifier associated with a color once with each pixel, and detaches the identifier among pixels separated by a predetermined distance from one another, from among pixels associated with an identical identifier. Moreover, the compression processing apparatus unifies an identifier related with approximation colors other than the identical color. For this reason, unwanted data contained in data after compression can be reduced, when a rectangular region containing a pixel of the identifier is extracted from the binary image to be compressed. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, an image forming apparatus, an image processing method, a computer program, and a recording medium that can compress image data with high efficiency by indexing the color of each pixel.

  An image processing apparatus that optically reads an image recorded on paper or the like to generate image data is widely used as a scanner apparatus, a facsimile apparatus, or a copying apparatus. Image data generated by the image processing apparatus is transmitted and received using facsimile communication, electronic mail, or the like, or stored in a database and used for various purposes. Image data generated by optically reading an image generally has a large amount of data. Therefore, compression of image data is indispensable for efficient transmission / reception or storage.

  One compression technique for realizing a high compression rate is a compression technique based on layer separation. The compression technique based on layer separation separates an image into a foreground layer representing characters and / or line drawings and a background layer representing other images, and compresses the foreground layer and the background layer using a compression method suitable for each, This is a technique for finally generating a compressed image.

The background layer is generally compressed using a lossy compression technique such as JPEG (Joint Photographic Experts Group). Since the lossy compression technique is easy to control the compression rate, it is possible to reduce the amount of data or prevent deterioration of the image quality depending on the application.
On the other hand, the foreground layer is generally compressed using a reversible compression technique such as JBIG (Joint Bi-level Image Experts Group), MMR (Modified Modified Read), or LZW (Lempel Ziv Welch). In the reversible compression technique, it is difficult to control the compression rate, so it is difficult to improve the compression rate.

  Patent Document 1 discloses a compression technique that can improve the compression rate. In the technique described in Patent Document 1, the color of each pixel included in a color image is represented by an index, color information including the number of pixels for each color represented by the index, and background color information generated using them. , Index integration.

JP 2004-229261 A

  The technique described in Patent Document 1 represents the same color included in one image with the same index, and binarized / compressed by extracting a rectangular area from the maximum and minimum coordinates of each color pixel in each index. Is doing. Accordingly, since the rectangular area cut out for one index includes an area unrelated to this index, unnecessary data is included in the compressed data. As a result, there occurs a problem that the amount of data after compression increases and the compression efficiency is poor.

  The present invention has been made to solve such a problem, and a main object of the present invention is to reduce the amount of data after compression by making it possible to extract an area containing a high percentage of pixels related to each index. An image processing apparatus, an image forming apparatus, an image processing method, a computer program, and a recording medium are provided.

  An image processing apparatus according to the present invention includes an accepting unit that accepts an image composed of a plurality of pixels, and an extraction that extracts pixels corresponding to a character and / or a line drawing from among pixels constituting the image accepted by the accepting unit. Means, a foreground layer generating means for generating a foreground layer representing the color of each pixel extracted by the extracting means by any of a plurality of identifiers, and characters and / or line drawings from the image received by the receiving means. A binary for generating a binary image composed of a pixel represented by a specific identifier and other pixels for each identifier included in the foreground layer generated by the background layer generating unit that generates the background layer and the foreground layer generating unit. An image generation unit; and a compression unit that compresses the binary image generated by the binary image generation unit and the background layer generated by the background layer generation unit, respectively, and compresses the image In the image processing apparatus that performs processing, the foreground layer generation unit is the same by an identifier corresponding unit that associates an identifier associated with a specific color with each pixel according to the color of each pixel, and an association by the identifier corresponding unit. Among the pixels associated with the identifiers, identifier separating means for separating the identifiers by associating different identifiers between pixels separated from each other by a predetermined distance on the image, association by the identifier correspondence means, and the identifier separation After the separation by the means, there is provided an identifier integrating means for integrating identifiers whose colors associated with the identifiers are similar within a predetermined range among different identifiers.

  In the image processing apparatus according to the present invention, the identifier integration unit may include different identifiers except for identifiers in which the associated colors are the same after association by the identifier association unit and separation by the identifier separation unit. Among these, identifiers whose colors associated with the identifiers are approximated within a predetermined range are integrated.

  In the image processing apparatus according to the present invention, the identifier separating unit is associated with a selecting unit that sequentially selects pixels according to a predetermined selection order, and the same identifier as the currently selected pixel among the selected pixels. A minimum calculation means for calculating a minimum distance between a pixel and the selected pixel; and a minimum distance calculated by the minimum calculation means that is associated with the selected pixel when the minimum distance is greater than a predetermined threshold. And a means for changing the identifier to a new identifier associated with the same color as the identifier.

  The image processing apparatus according to the present invention is configured such that the selection unit sequentially selects pixels with one direction on the image as a main scanning direction and a direction orthogonal to the one direction as a sub-scanning direction, The distance calculated by the minimum calculation means further includes means for switching between a distance in the main scanning direction, a distance in the sub scanning direction, and a distance in both the main scanning direction and the sub scanning direction.

  The image processing apparatus according to the present invention is characterized in that the color of each pixel includes information indicating the color intensities of a plurality of primary colors.

  The image processing apparatus according to the present invention is characterized in that the color of each pixel includes information indicating a single color intensity.

  An image forming apparatus according to the present invention includes the above-described image processing apparatus.

  The image processing method according to the present invention includes a step of accepting an image composed of a plurality of pixels, a step of extracting pixels corresponding to a character and / or a line drawing among pixels constituting the accepted image, A foreground layer generating step of generating a foreground layer representing a pixel color by any of a plurality of identifiers; a step of generating a background layer from which characters and / or line drawings are omitted from the received image; and the foreground layer In the image processing method, the method includes: generating a binary image including a pixel represented by a specific identifier and other pixels for each identifier; and compressing the generated binary image and the background layer, respectively. In the layer generation step, according to the color of each pixel, an identifier associated with a specific color is associated with each pixel, and the same identifier is used. A step of separating identifiers by associating different identifiers between pixels that are spaced apart from each other by a predetermined distance on the image, and after the association and separation of the identifiers, the identifiers among the different identifiers And integrating the identifiers whose colors that are associated with each other are approximated within a predetermined range.

  The computer program according to the present invention indicates, to a computer, a procedure for extracting pixels corresponding to a character and / or a line drawing from among pixels constituting an image, and the color of each extracted pixel by any one of a plurality of identifiers. A foreground layer generation procedure for generating a foreground layer, a procedure for generating a background layer from which characters and / or line drawings are omitted from the image, a pixel represented by a specific identifier and other information for each identifier included in the foreground layer A computer program for executing a process of compressing an image by executing a process including a procedure of generating a binary image including pixels and a procedure of compressing the generated binary image and a background layer, respectively. The foreground layer generation procedure is the same as the procedure of associating each pixel with an identifier associated with a specific color according to the color of each pixel. A procedure for separating identifiers by associating different identifiers between pixels that are separated from a predetermined distance on the image among pixels associated with a different child, and after association and separation of identifiers, And a procedure for integrating identifiers in which colors associated with the identifiers are approximated within a predetermined range.

  A recording medium according to the present invention stores the above-described computer program.

  In the present invention, the image processing apparatus first extracts pixels corresponding to characters and / or line drawings from the received image, and then generates a foreground layer representing the color of the extracted pixels by an identifier, Further, a background layer in which characters and / or line drawings are omitted from the image is generated, and further, a binary image in which a pixel related to each identifier is distinguished from other pixels is generated from the foreground layer. A compressed image is generated by compressing the value image and the background layer, respectively.

When the foreground layer is generated, the image processing apparatus temporarily associates an identifier associated with a color with each pixel, and then differs between pixels separated by a predetermined distance among pixels associated with the same identifier. Separate identifiers by associating identifiers. For this reason, pixels that have the same color but are associated with different identifiers are included in different binary images, and in each binary image, the pixels associated with each identifier are concentrated in a narrower region. .
After the association and separation of the identifiers, the image processing apparatus integrates the identifiers associated with the approximate colors. As a result, since the number of identifiers decreases, the number of binary images to be generated decreases. That is, the disadvantage that the number of identifiers increases excessively due to the separation of identifiers is suppressed.

  In the present invention, the image processing apparatus integrates the identifiers associated with the approximate colors after the association and separation of the identifiers. However, when integrating identifiers, the image processing apparatus does not integrate identifiers associated with the same color. For this reason, the inconvenience of unnecessarily integrating the identifiers separated in the preceding process is suppressed.

In the present invention, the image processing apparatus sequentially selects pixels constituting the image. Hereinafter, an identifier that is the same as the identifier associated with the currently selected pixel is referred to as the same identifier.
Next, the image processing apparatus obtains a minimum distance between a pixel associated with the same identifier among the selected pixels and the currently selected pixel, and when the obtained minimum distance is larger than a predetermined threshold, The identifier associated with the currently selected pixel is changed to a new identifier associated with the same color as the same identifier. As a result, identifiers with a long distance between pixels are appropriately separated.
On the other hand, when the determined minimum distance is smaller than the predetermined threshold (that is, when the distance between the pixels is short), the identifier of the currently selected pixel is not changed, and no separation of the identifier occurs.

  In the present invention, as the distance between the pixels, a distance in the main scanning direction, a distance in the sub scanning direction, and a distance in both the main scanning direction and the sub scanning direction can be used, and any of them may be used as necessary. By switching between these, the identifiers are separated according to the situation, such as when the image read by the scanner is rotated 90 ° to the right.

  In the present invention, a color image can be compressed by associating the color intensities of a plurality of primary colors such as red, green, and blue with an identifier and performing an image compression process.

  In the present invention, by compressing an image by associating a single color intensity with an identifier, it is possible to compress a grayscale monochrome image.

In the case of the image processing apparatus, the image forming apparatus, the image processing method, the computer program, and the recording medium according to the present invention, the pixels related to each identifier are concentrated in a narrower area in each binary image. Therefore, in the rectangular area including the pixels related to each identifier extracted from the binary image at the time of compression, the ratio of the pixels related to each identifier becomes higher. That is, unnecessary data included in the compressed data can be reduced.
In addition, by integrating the identifiers, the number of binary images is minimized.

As a result, the amount of compressed image data obtained by compressing an image can be reduced, and the compression efficiency of the image can be improved. In addition, the processing time required for compressing the image can be shortened.
Furthermore, when transmitting / receiving compressed image data, the present invention has excellent effects such as reducing the amount of data to be transmitted / received and shortening the time required for transmission / reception.

2 is a block diagram illustrating an internal functional configuration of the image forming apparatus according to Embodiment 1. FIG. It is a schematic diagram which shows the outline | summary of the process which a compression process part compresses an image. It is a schematic diagram which shows the outline | summary of the process which a compression process part compresses an image. It is a block diagram which shows the internal structure of a compression process part. It is a block diagram which shows the internal structure of a foreground color indexing process part. It is a conceptual diagram which shows the example of the image which a compression process part should compress. It is a conceptual diagram which shows the example of an initial index image. It is a conceptual diagram which shows the example of a separation index image. It is a conceptual diagram which shows the example of an integrated index image. It is a chart which shows the example of a separation index table and a separation index color table. It is a chart which shows the example of a separation coordinate table. It is a chart which shows the example of an integrated index table and an integrated index color table. It is a chart which shows the example of an integrated coordinate table. It is a flowchart which shows the procedure of the process which an index production | generation part performs. It is a flowchart which shows the procedure of the process which an index production | generation part performs. It is a flowchart which shows the procedure of the process which an index production | generation part performs. It is a flowchart which shows the procedure of the process of the subroutine of a coordinate table update process. It is a chart showing an example of a color reduction processing table. It is a graph for demonstrating the procedure of the process with respect to a 1st attention pixel. It is a chart for demonstrating the procedure of the process with respect to a 2nd attention pixel. It is a flowchart which shows the procedure of the process of the subroutine of an index color table update process. It is a schematic diagram which shows the relationship between an attention pixel and the minimum distance of a subscanning direction. It is a flowchart which shows the procedure of the process which an index integration part performs. It is a flowchart which shows the procedure of the process which an index integration part performs. It is a flowchart which shows the procedure of the process of the subroutine of various table update processes. It is a flowchart which shows the procedure of the process of the subroutine of various table update processes. It is a schematic diagram which shows the relationship between an attention pixel and the minimum distance of the main scanning direction. FIG. 28 is a schematic diagram illustrating a relationship between a target pixel and a minimum distance in the sub-scanning direction in an image obtained by rotating the image illustrated in FIG. It is a schematic diagram which shows the relationship between an attention pixel and each minimum distance of a main scanning direction and a subscanning direction. It is a flowchart which shows a part of process sequence using the other conditions which determine the distance between pixels using the minimum distance of each in a main scanning direction and a subscanning direction. It is a schematic diagram which shows the range of the pixel from which an index is isolate | separated by the process of S121-S126 shown in FIG. It is a conceptual diagram which shows the example of the produced | generated binary image. 6 is a block diagram showing an internal functional configuration of an image forming apparatus according to Embodiment 2. FIG. 10 is a chart showing an example of an index color table used by a compression processing unit according to Embodiment 2. 6 is a block diagram illustrating an internal functional configuration of an image forming apparatus according to Embodiment 3. FIG. FIG. 10 is a block diagram illustrating an internal functional configuration of a scanner apparatus including an image processing apparatus according to a fourth embodiment. FIG. 10 is a block diagram illustrating an internal configuration of an image processing apparatus according to a fifth embodiment.

  Hereinafter, the present invention will be described in detail with reference to the drawings illustrating embodiments thereof.

Embodiment 1.
In the first embodiment, the image processing apparatus of the present invention is a part of an image forming apparatus that forms a color image.
FIG. 1 is a block diagram illustrating an internal functional configuration of the image forming apparatus according to the first embodiment.
An image forming apparatus according to an embodiment of the present invention includes a color image input device 11 that optically reads a color image. The color image input device 11 includes image data and a compressed image corresponding to the read color image. A color image processing apparatus 2 that performs processing for generating data is connected. The color image processing device 2 functions as an image processing device in the present invention.

The color image processing device 2 transmits to the outside the color image output device 13 that outputs a color image based on the image data generated by the color image processing device 2, and the compressed image data generated by the color image processing device 2. A device 14 is connected.
An operation panel 12 that receives an operation from a user is connected to the color image input device 11, the color image processing device 2, the color image output device 13, and the transmission device 14.

The color image input device 11 is constituted by a scanner having a CCD (Charge Coupled Device) optical sensor, and a reflected light image from a color image recorded on a record carrier such as paper is represented by R (red) G ( Green) B (blue) is decomposed and read by a CCD, converted into RGB analog signals, and output to the color image processing apparatus 2.
The color image processing apparatus 2 performs predetermined signal processing on the RGB analog signals input from the color image input apparatus 11 to generate digital image data. The color image processing apparatus 2 generates image data composed of digital C (cyan) M (magenta) Y (yellow) K (black) signals and outputs the image data to the color image output apparatus 13. Furthermore, the color image processing apparatus 2 generates compressed image data obtained by compressing the generated digital image data by a compression process described later, and outputs the compressed image data to the transmission apparatus 14.

The color image output device 13 outputs a color image by forming an image on a record carrier by a method such as thermal transfer, electrophotography, or ink jet based on the image data input from the color image processing device 2.
The transmission device 14 can be connected to a communication network such as a public network (not shown), a LAN (Local Area Network), or the Internet, and transmits compressed image data to the outside via the communication network by a communication method such as facsimile or e-mail. To do.
The operation panel 12 includes a display unit such as a liquid crystal display that displays information necessary for operation of the image forming apparatus, and a reception unit such as a touch panel or a numeric keypad that receives an instruction for controlling the operation of the image forming apparatus by a user operation. It is configured to include.

  The color image processing device 2 converts an analog signal input from the color image input device 11 into a digital signal by an A / D conversion unit 20, and a shading correction unit 21, an input tone correction unit 22, a region separation processing unit 23, The color correction unit 24, the black generation and under color removal unit 25, the spatial filter processing unit 26, the output gradation correction unit 27, and the gradation reproduction processing unit 28 are sent in this order, and image data composed of digital CMYK signals is sent to the color image output device. 13 to output to 13. The color image processing device 2 is configured to send a digital signal from the region separation processing unit 23 to the compression processing unit 3 and to output compressed image data to the transmission device 14.

The A / D conversion unit 20 receives an RGB analog signal input from the color image input device 11 to the color image processing device 2, converts the RGB analog signal into a digital RGB signal, and converts the RGB signal into a shading correction unit 21. Output to.
The shading correction unit 21 performs processing for removing various distortions generated in the illumination system, the imaging system, and the imaging system of the color image input device 11 on the RGB signal input from the A / D conversion unit 20.
Next, the shading correction unit 21 outputs the RGB signal from which distortion has been removed to the input tone correction unit 22.

The input tone correction unit 22 adjusts the color balance for the RGB signals input from the shading correction unit 21. The RGB signal input from the shading correction unit 21 to the input gradation correction unit 22 is an RGB reflectance signal, and the input gradation correction unit 22 converts the RGB signal input from the shading correction unit 21 into a color image processing apparatus. 2 is converted into a signal such as a density signal that can be easily processed.
Next, the input tone correction unit 22 outputs the processed RGB signal to the region separation processing unit 23.

  The region separation processing unit 23 separates each pixel in the image represented by the RGB signal input from the input tone correction unit 22 into one of a character region, a halftone dot region, or a photo region, and based on the separation result, A region identification signal indicating to which region each pixel belongs is output to the black generation and under color removal unit 25, the spatial filter processing unit 26, and the gradation reproduction processing unit 28, respectively. The region separation processing unit 23 outputs the RGB signals input from the input tone correction unit 22 to the color correction unit 24 and the compression processing unit 3, respectively.

  The compression processing unit 3 receives image data composed of RGB signals from the region separation processing unit 23 and executes a process of generating compressed image data using an image processing method (described later) according to the embodiment of the present invention. The compression processing unit 3 stores the generated compressed image data in a storage unit 29 configured using a nonvolatile storage medium such as a hard disk. Further, the compression processing unit 3 outputs the generated compressed image data to the transmission device 14.

The color correction unit 24 converts the RGB signal input from the region separation processing unit 23 into a CMY signal, and in order to realize faithful color reproduction, color turbidity based on spectral characteristics of CMY color materials including unnecessary absorption components Is removed from the CMY signal.
Next, the color correction unit 24 outputs the CMY signal subjected to color correction to the black generation and under color removal unit 25.

  The black generation lower color removal unit 25 performs black generation processing for generating a K signal from the CMY three-color signals input from the color correction unit 24, and subtracts the K signal obtained by the black generation processing from the original CMY signal. Thus, the CMY three-color signal is converted into a CMYK four-color signal. As an example of the black generation process, there is a method of generating black using skeleton black. In this method, if the input / output characteristics of the skeleton curve are y = f (x), the data before conversion is C, M, Y, and the UCR (Under Color Removal) rate is α (0 <α <1), The data C ′, M ′, Y ′, and K ′ are expressed by the following equations (1) to (4).

K ′ = f (min (C, M, Y)) (1)
C ′ = C−αK ′ (2)
M ′ = M−αK ′ (3)
Y ′ = Y−αK ′ (4)

Here, the UCR rate α (0 <α <1) indicates how much CMY is reduced by replacing the portion where CMY overlaps with K. Equation (1) indicates that the K signal is generated in accordance with the smallest signal strength among the signal strengths of CMY.
Next, the black generation and under color removal unit 25 outputs the CMYK signal obtained by converting the CMY signal to the spatial filter processing unit 26.

The spatial filter processing unit 26 performs spatial filter processing using a digital filter on the image represented by the CMYK signal input from the black generation and under color removal unit 25 based on the region identification signal input from the region separation processing unit 23. As a result, the blurring or graininess deterioration of the image is improved. For example, for a region separated into characters by the region separation processing unit 23, the spatial filter processing unit 26 uses a filter with a high enhancement amount of high-frequency components to improve character reproducibility. To do. In addition, for the region separated into halftone dots by the region separation processing unit 23, the spatial filter processing unit 26 performs low-pass filter processing for removing the input halftone component.
Next, the spatial filter processing unit 26 outputs the processed CMYK signal to the output tone correction unit 27.

  The output tone correction unit 27 performs output tone correction processing based on the characteristics of the color image output device 13 on the CMYK signal input from the spatial filter processing unit 26, and outputs the CMYK signal after the output tone correction processing. Output to the gradation reproduction processing unit 28.

The gradation reproduction processing unit 28 performs halftone processing corresponding to the region on the CMYK signal input from the output gradation correction unit 27 based on the region identification signal input from the region separation processing unit 23. For example, for a region separated into characters by the region separation processing unit 23, the gradation reproduction processing unit 28 performs binarization or multi-value conversion on a high-resolution screen suitable for reproducing high-frequency components. Perform processing. In addition, for the region separated into halftone dots by the region separation processing unit 23, the gradation reproduction processing unit 28 performs binarization or multi-value processing on the screen with an emphasis on gradation reproducibility. .
Next, the gradation reproduction processing unit 28 outputs the processed image data to the color image output device 13.

  The color image output device 13 forms a CMYK color image on a record carrier such as paper on the basis of image data composed of CMYK signals input from the color image processing device 2. By forming an image based on the image data, the image forming apparatus forms an image based on the received image.

Next, the configuration of the compression processing unit 3 and the processing performed by the compression processing unit 3 will be described.
2 and 3 are schematic diagrams illustrating an outline of processing in which the compression processing unit 3 compresses an image.
The uncompressed image shown in Fig. 2 consists of a red text "TEST image" written on a white background, a blue text "This is a test image" written in a solid green area, and various colors. Including a picture with.
As shown in FIG. 2, the compression processing unit 3 separates the input image into a foreground layer representing characters and a background layer representing others. The background layer includes an image in which characters are omitted. The foreground layer includes information indicating the color of each character.

Further, as shown in FIG. 3, the compression processing unit 3 generates a binary image for each color from the foreground layer. For example, as illustrated in FIG. 3, the compression processing unit 3 performs a binary image A in which the pixel value of the character “TEST image” is “1” and the other pixel values are “0”, and “This is a test image. A binary image B is generated in which the pixel value of the character “is” is “1” and the other pixel values are “0”. In the binary image A, color information indicating the color (red) of the text of the “TEST image” is added, and in the binary image B, the color (blue) of the text “This is a test image” is added. The indicated color information is added.
The compression processing unit 3 generates compressed image data by compressing each binary image to which the background layer and color information are added.

FIG. 4 is a block diagram showing an internal configuration of the compression processing unit 3.
The compression processing unit 3 generates a foreground mask indicating a character pixel in the foreground mask generation processing unit 31 based on the image data input from the region separation processing unit 23, and the character pixel of the character in the foreground indexing processing unit 32. A foreground layer in which colors are represented by indexes (identifiers) is generated. Further, the compression processing unit 3 generates a background layer by the background layer generation processing unit 33, generates a binary image of each color from the foreground layer by the binary image generation processing unit 34, and generates a background layer and a background layer by the image compression processing unit 35. A compressed image data is generated by compressing a binary image of each color, and the compressed image data is output.

Below, each part of the compression process part 3 is demonstrated in detail.
The foreground mask generation processing unit 31 receives the image data input from the region separation processing unit 23 to the compression processing unit 3, and extracts pixels corresponding to characters as foregrounds from the pixels constituting the image represented by the received image data. A foreground mask indicating the position of the extracted pixel is generated. Pixels other than the extracted pixels are background pixels. The foreground mask generation processing unit 31 functions as an accepting unit and an extracting unit in the present invention.

Specifically, the foreground mask generation processing unit 31 performs a differentiation process on the luminance value of the pixel in one direction on the image, detects an edge portion where the luminance changes brightly and an edge portion where the luminance changes darkly, The character image is binarized based on the detected edge part. More specifically, the foreground mask generation processing unit 31 scans the pixels on the image in one direction, and sets the range from the appearance of the positive differential value to the appearance of the negative differential value as a character pixel. Extraction is performed to generate a binary image in which the pixel value of the extracted pixel is “1” and the other pixel values are “0”.
The binary image generated as described above is a foreground mask. The foreground mask is obtained, for example, by omitting color information from the foreground layer shown in FIGS.

  The foreground mask generation processing unit 31 outputs the generated foreground mask data and image data to the foreground indexing processing unit 32. Note that the compression processing unit 3 receives a region identification signal from the region separation processing unit 23, and the foreground mask generation processing unit 31 is configured to perform a process of extracting pixels included in the character region indicated by the region identification signal. Good.

FIG. 5 is a block diagram showing an internal configuration of the foreground color indexing processing unit 32.
The foreground indexing processing unit 32 functions as foreground layer generation means in the present invention. For this purpose, the foreground color indexing processing unit 32 includes an index generation unit 321, an index integration unit 322, and an index correction unit 323.

Here, features of the embodiment of the present invention will be described.
FIG. 6 is a conceptual diagram illustrating an example of an image to be compressed by the compression processing unit 3.
The x coordinate and y coordinate shown in the figure correspond to the main scanning direction and the sub scanning direction when the color image input device 11 reads a color image. The coordinate (x, y) indicates the position of the pixel on the image, and each pixel can be specified by the coordinate.
Pixel groups P1 to P5 in FIG. 6 are each a pixel group indicating the foreground, and pixel groups other than the pixel groups P1 to P5 are pixel groups indicating the background.

Here, the pixel groups P1 to P3 are pixel groups of the same color (for example, red), the pixel group P4 is a pixel group of a completely different color (for example, blue), and the pixel group P5 is the color of the pixel groups P1 to P3. It is assumed that the pixel group is an approximate color (for example, light red).
In this case, for example, an index “1” is associated with each pixel of the pixel groups P1 to P3, an index “2” is associated with each pixel of the pixel group P4, and an index is associated with each pixel of the pixel group P5. “3” is associated (index correspondence processing).

If a rectangular area having a minimum area including pixels with the same index is cut out from an image indexed in this way (hereinafter referred to as an initial index image), the index “1” has coordinates (2, 2) to A rectangular area with coordinates (11, 18) is cut out.
Hereinafter, an index related to the initial index image may be referred to as an initial index. In the case of the image shown in FIG. 6, the number of initial indexes is three.
The rectangular area of the extracted coordinates (2, 2) to coordinates (11, 18) is an area that does not include the pixel with the index “1” (for example, a rectangular area of coordinates (2, 8) to coordinates (11, 14). ) Widely. In other words, the rectangular area includes a large number of areas unrelated to the index “1”. For this reason, the data amount of each rectangular area becomes large, and the compression efficiency deteriorates.

Therefore, in order to improve the compression efficiency, the indexes are separated in consideration of the distance between pixels associated with the same initial index.
The pixel group P1 and the pixel group P2 are close to each other, but the pixel groups P1 and P2 and the pixel group P3 are far from each other. Therefore, the index “1” is directly associated with each pixel of the pixel groups P1 and P2, and the new index “4” is associated with each pixel of the pixel group P3 again (index separation process).

If a rectangular area having the minimum area including the pixels associated with the same index is cut out from the indexed image (hereinafter referred to as a separated index image), the index “1” has coordinates (2, 2) to a rectangular area of coordinates (11, 7) are cut out, and for the index “4”, a rectangular area of coordinates (6, 15) to coordinates (9, 18) is cut out.
Hereinafter, an index related to a separation index image may be referred to as a separation index. In the case of the image shown in FIG. 6, the number of separation indexes is four, which is larger than the number of initial indexes.
After performing the index separation process, there is a possibility that the number of indexes will be very large (that is, the number of rectangular areas to be cut out will be very large). For this reason, the total amount of data in the rectangular area is increased, and the compression efficiency is deteriorated.

Therefore, in order to reduce the number of indexes, indexes that are associated with approximate colors among different indexes are integrated.
Accordingly, index integration is executed for the pixel groups P1 and P2 and the pixel group P5 (index integration processing).
Here, the number of pixels in the pixel groups P1 and P2 is compared with the number of pixels in the pixel group P5 in order to suppress deterioration in image quality due to the execution of the index integration process. This is because an image in which the smaller number of pixels is integrated into the larger number of pixels is less noticeable in the color change of each pixel than an image in which the larger number of pixels is integrated into the smaller number of pixels. It is.
Therefore, the pixel group P5 is integrated with the pixel groups P1 and P2, and the same index “1” as that of the pixel groups P1 and P2 is associated with each pixel of the pixel group P5 again.

When a rectangular area including pixels associated with the same index is cut out from an image indexed in this way (hereinafter referred to as an integrated index image), coordinates (2, 2) for the index “1” A rectangular area having coordinates (15, 11) is cut out.
Hereinafter, an index related to the integrated index image may be referred to as an integrated index. In the case of the image shown in FIG. 6, the number of integrated indexes is 3, which is smaller than the number of separation indexes.

In the present embodiment, in the index integration process, indexes associated with the same color are not integrated. This is because if the pixel groups P1, P2 and the pixel group P3 are integrated, the execution result of the index separation process is not reflected in the integrated index image.
If there is no approximate color, or if the number of indices cannot be reduced sufficiently by simply integrating the indexes associated with the approximate colors, the indexes associated with the same color are integrated. May be. In this case, it is desirable to integrate those having a relatively short distance between pixels.

As a result, the index “1” is associated with the pixel groups P1, P2, and P5, the index “4” is associated with the pixel group P3, and the index “2” is associated with the pixel group P4, and a rectangular area is cut out for each index. For this reason, the area of the rectangular area including the same index is minimized, and the number of indexes is minimized.
Then, such an integrated index image is output as a foreground layer.

  The index “3” becomes an empty index by the index integration process, but after the index integration process is executed, the index “4” is converted to the index “3”, and the index “4” is changed to an empty index. You can pack the index. Note that the index “3” may be left as an empty index and the indexes “1”, “2”, and “4” may be used.

7, 8, and 9 are conceptual diagrams illustrating examples of the initial index image, the separation index image, and the integrated index image.
In the present embodiment, the data of each of the separation index image and the integrated index image is generated, but the data of the initial index image is not generated. However, the initial index image is also illustrated for comparison with the separated index image.
The x coordinate shown in FIGS. 7 to 9 corresponds to the position of the pixel in the main scanning direction in the image represented by the image data input to the foreground color indexing processing unit 32, and the y coordinate is the pixel in the sub scanning direction. Corresponds to the position of. The main scanning direction and the sub scanning direction of each index image correspond to the main scanning direction and the sub scanning direction when the color image is read by the color image input device 11. The coordinate (x, y) indicates the position of the pixel on the image, and each pixel can be specified by the coordinate.

  In each index image, an index indicating each color (for example, red, blue, etc.) is associated with each pixel set as the foreground in the foreground mask. The index indicating the foreground has any numerical value of “1” or more. In each index image, an index indicating a color (for example, white) different from the foreground color is associated with each pixel that is set as the background in the foreground mask. The index indicating the background is “0”.

10 to 13 are diagrams illustrating examples of an index table, an index color table, and a coordinate table.
The index table is a table in which addresses and indexes are associated with each other.

  The index color table is a table in which addresses, color information represented by color intensities of RGB colors, and the number of pixels are associated with each other. The address of the index color table corresponds to the index of the index table. In other words, the index color table is a table that associates an index, color information associated with the index, and the number of pixels associated with the index. The index “0” indicating the background is associated with color information indicating white color and the number of pixels 243. In the following, the color information of each color is represented by RGB values (R color intensity, G color intensity, B color intensity). For example, white is represented as an RGB value (255, 255, 255).

  The coordinate table is a table in which addresses are associated with minimum x coordinate values, minimum y coordinate values, maximum x coordinate values, and maximum y coordinate values. The address of the coordinate table corresponds to the index of the index table. In other words, the coordinate table includes the minimum coordinate value and the maximum coordinate value in the main scanning direction and the minimum coordinate value and the maximum coordinate value in the sub-scanning direction of the rectangular area on the image where the index and the pixel associated with the index exist. It is a table in which coordinate values are associated.

The separation index table shown in FIG. 10A, the separation index color table shown in FIG. 10B, and the separation coordinate table shown in FIG. 11 correspond to the separation index image shown in FIG.
The integrated index table shown in FIG. 12A, the integrated index color table shown in FIG. 12B, and the integrated coordinate table shown in FIG. 13 correspond to the integrated index image shown in FIG.
The address of the integrated index table corresponds to the index of the separation index table. In other words, the integrated index table is a table in which the separation index and the integrated index are associated with each other.

In the separation index image shown in FIG. 8, the pixel of index “2” (hereinafter referred to as the first blue pixel group) included in the rectangular area of coordinates (7, 2) to coordinates (11, 7), and coordinates (10 , 16) to the pixel of index “6” (hereinafter referred to as the second blue pixel group) included in the rectangular area of coordinates (12, 16) is a pixel having an RGB value (0, 0, 255) ( Blue pixels).
Therefore, as shown in FIG. 7, in the initial index image, the index “2” is associated with both the first blue coordinate group and the second blue coordinate group.

However, as can be seen from the separation coordinate table shown in FIG. 11, the minimum distance in the sub-scanning direction between the first blue pixel group and the second blue pixel group is the minimum y coordinate of the second blue pixel group. The subtraction result “9” obtained by subtracting the maximum y-coordinate value “7” of the first blue pixel group from the value “16” is very far.
For this reason, in the separation index image shown in FIG. 8, the index related to the first blue coordinate group and the index related to the second blue coordinate group are separated, and different indexes are associated with each other.

In the separation index image shown in FIG. 8, the pixel of index “5” (hereinafter referred to as the first green pixel group) included in the rectangular area of coordinates (1, 14) to coordinates (3, 18) has an RGB value ( 120, 240, 120) pixels (light green pixels). Also, the pixel of index “7” (hereinafter referred to as the second green pixel group) included in the rectangular area of coordinates (4, 18) to coordinates (4, 19) has an RGB value (135, 240, 135). A pixel (a lighter green pixel).
The color of the first green pixel group and the color of the second green pixel group are very close. Moreover, the number of pixels “11” in the first green pixel group is larger than the number of pixels “2” in the second green pixel group.

Accordingly, the index “7” associated with the address “7” in the separation index table shown in FIG. 10A is changed to the index “5” in the integrated index table shown in FIG.
In the integrated index color table shown in FIG. 12B, the color information and the number of pixels associated with the address “7” are respectively changed to default values (“0” in the present embodiment). Further, the number of pixels associated with the address “5” is changed to “13”. The number of pixels “13” is the number of pixels “11” associated with the address “5” and the number of pixels “2” associated with the address “7” in the separation index color table shown in FIG. Is the addition result.

  Furthermore, in the integrated coordinate table shown in FIG. 13, various coordinate values associated with the address “7” are changed to default values. The default value of the minimum x coordinate value is “4095” which is a numerical value equal to or larger than the number of pixels in the main scanning direction of the image, and the default value of the minimum y coordinate value is a numerical value equal to or larger than the number of pixels in the sub scanning direction of the image. “8191”, the default value of the maximum x coordinate value is “0”, and the default value of the maximum y coordinate value is “0”. Here, a default value is set based on the number of pixels with an A3 size and a resolution of 300 dpi.

  Furthermore, in the integrated coordinate table, various coordinate values associated with the address “5” are the minimum x coordinate value “1” and the minimum y coordinate value of the rectangular area where the pixel associated with the index “5” is present. It is changed to “14”, the maximum x coordinate value “4”, and the maximum y coordinate value “19”. Here, as can be understood by referring to the separation coordinate table shown in FIG. 11, the minimum x coordinate value and the minimum y coordinate value of the index “5” in the integrated coordinate table shown in FIG. 13 are the index “5” in the separation coordinate table. , “7” equal to the minimum value of the minimum x coordinate value and the minimum y coordinate value of each, and the maximum x coordinate value and the maximum y coordinate value of the index “5” in the integrated coordinate table are the index “5” in the separated coordinate table. It is equal to the maximum value among the maximum x coordinate value and the maximum y coordinate value of “7” and “7”, respectively.

In the index generation unit 321 illustrated in FIG. 5, index correspondence processing and index separation processing are executed based on foreground layer data and image data. The index generation unit 321 functions as an identifier correspondence unit and an identifier separation unit in the present invention.
In the index correspondence processing, an index associated with a specific color is associated with each pixel according to the color of each pixel.
In the index separation process, among the pixels associated with the same index in the index correspondence process, the index is separated between pixels that are separated from the predetermined distance on the image. In the present embodiment, the minimum distance Δy shown in FIG. 22 described later is used as the predetermined distance.

  However, the index generation unit 321 is configured to generate indexed index image data by executing index correspondence processing and index separation processing for each pixel of image data for one sheet (FIGS. 14 to 16 described later). Refer to the flowchart shown in FIG. In other words, the index generation unit 321 executes index correspondence processing on one piece of image data to generate initial index image data, and then executes index separation processing on the initial index image data. It is not the structure which produces | generates the data of a separation index image.

Therefore, the index generation unit 321 outputs separation index image data, a separation index table, a separation index color table, and a separation coordinate table as shown in FIGS. 8, 10, and 11.
The index generation unit 321 also generates an initial index table for temporarily storing the initial index. The initial index stored in the initial index table corresponds to the address of the separation index table.

  The initial index table is not used in the subsequent stage of the index generation unit 321. In the flowcharts shown in FIGS. 14 to 16, the index generation unit 321 does not generate the separation index table in S11, and determines NO in S13, and then rewrites the data in the initial index table to rewrite the separation index table. A table may be generated. In this case, since only one of the initial index table and the separation index table needs to be stored in a memory (not shown), for example, hardware resources can be saved.

14 to 16 are flowcharts showing a procedure of processing performed by the index generation unit 321.
As illustrated in FIG. 14, the index generation unit 321 generates an initial index table, a separation index table, a separation index color table, a separation coordinate table, and a separation index image (S11).
By executing the process of S11, addresses “0”, “1”, “2”,... And a default value “0” of the initial index are stored in the initial index table. , Addresses “0”, “1”, “2”,... And the default value of the separation index are stored. The default value for the isolation index is equal to the address associated with the isolation index.

In addition, the separation index color table stores addresses “0”, “1”, “2”,..., A default value of RGB values, and a default value “0” of the number of pixels. The default RGB value associated with the address “0” is the RGB value (255, 255, 255), but the default RGB value associated with an address other than the address “0” is the RGB value (0, 0, 0). ).
In the separation coordinate table, addresses “0”, “1”, “2”,..., And default values of various coordinate values are stored.
Each pixel of the separation index image is given an index “0” (that is, a background index) as a default value.

After the processing of S11 ends, the index generation unit 321 resets the initial index counter idxcnt and the sub-scanning counter ycnt to “0” (S12). The initial index counter idxcnt is a counter for assigning an initial index in this process. The sub scanning counter ycnt is a counter indicating the pixel position in the sub scanning direction on the image of the target pixel.
Next, the index generation unit 321 determines whether or not the sub-scanning counter ycnt is smaller than the number of pixels in the sub-scanning direction of the image (S13). If the sub-scanning counter ycnt is equal to or greater than the number of pixels in the sub-scanning direction (NO in S13), the index generation unit 321 ends this processing because all input of image data for one sheet has been completed.

If the sub-scanning counter ycnt is smaller than the number of pixels in the sub-scanning direction (YES in S13), the index generating unit 321 resets the main scanning counter xcnt to “0” (S14). The main scanning counter xcnt is a counter indicating the pixel position in the main scanning direction on the image of the target pixel.
Next, the index generation unit 321 determines whether or not the main scanning counter xcnt is smaller than the number of pixels in the main scanning direction of the image (S15). When the main scanning counter xcnt is equal to or larger than the number of pixels in the main scanning direction (NO in S15), since the selection of all the pixels in the main scanning direction for one line has been completed, the index generation unit 321 increments the sub scanning counter ycnt. (S16), the process is returned to S13.

When the main scanning counter xcnt is smaller than the number of pixels in the main scanning direction (YES in S15), the index generation unit 321 executes the processing after S17. At this time, the position of the target pixel is expressed by the main scanning counter xcnt and the sub scanning counter ycnt. In the present embodiment, the pixel of interest (xcnt, ycnt) is the currently selected pixel. In addition, each pixel selected before the target pixel (xcnt, ycnt) is selected is a selected pixel.
The index generation unit 321 refers to the foreground mask and determines whether the pixel of interest (xcnt, ycnt) is a foreground pixel (S17).

When the target pixel (xcnt, ycnt) is a background pixel (NO in S17), the index generating unit 321 substitutes “0” for the address addr (S18), and updates the separation index color table for the address addr (S18). S19). In the process of S19, the index generation unit 321 increments the number of pixels (that is, the number of background pixels) associated with the address addr.
Next, the index generation unit 321 updates the separation coordinate table for the address addr by calling a subroutine for coordinate table update processing (see FIG. 17 described later) (S20).
After the process of S20 ends, the index generation unit 321 increments the main scanning counter xcnt (S21), and returns the process to S15.

FIG. 17 is a flowchart showing the procedure of the subroutine of the coordinate table update process.
In the following, the minimum x coordinate value, the minimum y coordinate value, the maximum x coordinate value, and the maximum y coordinate value associated with the address addr are represented as a minimum x coordinate value xmin [addr], a minimum y coordinate value ymin [addr], The maximum x coordinate value xmax [addr] and the maximum y coordinate value ymax [addr] are described.
The index generation unit 321 determines whether or not the minimum x coordinate value xmin [addr] recorded in the separation coordinate table is larger than the main scanning counter xcnt (S31).
When the minimum x coordinate value xmin [addr] is larger than the main scanning counter xcnt (YES in S31), the index generation unit 321 substitutes the main scanning counter xcnt for the minimum x coordinate value xmin [addr] associated with the address addr. (S32).

When S32 ends, or when the minimum x coordinate value xmin [addr] is equal to or smaller than the main scanning counter xcnt in S31 (NO in S31), the index generation unit 321 records the minimum y coordinate value ymin recorded in the separation coordinate table. It is determined whether [addr] is larger than the sub-scanning counter ycnt (S33).
When the minimum y coordinate value ymin [addr] is larger than the sub-scanning counter ycnt (YES in S33), the index generation unit 321 substitutes the sub-scanning counter ycnt for the minimum y-coordinate value ymin [addr] associated with the address addr. (S34).

When S34 ends, or when the minimum y coordinate value ymin [addr] is equal to or smaller than the sub-scanning counter ycnt in S33 (NO in S33), the index generation unit 321 determines the maximum x coordinate value xmax recorded in the separation coordinate table. It is determined whether [addr] is smaller than the main scanning counter xcnt (S35).
When the maximum x coordinate value xmax [addr] is smaller than the main scanning counter xcnt (YES in S35), the index generation unit 321 substitutes the main scanning counter xcnt for the maximum x coordinate value xmax [addr] associated with the address addr. (S36).

When S36 ends, or when the maximum x coordinate value xmax [addr] is greater than or equal to the main scanning counter xcnt in S35 (NO in S35), the index generation unit 321 determines the maximum y coordinate value ymax recorded in the separation coordinate table. It is determined whether [addr] is smaller than the sub-scanning counter ycnt (S37).
When the maximum y coordinate value ymax [addr] is smaller than the sub-scanning counter ycnt (YES in S37), the index generation unit 321 substitutes the sub-scanning counter ycnt for the maximum y-coordinate value ymax [addr] associated with the address addr. (S38).

  When S38 ends, or when the maximum y coordinate value ymax [addr] is equal to or greater than the sub-scanning counter ycnt in S37 (NO in S37), the index generation unit 321 ends the subroutine of the coordinate table update process, Return processing to the main routine.

As shown in FIG. 14, when the target pixel (xcnt, ycnt) is a foreground pixel (YES in S17), as shown in FIG. 15, the index generating unit 321 displays color information of the target pixel (xcnt, ycnt). Based on the above, a color reduction process is performed (S41).
As the color reduction processing in S41, for example, the method described in Patent Document 1 can be used. The color reduction process in this case is a process of performing color reduction to a predetermined number of colors for the color information of each pixel included in the image data. For example, the number of bits of color information representing the color intensity of each RGB color by 8 bits is reduced to the number of bits such as 2-2-2, 3-3-3, or 3-3-3 bits. The method for selecting the number of bits can be arbitrarily selected depending on the accuracy of the color determination.

FIG. 18 is a chart showing an example of the color reduction processing table.
In the present embodiment, when the number of bits of RGB color information is 2-2-2, color reduction processing is performed using the color reduction processing table shown in FIG. In this case, each RGB color has 4 gradations.
In the color reduction processing table, the R color intensity, the G color intensity, and the B color intensity are associated with the R color reduction result col_r, the G color reduction result col_g, and the B color reduction result col_b. Such a color reduction processing table is stored in advance in a memory (not shown).
In S41, the index generation unit 321 refers to the color reduction processing table and acquires the color reduction results col_r, col_g, and col_b based on the color information of the pixel of interest (xcnt, ycnt).

Next, the index generation unit 321 calculates an address coladr to the initial index table using the following equation (5) in order to associate the color reduction result with the initial index (S42). As a result, the address coladr takes any value within the range of “0” or more and “63” or less. That is, up to 64 initial indexes can be assigned to the color information.
coladr = col_r + col_g × 4 + col_b × 16 (5)

A specific example of the address coladr to the initial index table is given below.
The target pixel (xcnt, ycnt) = (2, 2) is an RGB value (128, 0, 128) (see the separation index image shown in FIG. 8 and the separation index color table shown in FIG. 10B). In this case, referring to the color reduction processing table shown in FIG. 18, since the color reduction results col_r, col_g, col_b = 2, 0, 2, the address coladr to the initial index table is calculated as in the following equation (6). Is done.
coladr = 2 + 0 × 4 + 2 × 16 = 34 (6)

  Hereinafter, the target pixel (xcnt, ycnt) = (2, 2) is referred to as a first target pixel (2, 2). The first target pixel (2, 2) is an example of a pixel whose index is not separated. For this reason, the separation index associated with the first target pixel (2, 2) is equal to the initial index.

The pixel of interest (xcnt, ycnt) = (10, 16) is an RGB value (0, 0, 255). In this case, since the color reduction results col_r, col_g, col_b = 0, 0, 3, the address coladr to the initial index table is calculated as in the following equation (7).
coladr = 0 + 0 × 4 + 3 × 16 = 48 (7)

  Hereinafter, the target pixel (xcnt, ycnt) = (10, 16) is referred to as a second target pixel (10, 16). The second pixel of interest (10, 16) is an example of a pixel from which an index is separated. For this reason, the separation index associated with the second pixel of interest (10, 16) is different from the initial index.

Next, the index generation unit 321 refers to the initial index table, reads the initial index tmpidx [coladr] at the address coladr, and determines whether or not the read initial index tmpidx [coladr] is “0” (S43). . The determination process of S43 is a process of determining whether the initial index tmpidx [coladr] is unregistered or registered.
When the initial index tmpidx [coladr] is “0” (YES in S43), the initial index tmpidx [coladr] is not registered. Therefore, the index generation unit 321 increments the initial index counter idxcnt (S44).

FIGS. 19 and 20 are tables for explaining the processing procedure for the first pixel of interest (2, 2) and the second pixel of interest (10, 16). 19A and 19B, (a) shows an initial index table, (b) shows a separation coordinate table, and (c) shows a separation index color table.
As shown in FIG. 19A, the first target pixel (2, 2) having the address coladr = 34 has the initial index tmpidx [coladr] = 0. The initial index counter idxcnt = 1. Therefore, the initial index “1” is newly assigned to the first target pixel (2, 2).

As shown in FIG. 15, the index generating unit 321 updates the initial index tmpidx [coladr] associated with the address coladr of the initial index table to the value of the initial index counter idxcnt (S45).
In the case of FIG. 19A, the index generation unit 321 updates the initial index “0” associated with the address “34” of the initial index table to “1”.

Next, as shown in FIG. 15, the index generation unit 321 updates the separated index image (S46). Specifically, the index generation unit 321 overwrites the pixel value of the target pixel (xcnt, ycnt) in the separated index image with the value of the initial index counter idxcnt.
Further, the index generating unit 321 updates the separation coordinate table with respect to the address addr by substituting the initial index tmpidx [coladr] for the address addr (S47) and calling a coordinate table update processing subroutine (see FIG. 17) ( S48).

  As shown in FIG. 19A, the initial index tmpidx [coladr] = 1 associated with the first pixel of interest (2, 2). Therefore, by executing the processing of S46, the pixel value of the pixel at the coordinate (2, 2) in the separation index image becomes “1” as shown in FIG. Further, by executing the processing of S48, the separation coordinate table is updated as shown in FIG. As a result, in the separation coordinate table, the minimum x coordinate value “2”, the minimum y coordinate value “2”, the maximum x coordinate value “2”, and the maximum y coordinate value “2” are stored in association with the address “1”. Is done.

Furthermore, the index generation unit 321 updates the separation index color table with respect to the address addr by calling a subroutine (see FIG. 21 described later) of the index color table update process (S49).
As shown in FIG. 19A, the initial index tmpidx [coladr] = 1 associated with the first pixel of interest (2, 2). For this reason, by executing the processing of S49, the separation index color table is updated as shown in FIG. Prior to the execution of the process of S49, the color information associated with the address “1” is an RGB value (0, 0, 0) and the number of pixels is “0”. In the table, the RGB value (128, 0, 128) of the first target pixel (2, 2) and the number of pixels “1” are stored in association with the address “1” (shown in FIG. 21 described later). (See S52 and S53).

FIG. 21 is a flowchart showing the procedure of the subroutine of the index color table update process.
The index generation unit 321 refers to the separation index color table and determines whether or not the number of pixels associated with the address addr is “0” (S51). The determination process of S51 is a process of determining whether the RGB value associated with the initial index “addr” in the initial index table is unregistered or registered.

  When the number of pixels is “0” (YES in S51), since the RGB value is not registered, the index generation unit 321 stores the RGB value of the pixel of interest (xcnt, ycnt) in the separation index color table (S52). ) The number of pixels is incremented (S53), the index color table update process subroutine is terminated, and the process returns to the main routine.

When executing the processing of S52, the color intensities rcol [addr], gcol [addr], and bcol [addr] of RGB colors to be associated with the initial index “addr” are the color intensities rcol [attention of the pixel of interest (xcnt, ycnt). Pixels], gcol [target pixel], and bcol [target pixel] are used to calculate the following formulas (8) to (10).
rcol [addr] <-rcol [target pixel] (8)
gcol [addr] <-gcol [target pixel] (9)
bcol [addr] <-bcol [target pixel] (10)

When the number of pixels is not “0” (NO in S51), since the RGB value has already been registered, the index generation unit 321 reads the RGB value associated with the address addr from the separation index color table, and reads the read RGB Based on the value and the RGB value of the target pixel (xcnt, ycnt), the average value of the color intensities of the RGB colors is calculated (S54).
Next, the index generation unit 321 stores the average value of the color intensities of the RGB colors calculated in S54 in the separation index color table in association with the address addr (S55), and moves the process to S53.

  In the case of NO in S51, the index generation unit 321 may execute the process of S53 without performing the processes of S54 and S55. This is because the color intensities of RGB with the same color reduction result are already stored in the separation index color table.

When executing the processing of S54, the color intensities rcol [addr], gcol [addr], bcol [addr] of the RGB colors to be associated with the initial index “addr” are already associated with the initial index “addr”. Using the color intensities rcol [addr], gcol [addr], and bcol [addr] of each color, the following equations (11) to (13) are calculated.
rcol [addr] ← {rcol [addr] + rcol [target pixel]} / 2 (11)
gcol [addr] ← {gcol [addr] + gcol [target pixel]} / 2 (12)
bcol [addr] ← {bcol [addr] + bcol [target pixel]} / 2 (13)

By the way, although the process of S54 is very simple, every time the average value is calculated, the color intensities rcol [addr], gcol [addr], and bcol [addr] may include an error.
In order to avoid such inconvenience, for example, in the process of S54, the index generation unit 321 uses the following formulas (14) to (16) instead of the above formulas (11) to (13). The color intensities rcol [addr], gcol [addr], bcol [addr] are calculated, and the calculated color intensities rcol [addr], gcol [addr], bcol [addr] are stored in the separation index color table in the process of S55. It is possible to memorize it.

rcol [addr] ← rcol [addr] + rcol [target pixel] (14)
gcol [addr] ← gcol [addr] + gcol [target pixel] (15)
bcol [addr] ← bcol [addr] + bcol [target pixel] (16)
Such equations (14) to (16) are obtained by simply adding the color intensities of the target pixel (xcnt, ycnt).

  Then, the index generation unit 321 executes a process of updating the separation index color table when NO in S13, that is, when all input of image data for one sheet is completed. At this time, the index generation unit 321 sequentially generates the variables tempad = 1, 2,..., And uses the number of pixels cnt [tempad] associated with the address tempad in the separation index color table to obtain the final color intensity rcol [ tempad], gcol [tempad], bcol [tempad] are calculated as in the following equations (17) to (19). Finally, the index generation unit 321 stores the calculated color intensities rcol [tempad], gcol [tempad], and bcol [tempad] in the separation index color table.

rcol [tempad] <-rcol [tempad] / cnt [tempad] (17)
gcol [tempad] <-gcol [tempad] / cnt [tempad] (18)
bcol [tempad] <-bcol [tempad] / cnt [tempad] (19)

  When the process of S49 shown in FIG. 15 ends, the process for the pixel of interest (xcnt, ycnt) ends. Therefore, the index generation unit 321 returns the process to S21 illustrated in FIG. After that, the index generation unit 321 increments the main scanning counter xcnt in S21, and then executes the processing after S15 for the pixel of interest (xcnt, ycnt).

As shown in FIG. 15, when the initial index tmpidx [coladr] is not “0” (NO in S43), the initial index tmpidx [coladr] is already registered. Therefore, as shown in FIG. 16, the index generation unit 321 substitutes the initial index tmpidx [coladr] for the address addr (S61), and refers to the separation coordinate table to determine the maximum y coordinate associated with the address addr. The value ymax [addr] is read (S62).
Next, the index generation unit 321 substitutes the result obtained by subtracting the maximum y coordinate value ymax [addr] from the sub-scanning counter ycnt for the minimum distance Δy (S63), and determines whether the minimum distance Δy is greater than a predetermined threshold lm_y. (S64).
Here, in this embodiment, the predetermined threshold value lm_y = 3.

FIG. 22 is a schematic diagram showing the relationship between the target pixel (xcnt, ycnt) and the minimum distance Δy in the sub-scanning direction.
The position of the target pixel (xcnt, ycnt) is designated by coordinates (xcnt, ycnt) on the image. The index of the pixel of interest (xcnt, ycnt) is the initial index tmpidx [coladr], as can be seen by referring to the initial index table.
The area indicated by hatching in FIG. 22 is an area occupied by pixels associated with the same index (hereinafter referred to as the same index) as the index of the target pixel (xcnt, ycnt). The minimum and maximum coordinate values of the pixels included in this area are the minimum x coordinate value xmin [addr], the minimum y coordinate value ymin [addr], the maximum x coordinate value xmax [addr], and the maximum y coordinate value ymax [ addr]. However, the description of “[addr]” is omitted in FIG.

  In the present embodiment, the difference between the y coordinate values of the pixel having the maximum y coordinate and the target pixel (xcnt, ycnt) among the pixels included in this region is defined as the minimum distance Δy in the sub-scanning direction. . Since the target pixel (xcnt, ycnt) is selected in order from the position of the coordinates (0, 0), the minimum distance Δy is associated with the same index among the target pixel (xcnt, ycnt) and the selected pixel. It is the minimum distance between the pixels.

  As shown in FIG. 20A, the second target pixel (10, 16) with coladr = 48 has an initial index tmpidx [coladr] = 2. Therefore, “2” is assigned to the address addr in the process of S61, and the separated coordinate table shown in FIG. 20A is referred to in the process of S62, and the maximum y coordinate value ymax [addr] = 7 is read. It is. Further, the minimum distance Δy = 16−7 = 9 is calculated in the process of S63, and it is determined in S64 that the minimum distance Δy is greater than the predetermined threshold value lm_y.

That is, the determination process of S64 is a process of determining whether or not the target pixel (xcnt, ycnt) is separated from the other pixel associated with the same index by a predetermined distance.
If the minimum distance Δy> the predetermined threshold lm_y (YES in S64), the index generating unit 321 moves the process to S44 shown in FIG.

As shown in FIGS. 20B and 20C, the value of the initial index counter idxcnt immediately before processing the second pixel of interest (10, 16) is “5”. The value of the counter idxcnt is “6”.
Therefore, as shown in FIG. 20A, the second target pixel (10, 16) with coladr = 48 is newly assigned with the separation index “6” instead of the initial index “2”.
By executing the process of S45, the index generation unit 321 updates the initial index “2” associated with the address “48” of the initial index table to “6”.

Therefore, by executing the processing of S46, the pixel value of the pixel at the coordinates (10, 16) in the separation index image becomes “6” as shown in FIG. The pixel value “2” is not given to the pixel at the coordinates (10, 16).
Furthermore, by executing the processing of S48, the separation coordinate table is updated as shown in FIG. As a result, in the separation coordinate table, the minimum x coordinate value “10”, the minimum y coordinate value “16”, the maximum x coordinate value “10”, and the maximum y coordinate value “16” are stored in association with the address “6”. Is done.

Furthermore, by executing the process of S49, the separation index color table is updated as shown in FIG. Prior to the execution of the process of S49, the color information associated with the address “6” is an RGB value (0, 0, 0) and the number of pixels is “0”. In the table, the RGB value (0, 0, 255) and the number of pixels “1” are stored in association with the address “6”.
In the separation coordinate table and the separation index color table, various numerical values associated with the address “2” are not changed.

  After the index separation is performed as described above, the initial index “2” is not associated with the target pixel (xcnt, ycnt). This is because the minimum distance Δy between the target pixel (xcnt, ycnt) selected after the second target pixel (10, 16) and the pixel associated with the index “2” is always the predetermined threshold lm_y. It is because it is larger than.

As shown in FIG. 16, when the minimum distance Δy ≦ the predetermined threshold lm_y (NO in S64), the index generation unit 321 updates the separated index image (S65). Specifically, the index generation unit 321 overwrites the initial index tmpidx [coladr] on the pixel value of the target pixel (xcnt, ycnt) in the separated index image.
After the process of S65 is completed, the index generating unit 321 moves the process to S48 shown in FIG.

  By the way, the separation index stored in the separation index color table may be a default value. This is because the addresses of the separation index color table and the separation coordinate table are equal to the separation index when NO in S13, that is, when all the input of image data for one sheet is completed.

  In the index integration unit 322 illustrated in FIG. 5, index integration processing is executed based on the data of the separation index image, the separation index table, the separation index color table, and the separation coordinate table. The index integration unit 322 functions as an identifier integration unit in the present invention.

In the index integration process, among the pixels associated with different indexes by executing the index association process and the index separation process, indexes whose colors are approximated within a predetermined range are integrated.
The index integration unit 322 outputs the integrated index table, the integrated index color table, and the integrated coordinate table shown in FIGS. 12 and 13.

  For the first color associated with one separation index and the second color associated with another separation index, refer to the separation index table and separation index color table as shown in FIGS. Easy to understand. Also, the number of pixels associated with one separation index (hereinafter referred to as the first pixel number) and the number of pixels associated with other separation indexes (hereinafter referred to as the second pixel number). Is easily understood by referring to the separation index table and the separation index color table.

When the first color and the second color are close to each other, index integration is performed. At this time, in order to suppress degradation of image quality due to the integration of the indexes, the first pixel number and the second pixel number are compared so that the index with the smaller number of pixels is integrated into the index with the larger number. To do.
If the first color and the second color are not approximate, the index is not integrated.

FIG. 23 and FIG. 24 are flowcharts showing a procedure of processing performed by the index integration unit 322.
The index integration unit 322 sets “1” to the counter i (S71). The counter i indicates the first index “i”.
Next, the index integration unit 322 determines whether the counter i is equal to or less than the initial index counter idxcnt (S72).
The value of the initial index counter idxcnt is the number of separation indexes stored in the separation index table. Therefore, if i> idxcnt (NO in S72), since the processing for all the separated indexes is completed, the index integration unit 322 ends this processing.

If i ≦ idxcnt (YES in S72), the index integration unit 322 calculates the counter j using the following equation (20) (S73). The counter j indicates the first index “i” and the second index “j” for comparing the color information.
j = i + 1 (20)

Next, the index integration unit 322 determines whether the counter j is equal to or less than the initial index counter idxcnt (S74).
If j> idxcnt (NO in S74), since the comparison of the color information between the first index “i” and all the separation indexes is completed, the index integration unit 322 increments the counter i (S75). The process returns to S72.

  When j ≦ idxcnt (YES in S74), the index integration unit 322 refers to the separation index table and the separation index color table, and determines the color information of the first index “i” and the second index “j”. Color information is read out (S76). Specifically, the index integration unit 322 refers to the separation index color table, reads the color intensities rcol [i], gcol [i], bcol [i] of each color of RGB using the counter i as an address, and addresses the counter j as an address. As a result, the color intensities rcol [j], gcol [j], and bcol [j] of each RGB color are read out.

Next, the index integration unit 322 determines whether the color information of the first index “i” and the color information of the second index “j” are approximated within a predetermined range.
For this purpose, the index integration unit 322 calculates the calculation result Δcol using the following equation (21) (S77). The calculation result Δcol is the addition result of the absolute values of the differences in color intensity of each of the RGB colors of the first index “i” and the second index “j”.

Δcol = | rcol [i] −rcol [j] |
+ | Gcol [i] -gcol [j] |
+ | Bcol [i] -bcol [j] | ... (21)

When the calculation result Δcol is within the predetermined range, the index integration unit 322 determines that the color information of the first index “i” and the color information of the second index “j” are approximated within the predetermined range. To do. When the calculation result Δcol is “0”, the index integration unit 322 determines that the color information of the first index “i” and the color information of the second index “j” are the same color.
Therefore, as shown in FIG. 24, the index integration unit 322 determines whether or not the calculation result Δcol in S77 is equal to or less than a threshold th_col (S78). In the present embodiment, the threshold th_col = 40.

  Specific examples are described below. In the separation index color table shown in FIG. 10B, when the color information is read with the counter “5” as an address, the color intensity rcol [5] is “120”, gcol [5] is “240”, and bcol [5 ] Is “120”. When the color information is read using the counter “7” as an address, the color intensity rcol [7] is “135”, gcol [7] is “240”, and bcol [7] is “135”. Accordingly, the calculation result Δcol is calculated as shown in Expression (22).

Δcol = | rcol [5] −rcol [7] |
+ | Gcol [5] -gcol [7] |
+ | Bcol [5] -bcol [7] |
= | 120-135 | + | 240-240 | + | 120-135 |
= 30 ... (22)
Since the calculation result Δcol is “30”, that is, the threshold th_col = 40 or less and not “0”, the index integration unit 322 has the color information of the first index “5” and the second index “7”. It is determined that the color information approximates within a predetermined range and is not the same color.

  In the separation index color table shown in FIG. 10B, when the color information is read with the counter “2” as an address, the color intensity rcol [2] is “0”, gcol [2] is “0”, bcol. [2] is “255”. When color information is read using the counter “6” as an address, the color intensity rcol [6] is “0”, gcol [6] is “0”, and bcol [6] is “255”. Therefore, the calculation result Δcol is calculated as shown in Equation (23).

Δcol = | rcol [2] −rcol [6] |
+ | Gcol [2] -gcol [6] |
+ | Bcol [2] -bcol [6] |
= | 0-0 | + | 0-0 | + | 255-255 |
= 0 (23)
Since the calculation result Δcol is “0”, the index integration unit 322 determines that the color information of the first index “2” and the color information of the second index “6” are the same color.

If Δcol ≦ th_col or less (YES in S78), the index integration unit 322 determines whether the calculation result Δcol in S77 is “0” (S79).
When Δcol ≠ 0 (NO in S79), the index integration unit 322 determines the integration of the first index “i” and the second index “j”, and then the first index “i”. Are integrated into the second index “j” or the second index “j” is integrated into the first index “i”.

For this purpose, the index integration unit 322 refers to the separation index table and the separation index color table, and the number of pixels cnt [i] associated with the first index “i” and the second index “j”. Read out the number of pixels cnt [j] associated with (S80). Specifically, the index integration unit 322 refers to the separation index color table, reads the number of pixels cnt [i] using the counter i as an address, and reads the number of pixels cnt [j] using the counter j as an address.
Furthermore, the index integration unit 322 determines that the number of pixels cnt [i] associated with the first index “i” is greater than or equal to the number of pixels cnt [j] associated with the second index “j”. It is determined whether or not there is (S81).

  Specific examples are described below. In the separation index color table shown in FIG. 10B, the number of pixels cnt [5] read using the counter “5” as an address is “11”, and the number of pixels cnt [5] read using the counter “7” as an address. Is “2”. Therefore, since cnt [5] ≧ cnt [7], the index integration unit 322 determines to integrate the second index “7” with the first index “5”.

When cnt [i] ≧ cnt [j] (YES in S81), the index integration unit 322 counts the counters k1 and k2 in order to integrate the second index “j” with the first index “i”. Substituting i and j (S82) and calling various table update processing subroutines (see FIGS. 25 and 26 described later) (S83), the indexes are integrated.
When cnt [i] <cnt [j] (NO in S81), the index integration unit 322 counts the counters k1 and k2 in order to integrate the first index “i” with the second index “j”. j, i are substituted (S84), and the process proceeds to S83.

After the process of S83 is completed, the index integration unit 322 executes the process of S85 described later.
If Δcol> th_col (NO in S78), the index integration unit 322 does not integrate the first index “i” and the second index “j”, and executes the process of S85. Furthermore, when Δcol = 0 (YES in S79), the index integration unit 322 performs the process of S85 without integrating the first index “i” and the second index “j”.
The index integration unit 322 increments the counter j (S85), and returns the process to S74 shown in FIG.

25 and 26 are flowcharts showing the procedure of the subroutine of the various table update processing.
As shown in FIG. 25, the index integration unit 322 substitutes the index idx [k1] read with the counter k1 as an address into the index idx [k2] with the counter k2 as an address in the separation index table (S101). The separation index table is updated by the process of S101.

Next, the index integration unit 322 reads out the pixel numbers cnt [k1] and cnt [k2] using the counters k1 and k2 as addresses in the separation index color table (S102), and the read pixel number cnt [k1]. The addition result obtained by adding the number of pixels cnt [k2] is substituted into the number of pixels cnt [k1] associated with the counter k1 (S103), and the number of pixels cnt [k2] associated with the counter k2 is further set to a default value. Reset to “0” (S104).
Next, the index integration unit 322 resets the color intensities rcol [k2], gcol [k2], and bcol [k2] of each of the RGB colors associated with the counter k2 to the default value “0” in the separation index color table ( S105). The separation index color table is updated by the processing of S102 to S105.

  In the following, in the separation coordinate table, the minimum x coordinate value, the minimum y coordinate value, the maximum x coordinate value, and the maximum y coordinate value associated with the address “k1” are represented by the minimum x coordinate value xmin [k1] and the minimum y coordinate value. The coordinate value ymin [k1], the maximum x coordinate value xmax [k1], and the maximum y coordinate value ymax [k1] are described. Similarly, the minimum x coordinate value, the minimum y coordinate value, the maximum x coordinate value, and the maximum y coordinate value associated with the address “k2” are set as the minimum x coordinate value xmin [k2] and the minimum y coordinate value ymin [k2]. ], The maximum x coordinate value xmax [k2], and the maximum y coordinate value ymax [k2].

After the process of S105, as shown in FIG. 26, the index integration unit 322 determines whether or not the minimum x coordinate value xmin [k1] recorded in the separation coordinate table is larger than the minimum x coordinate value xmin [k2]. (S106).
When the minimum x coordinate value xmin [k1] is larger than the minimum x coordinate value xmin [k2] (YES in S106), the index integration unit 322 minimizes the minimum x coordinate value xmin [k1] associated with the address “k1”. The x coordinate value xmin [k2] is substituted (S107).

When S107 ends, or when the minimum x coordinate value xmin [k1] is equal to or smaller than the minimum x coordinate value xmin [k2] in S106 (NO in S106), the index integration unit 322 records the minimum recorded in the separation coordinate table. It is determined whether the y-coordinate value ymin [k1] is larger than the minimum y-coordinate value ymin [k2] (S108).
When the minimum y coordinate value ymin [k1] is larger than the minimum y coordinate value ymin [k2] (YES in S108), the index integration unit 322 minimizes the minimum y coordinate value ymin [k1] associated with the address “k1”. The y coordinate value ymin [k2] is substituted (S109).

When S109 ends, or when the minimum y coordinate value ymin [k1] is equal to or smaller than the minimum y coordinate value ymin [k2] in S108 (NO in S108), the index integration unit 322 records the maximum recorded in the separation coordinate table. It is determined whether or not the x coordinate value xmax [k1] is smaller than the maximum x coordinate value xmax [k2] (S110).
When the maximum x-coordinate value xmax [k1] is smaller than the maximum x-coordinate value xmax [k2] (YES in S110), the index integration unit 322 maximizes the maximum x-coordinate value xmax [k1] associated with the address “k1”. The x coordinate value xmax [k2] is substituted (S111).

When S111 ends, or when the maximum x-coordinate value xmax [k1] is greater than or equal to the maximum x-coordinate value xmax [k2] in S110 (NO in S110), the index integration unit 322 records the maximum recorded in the separation coordinate table. It is determined whether the y coordinate value ymax [k1] is smaller than the maximum y coordinate value ymax [k2] (S112).
When the maximum y coordinate value ymax [k1] is smaller than the maximum y coordinate value ymax [k2] (YES in S112), the index integration unit 322 sets the maximum y coordinate value ymax [k1] associated with the address “k1” to the maximum. The y coordinate value ymax [k2] is substituted (S113).

When S113 ends, or when the maximum y-coordinate value ymax [k1] is greater than or equal to the maximum y-coordinate value ymax [k2] in S112 (NO in S112), the index integration unit 322 determines the minimum x associated with the counter k2. The coordinate value xmin [k2], the minimum y coordinate value ymin [k2], the maximum x coordinate value xmax [k2], and the maximum y coordinate value ymax [k2] are reset to default values (S114). The separation coordinate table is updated by the processing of S106 to S114.
After the process of S114 ends, the index integration unit 322 ends the subroutine of the coordinate table update process and returns the process to the main routine.

Specific examples are described below. When the second index “7” is integrated with the first index “5”, the process of S101 is executed, so that it is associated with the address “7” in the separation index table shown in FIG. The index “7” is overwritten with the index “5” read with the counter “5” as an address.
Therefore, when the integrated index table shown in FIG. 12A is referenced using the counter “7” as an address, the read index is “5”, unlike the case where the separated index table is referenced.
On the other hand, when the integrated index table is referenced using the counter “5” as an address, the read index is “5”, as in the case of referring to the separated index table.

In addition, by executing the processing of S103 and S104, the number of pixels “11” associated with the address “5” in the separation index color table shown in FIG. 10B is read using the counter “7” as the address. Overwritten with the addition result “13” with the number of pixels “2”. After the overwriting related to the address “5” is completed, the pixel number “2” associated with the counter “7” is overwritten with the default value “0”.
Therefore, when the integrated index color table shown in FIG. 12B is referenced using the counters “5” and “7” as addresses, the number of pixels to be read is “13”, unlike the case of referring to the separation index table. “0”.

Further, by executing the process of S105, the RGB values (135, 240, 135) associated with the counter “7” in the separation index color table shown in FIG. , 0, 0).
Therefore, when the integrated index table shown in FIG. 12B is referred to with the counter “7” as an address, the read RGB values are RGB values (0, 0, 0), unlike the case of referring to the separation index table. It is.
On the other hand, when the integrated index table is referenced using the counter “5” as an address, the read RGB values are RGB values (120, 240, 120), as in the case of referring to the separation index table.

  Furthermore, in the separation coordinate table shown in FIG. 11, the maximum x coordinate value xmax [5] and the maximum y coordinate value ymax [5] associated with the address “5” are set to the address “7” by the processing of S106 to S114. The minimum x-coordinate value xmin [7], the minimum y-coordinate value ymin [7] associated with the address “7” is replaced with the associated maximum x-coordinate value xmax [7] and maximum y-coordinate value ymax [7]. The maximum x coordinate value xmax [7] and the maximum y coordinate value ymax [7] are reset to default values, respectively.

Therefore, when the integrated coordinate table shown in FIG. 13 is referenced using the counter “5” as an address, unlike the case of referring to the separated coordinate table, the minimum x-coordinate value xmin [5] and the minimum y-coordinate value ymin [ 5], the maximum x-coordinate value xmax [5], and the maximum y-coordinate value ymax [5] are “1”, “14”, “4”, and “19”.
Similarly, when referring to the integrated coordinate table using the counter “7” as an address, unlike the case of referring to the separated coordinate table, the minimum x-coordinate value xmin [7], the minimum y-coordinate value ymin [7], The maximum x coordinate value xmax [7] and the maximum y coordinate value ymax [7] are “4095”, “8191”, “0”, and “0”.

  As shown in FIG. 25, in the various table update processes in the present embodiment, the color intensities rcol [k1], gcol [k1], and bcol [k1] of each of the RGB colors associated with the counter k1 are the original values. Maintained. However, the index integration unit 322 may recalculate and update the color intensities rcol [k1], gcol [k1], and bcol [k1] before executing the processes of S103 to S105. In this case, the color intensities rcol [k1], gcol [k1], and bcol [k1] are calculated using the following equations (24) to (27).

cnt12 = cnt [k1] + cnt [k2] (24)
rcol [k1] = {rcol [k1] × cnt [k1] + rcol [k2] × cnt [k2]} / cnt12 (25)
gcol [k1] = {gcol [k1] × cnt [k1] + gcol [k2] × cnt [k2]} / cnt12 (26)
bcol [k1] = {bcol [k1] × cnt [k1] + bcol [k2] × cnt [k2]} / cnt12 (27)

When the new color intensities rcol [k1], gcol [k1], and bcol [k1] obtained using the equations (24) to (27) as described above are obtained, the obtained new color intensities rcol [k1] ], Gcol [k1], bcol [k1] include the original color intensities rcol [k1], gcol [k1], bcol [k1] and the original color intensities rcol [k2], gcol [k2], bcol [k2]. ] Are reflected according to the number of original pixels cnt [k1] and cnt [k2].
For this reason, the index integration unit 322 is limited to a configuration that integrates an index with a small number of pixels into an index with a large number of pixels (that is, a configuration that executes S81 shown in FIG. 24 and one of S82 and S84). Instead, the configuration may be such that the index of the address “j” is integrated into the index of the address “i” (that is, S82 and S84 are not executed and S82 is always executed).

  By executing the processing as described above, in the case of NO in S72 shown in FIG. 23, that is, when the processing for all the separation indexes is completed, the separation index table shown in FIG. 10 (a) is shown in FIG. 10 (b). The separated index color table shown in FIG. 11 and the separated coordinate table shown in FIG. 11 are the integrated index table shown in FIG. 12A, the integrated index color table shown in FIG. 12B, and the integrated coordinate table shown in FIG. Yes.

  After index integration, if the number of indexes stored in the integrated index table is greater than a predetermined upper limit number of indexes, the integrated index table, the integrated index color table, and the integrated coordinate table are separated as index tables. The index integration unit 322 again shows the index index unit 322 as shown in FIGS. 23 and 24 by considering the separation index color table and the separation coordinate table and appropriately increasing the threshold th_col (that is, relaxing the integration condition). What is necessary is just to perform a process. At this time, it is not necessary to return various data to the previous process (the process executed by the index generation unit 321).

  However, when the threshold th_col is excessively increased, the image quality may be significantly degraded. This is because not only the light red pixel changes to a red pixel, but also a significant color change may occur such that a purple pixel changes to a red pixel. In order to eliminate such an inconvenience, it is conceivable to relax the separation conditions instead of relaxing the integration conditions. However, in this case as well, there is no need to return various data to the previous process (the process executed by the index generation unit 321).

  First, the index integration unit 322 appropriately increases the minimum distance Δy (that is, relaxes the separation condition). Next, the index integration unit 322 refers to the integrated index color table and extracts an index of the same color. Furthermore, the index integration unit 322 refers to the integrated coordinate table to obtain the distance between the pixels associated with each index, and if the obtained distance is equal to or less than the minimum distance Δy, the indexes of the same color are integrated. The integrated index table, the integrated index color table, and the integrated coordinate table are corrected according to the integration result.

In the index correction unit 323 illustrated in FIG. 5, index correction processing is performed based on the data of the separated index image, the integrated index table, the integrated index color table, and the integrated coordinate table.
In the index correction process, data of an integrated index image (ie, foreground layer) is generated.

Specifically, the index correction unit 323 performs processing similar to, for example, S12 to S16 illustrated in FIG. 14 and sequentially selects a target pixel (xcnt, ycnt) in the separated index image. Next, the index correction unit 323 reads the pixel value (that is, the separation index) of the target pixel (xcnt, ycnt) based on the data of the separation index image. Furthermore, the index correction unit 323 refers to the integrated index table and reads the integrated index using the read separation index as an address. Finally, the index correction unit 323 overwrites the read integrated index on the pixel value of the target pixel (xcnt, ycnt).
As a result of the above processing, the separation index image shown in FIG. 8 becomes the integrated index image shown in FIG.

  In addition, the structure which the index correction | amendment part 323 performs may be performed when the index integration part 322 is NO in S72 shown in FIG. 23 (namely, when the process with respect to all the separation indexes is completed). In this case, it is not necessary to provide the index correction unit 323 in the foreground color indexing processing unit 32.

In addition, the foreground color indexing processing unit 32 may include an index corresponding unit that executes an index correspondence process and an index separation unit that executes an index separation process instead of the index generation unit 321.
However, if configured in this way, it is necessary to store both the relationship between the initial index and the separation index and the relationship between the separation index and the integrated index. Therefore, the foreground color indexing processing unit 32 must have a memory with a large storage capacity. Moreover, the process which manages the relationship between indexes is complicated.

  On the other hand, in the configuration of the present embodiment, index separation processing is continuously executed immediately after the index association processing in the index generation unit 321. In other words, the separation index can be associated with each pixel before the initial index is associated with all the pixels. For this reason, the foreground color indexing processing unit 32 may include a memory having a storage capacity capable of storing the relationship between the separation index and the integrated index. In addition, the process of managing the association between indexes is simple.

By the way, in the present embodiment, the foreground color indexing processing unit 32 executes the index association for each pixel, the index separation based on the distance between the pixels, and the integration of the indexes of the approximate colors in this order. To do.
If the foreground color indexing processing unit 32 executes the index association for each pixel, the integration of the indexes of the approximate colors, and the index separation based on the distance between the pixels in this order, the following is performed. There is a risk of inconvenience.

In order to prevent the deterioration of the image quality, it is necessary to integrate the colors that are closest to each other. For this reason, it is necessary to once associate an initial index with respect to all pixels. That is, the index association process and the integration process cannot be executed continuously.
In addition, since the separation process is performed after the integration process, the number of indexes may increase excessively after the separation. In this case, since it is necessary to reduce the number of indexes by performing integration processing again (that is, returning various data to the previous processing), management of various data becomes complicated.

FIG. 27 is a schematic diagram illustrating the relationship between the pixel of interest (xcnt, ycnt) and the minimum distance Δx in the main scanning direction. FIG. 28 is a schematic diagram showing the relationship between the pixel of interest (xcnt, ycnt) and the minimum distance Δx in the sub-scanning direction in an image obtained by rotating the image shown in FIG. 27 rightward by 90 °.
FIG. 29 is a schematic diagram showing the relationship between the target pixel (xcnt, ycnt), the minimum distance Δx in the main scanning direction, and the minimum distance Δy in the sub-scanning direction.
The images shown in FIGS. 27 to 29 correspond to the images shown in FIG.

  The index generation unit 321 may be configured to use the minimum distance Δx illustrated in FIG. 27 instead of using the minimum distance Δy illustrated in FIG. 22 when executing the processes illustrated in FIGS. 14 to 16. At this time, the index generation unit 321 reads the maximum x-coordinate value xmax [addr] associated with the address addr with reference to the separation coordinate table in the process of S62 shown in FIG. The result of subtracting the maximum x coordinate value xmax [addr] from the scan counter xcnt is substituted into the minimum distance Δx, and it is determined in step S64 whether the minimum distance Δx is greater than a predetermined threshold lm_x.

  When the minimum distance Δx> the predetermined threshold value lm_x (that is, YES in S64), the index generation unit 321 moves the process to S44 to separate the indexes. When the minimum distance Δx ≦ the predetermined threshold lm_x (that is, NO in S64), the index generation unit 321 moves the process to S65 without separating the indexes.

In addition, when executing the processes shown in FIGS. 14 to 16, the index generation unit 321 uses the minimum distance Δx and the minimum distance Δy shown in FIG. 29 instead of using either the minimum distance Δy or the minimum distance Δy. A configuration using both is also acceptable. At this time, the index generation unit 321 reads the maximum x coordinate value xmax [addr] and the maximum y coordinate value ymax [addr] by referring to the separation coordinate table in the process of S62 shown in FIG. 16, and the process of S63 Thus, the minimum distance Δx and the minimum distance Δy are respectively calculated, and it is determined whether or not the minimum distance Δx is larger than the predetermined threshold lm_x and the minimum distance Δy is larger than the predetermined threshold lm_y in the process of S64.
Here, the predetermined threshold value lm_x may be the same value as the predetermined threshold value lm_y, or may be a different value.

  When the minimum distance Δx> the predetermined threshold lm_x and the minimum distance Δy> the predetermined threshold lm_y (that is, YES in S64), the index generating unit 321 moves the process to S44 to separate the indexes. When the minimum distance Δy ≦ predetermined threshold lm_y or the minimum distance Δx ≦ predetermined threshold lm_x (that is, NO in S64), the index generating unit 321 moves the process to S65 without separating the index.

Incidentally, as shown in FIG. 27, the process of determining the distance between the pixels using the minimum distance Δx in the main scanning direction is a process of rotating the image read by the color image input device 11 by 90 ° to the right by image processing. This is useful when the foreground color indexing processing unit 32 performs processing after the processing.
In the case of the image shown in FIG. 28, the minimum distance Δx in the sub-scanning direction has the same value as the minimum distance Δx in the main scanning direction shown in FIG.
The process of rotating the image 90 ° to the right accepts an instruction to rotate the image 90 ° to the right when the color image input device 11 automatically determines the top and bottom of the image read or when the user operates the operation panel 12. This is done when

FIG. 30 is a flowchart showing a part of a processing procedure using other conditions for determining the distance between pixels using the minimum distance Δx in the main scanning direction and the minimum distance Δy in the sub-scanning direction.
When the index generation unit 321 performs the processing shown in FIGS. 14 to 16 and uses the minimum distance Δx and the minimum distance Δy shown in FIG. 29, instead of executing the processing of S64, FIG. 30 may be configured to execute the processing of S121 to S126 shown in FIG.
As described above, the index generation unit 321 calculates the minimum distance Δx and the minimum distance Δy in the process of S63 illustrated in FIG.

Next, the index generation unit 321 determines whether or not the value of the minimum distance Δy in the sub-scanning direction is greater than a predetermined threshold value lm_y (S121).
When the minimum distance Δy is larger than the predetermined threshold lm_y (YES in S121), the index generation unit 321 determines whether or not the main scanning counter xcnt is smaller than the minimum x coordinate value xmin [addr] (S122).
When the main scanning counter xcnt is smaller than the minimum x coordinate value xmin [addr] (YES in S122), the index generating unit 321 moves the process to S44 to separate the indexes.

When the main scanning counter xcnt is greater than or equal to the minimum x coordinate value xmin [addr] (NO in S122), the index generation unit 321 determines whether or not the main scanning counter xcnt is greater than the maximum x coordinate value xmax [addr]. (S123).
When the main scanning counter xcnt is larger than the maximum x coordinate value xmax [addr] (YES in S123), the index generating unit 321 moves the process to S44.

When the minimum distance Δy is equal to or smaller than the predetermined threshold lm_y (NO in S121), or when the main scanning counter xcnt is equal to or smaller than the maximum x coordinate value xmax [addr] (NO in S123), the index generating unit 321 performs main scanning. It is determined whether or not the value of the minimum direction distance Δx is greater than a predetermined threshold value lm_x (S124).
When the minimum distance Δx is larger than the predetermined threshold lm_x (YES in S124), the index generation unit 321 determines whether or not the sub-scanning counter ycnt is smaller than the minimum y coordinate value ymin [addr] (S125).
If the sub-scanning counter ycnt is smaller than the minimum y coordinate value ymin [addr] (YES in S125), the index generation unit 321 moves the process to S44.

When the sub-scanning counter ycnt is greater than or equal to the minimum y-coordinate value ymin [addr] (NO in S125), the index generation unit 321 determines whether or not the sub-scanning counter ycnt is greater than the maximum y-coordinate value ymax [addr]. (S126).
When the sub-scanning counter ycnt is larger than the maximum y coordinate value ymax [addr] (YES in S126), the index generating unit 321 moves the process to S44.
When the minimum distance Δx is equal to or smaller than the predetermined threshold lm_x (NO in S124), or when the sub-scanning counter ycnt is equal to or smaller than the maximum y coordinate value ymax [addr] (NO in S126), the index generating unit 321 stores the index. Without separation, the process proceeds to step S65.

FIG. 31 is a schematic diagram illustrating a range of pixels in which indexes are separated by the processing of S121 to S126 illustrated in FIG.
In FIG. 31, the area indicated by hatching rising to the right is an area occupied by a pixel associated with a specific index (hereinafter referred to as a specific area). In the range indicated by D1 or D2 in the figure, when there is a pixel associated with the same index as the pixel in the specific area, the index is separated and a new index is associated with the pixel.
Since the pixels within the range indicated by D3 in the figure are scanned before the pixels within the specific area, there is no pixel associated with the same index as the pixels within the specific area.

  As described above, the index generation unit 321 also performs the processing using the distance between the pixels in the main scanning direction and the sub-scanning direction as the distance between the pixels. A foreground layer can be created while reassigning different indexes. The processing using both the distance in the main scanning direction and the sub-scanning direction as the distance between the pixels is the same whether the image read by the color image input device 11 is rotated by image processing or when the image is not rotated. The effect of.

  When the operation panel 12 receives a switching instruction by an operation from a user, the index generation unit 321 performs processing using a distance in the sub-scanning direction as a distance between pixels, processing using a distance in the main scanning direction, A function of switching between processing using distances in both the main scanning direction and the sub-scanning direction is provided.

  By the way, in this embodiment, the case where each of the predetermined threshold value lm_y and the predetermined threshold value lm_x is set to “3” in advance is exemplified, but in practice, it is suitable for an image read by the color image input device 11. A threshold value may be used.

  For example, as the predetermined threshold value lm_y in the sub-scanning direction and the predetermined threshold value lm_x in the main scanning direction, values of 30% of the number of pixels in the sub-scanning direction and the number of pixels in the main scanning direction are set in the index generation unit 321. When the color image input device 11 reads an A3 size image at 600 dpi, the number of pixels in the sub-scanning direction is “10000” and the number of pixels in the main scanning direction is “7000”, so the predetermined threshold lm_y = 3000 and the predetermined The threshold lm_x = 2100 is set in the index generation unit 321. When the color image input device 11 reads an A4 size image at 600 dpi, the number of pixels in the sub-scanning direction is “7000” and the number of pixels in the main-scanning direction is “5000”. Therefore, the predetermined threshold lm_y = 2100 and the predetermined threshold value lm_x = 1500.

  Furthermore, the predetermined threshold value lm_y and the predetermined threshold value lm_x that have been set may be adjusted according to the number of indexes stored in the separation index table.

  After the above-described processing ends, the foreground color indexing processing unit 32 outputs the integrated index image (ie, foreground layer) data, the integrated index color table, the integrated coordinate table, and the image data to the background layer generation processing unit 33. . Hereinafter, the integrated index, the integrated index color table, and the integrated coordinate table are simply referred to as an index, an index color table, and a coordinate table.

The background layer generation processing unit 33 performs processing for generating a background layer as shown in FIG. 2 from the input image data and the foreground layer. The background layer generation processing unit 33 functions as background layer generation means in the present invention.
Specifically, the background layer is created by replacing the color information of the foreground pixels included in the image data with the average value of the color information of the background pixels around the foreground pixels. However, for the foreground pixel in which no background pixel exists in the vicinity, the background layer generation processing unit 33 calculates the color information with the color information value or the average value of the other foreground pixels in which the color information is replaced with the average value of the background pixel. Perform the replacement process.

Here, the background pixel is a pixel associated with the index “0” in the foreground layer, and the foreground pixel is a pixel associated with another index.
The background layer generation processing unit 33 may perform a process of distinguishing the foreground pixels and the background pixels using the foreground mask generated by the foreground mask generation processing unit 31.

  Next, the background layer generation processing unit 33 outputs the generated background layer data, foreground layer data, index color table, and coordinate table to the binary image generation processing unit 34.

  The binary image generation processing unit 34 generates a binary image for each index other than “0” included in the foreground layer based on the input foreground layer data and the coordinate table. Specifically, a binary image in which the pixel value of a pixel associated with a specific index is “1” and the pixel value of another pixel is “0” is generated, and all indexes other than “0” are generated. Repeat the generation of the binary image.

FIG. 32 is a conceptual diagram illustrating an example of a generated binary image.
In the example shown in FIG. 32A, the pixels associated with the index “5” in the foreground layer shown in FIG. 9 (the indexes “5” and “7” in the separated index image shown in FIG. 8 can be associated). This is a binary image in which the pixel value of each pixel is “1” and the pixel values of other pixels are “0”.
Also, the example shown in FIG. 32B is a pixel associated with an index “2” in the foreground layer shown in FIG. 9 (a pixel associated with an index “2” in the initial index image shown in FIG. 7). The remainder is a binary image in which the pixel value of the separated index image shown in Fig. 8 that is associated with the index "6" is "1" and the pixel values of the other pixels are "0". It is.

A binary image is similarly generated even if the indexes “1”, “3”, “4”, and “6” in the foreground layer shown in FIG.
As a result, the same number of binary images as the number of indexes other than “0” included in the foreground layer are generated (six for the foreground layer shown in FIG. 9).
Each generated binary image is associated with each index. The binary image generation processing unit 34 outputs the generated binary image, background layer, index color table, and coordinate table to the image compression processing unit 35. The binary image generation processing unit 34 functions as a binary image generation unit in the present invention.

The image compression processing unit 35 generates compressed image data by compressing each binary image, background layer, index color table, and coordinate table.
Specifically, the image compression processing unit 35 is a rectangle whose pixel range is limited by the minimum x coordinate value, the minimum y coordinate value, the maximum x coordinate value, and the maximum y coordinate value recorded in the coordinate table for each index. A region is extracted from each binary image, and a rectangular region extracted from each binary image is compressed using a reversible compression technique such as MMR.

Further, the image compression processing unit 35 compresses the background layer using an irreversible compression technique such as JPEG, and the color information indicating the color indicated by each index and the pixel of each index are displayed on the image in each rectangular area after compression. The range of the coordinates occupied by (ie, the position on the image of the rectangular area related to each index) is added, and all the compressed data is integrated into one data file, thereby generating compressed image data.
The image compression processing unit 35 functions as compression means in the present invention.

The image compression processing unit 35 may have a function of changing the data compression format in accordance with a predetermined instruction received on the operation panel 12.
In addition, the image compression processing unit 35 does not extract the rectangular area, but the foreground color indexing processing unit 32 or the binary image generation processing unit 34 extracts the rectangular area and outputs the extracted rectangular area to the subsequent stage. Good.

The compression processing unit 3 generates compressed image data obtained by compressing the image input from the color image input device 11 by the above processing, and outputs the compressed image data to the storage unit 29 or the transmission device 14.
The storage unit 29 stores the compressed image data, reads out the stored compressed image data according to a predetermined instruction received by the operation panel 12, and outputs the compressed image data to the transmission device 14.
The transmission device 14 transmits the compressed image data to the outside in accordance with a predetermined instruction received on the operation panel 12.

  As described above in detail, the image processing apparatus according to the embodiment of the present invention extracts pixels corresponding to characters from an image, and associates each pixel corresponding to the character with an index indicating the color of the pixel as a numerical value. Generating a foreground layer, generating a background layer in which characters are omitted from the image, generating a binary image in which for each index, a pixel associated with each index and other pixels are distinguished from the foreground layer, Compressed image data is generated by compressing all binary images and background layers.

When generating the foreground layer, the image processing apparatus associates an index corresponding to the color with each pixel, and if there is already a pixel associated with the same index as the target pixel, the index is determined in advance. It is determined whether the distance is more than. Further, the image processing apparatus associates indexes having different values while indicating the same color as the index of the target pixel if the indexes are separated by a predetermined distance or more, and the same if the indices are not separated by a predetermined distance or more. Associate the index of.
Further, the image processing apparatus integrates indexes indicating colors that are similar to each other in each index, and does not integrate indexes that indicate the same color.

  When a binary image is generated from the foreground layer, pixels that have the same color but are associated with different indexes are included in different binary images, and in each binary image, the pixel value The “1” pixels are concentrated in a narrower region than in the conventional technique. Therefore, in the rectangular area including the pixels related to each index extracted from each binary image, the ratio of the pixels associated with each index is higher than that of the related art, and unnecessary data included in the compressed data Can be reduced.

  For example, when the colors of two regions separated on the image are the same, the conventional technique extracts a rectangular region including the two regions, whereas creating compressed data including pixels between the regions, In the present invention, different indexes are associated with the two regions, and rectangular regions including each region are individually extracted and compressed individually, so that the compressed data does not include pixels between the regions.

  As a result, since the index is efficiently included in the rectangular area (pixels that are not the index are minimized), it is not necessary to transfer unnecessary image area data between the two index areas when performing subsequent processing. The data transfer amount can be reduced and the processing speed can be improved. In addition, since the respective rectangular areas are also reduced, the possibility that the rectangular areas overlap each other is reduced, and the speed at which the compressed data is expanded and printed can be improved.

  As described above, in the present invention, unnecessary data included in the compressed data can be reduced, so the amount of compressed image data obtained by compressing a color image is reduced, and the compression efficiency of the color image is improved. Can be made. Therefore, it is possible to shorten the processing time required for compressing the color image. Further, when transmitting / receiving compressed image data, the amount of data to be transmitted / received can be reduced, and the time required for transmission / reception can be shortened.

Embodiment 2. FIG.
In the first embodiment, the image forming apparatus of the present invention is an apparatus that forms a color image. In the second embodiment, the image forming apparatus of the present invention is an apparatus that forms a monochrome image. Show.
FIG. 33 is a block diagram illustrating an internal functional configuration of the image forming apparatus according to the second embodiment.
The image forming apparatus includes a gray image input device 15 that optically reads an image in a monochrome gray scale. The gray image input device 15 includes a monochrome image processing that is an image processing device according to an embodiment of the present invention. A device 4 is connected.

The monochrome image processing device 4 is connected to a monochrome image output device 16 that outputs a monochrome image based on the image data generated by the monochrome image processing device 4 and a transmission device 14.
An operation panel 12 is connected to the gray image input device 15, the monochrome image processing device 4, the monochrome image output device 16, and the transmission device 14.
The gray image input device 15 is composed of a scanner equipped with a CCD optical sensor, reads a reflected light image from an image recorded on a record carrier such as paper by the CCD, and is grayscale with a green (G) single color. Are converted into analog signals and input to the monochrome image processing apparatus 4.

The monochrome image processing device 4 performs image processing on the G single-color analog signal input from the gray image input device 15 to generate digital image data. The monochrome image processing device 4 generates image data composed of digital K signals and outputs the image data to the monochrome image output device 16. Further, the monochrome image processing device 4 generates compressed image data obtained by compressing the generated digital image data by compression processing, and outputs the compressed image data to the transmission device 14.
The monochrome image output device 16 outputs a monochrome image based on the image data input from the monochrome image processing device 4.
The configurations of the operation panel 12 and the transmission device 14 are the same as those in the first embodiment.

  The monochrome image processing device 4 converts the analog signal input from the gray image input device 15 into a digital signal by the A / D conversion unit 40, and converts the shading correction unit 41, the input gradation correction unit 42, the region separation processing unit 43, The spatial filter processing unit 46, the output tone correction unit 47, and the tone reproduction processing unit 48 are sent in this order, and image data composed of digital K signals is output to the monochrome image output device 16. The monochrome image processing device 4 is configured to send a digital signal from the region separation processing unit 43 to the compression processing unit 45 and to output compressed image data to the transmission device 14.

The A / D conversion unit 40 receives the analog signal input from the gray image input device 15, converts the analog G signal into a digital G signal, and outputs the digital G signal to the shading correction unit 41.
The shading correction unit 41 performs processing for removing various distortions on the G signal input from the A / D conversion unit 40, and outputs the G signal from which the distortion has been removed to the input tone correction unit 42.
The input tone correction unit 42 adjusts the balance of the intensity of the G signal input from the shading correction unit 41, and further converts the G signal into a signal such as a density signal that can be easily processed by the monochrome image processing device 4. The subsequent G signal is output to the region separation processing unit 43.

The region separation processing unit 43 separates each pixel in the image represented by the G signal input from the input tone correction unit 42 into one of a character region, a dot region, and a photo region, and each pixel is A region identification signal indicating whether the image belongs to the region is output to the spatial filter processing unit 46 and the gradation reproduction processing unit 48, respectively, and the G signal is output to the spatial filter processing unit 46 and the compression processing unit 45, respectively.
The spatial filter processing unit 46 performs spatial filter processing on the image represented by the input G signal based on the region identification signal, and outputs the processed G signal to the output tone correction unit 47.
The output tone correction unit 47 performs output tone correction processing based on the characteristics of the monochrome image output device 16 for the G signal input from the spatial filter processing unit 46, and performs tone reproduction processing on the processed G signal. To the unit 48.

The gradation reproduction processing unit 48 performs halftone processing corresponding to the region on the G signal input from the output gradation correction unit 47 based on the region identification signal, and converts the processed G signal to K Image data consisting of signals is output to the monochrome image output device 16.
The compression processing unit 45 receives the image data composed of the G signal from the region separation processing unit 43, executes a process of generating compressed image data using the image processing method according to the embodiment of the present invention, and the compressed image data Is stored in the storage unit 29, and the compressed image data is output to the transmission device 14. The image processing method executed by the compression processing unit 45 is executed by the compression processing unit 3 according to the first embodiment except that the color information is not information indicating the color intensity of each RGB color but information indicating the intensity of the G single color. This is equivalent to the image processing method to be performed.

FIG. 34 is a table showing an example of an index color table used by the compression processing unit 45 according to the second embodiment.
Each address is associated with information indicating the intensity of the G single color as color information, and the address value is equal to the index value. In addition, “0”, which is color information indicating white, is associated with the index “0” corresponding to the background pixel.
The compression processing unit 45 generates compressed image data by executing an image processing method similar to the image processing method executed by the compression processing unit 3 of the first embodiment using the index color table as shown in FIG. To do.

As described above in detail, the image processing apparatus according to the present embodiment performs monochrome image processing by performing image processing similar to that of Embodiment 1 on a monochrome image whose pixel color information indicates a single color intensity. Compressed image data is generated.
Also in the present embodiment, when the foreground layer is generated, the image processing apparatus associates an index corresponding to the color with each pixel, and there is already a pixel associated with the same index as the target pixel. Determines whether these indexes are more than a predetermined distance apart. Further, the image processing apparatus associates indexes having different values while indicating the same color as the index of the target pixel if the indexes are separated by a predetermined distance or more, and the same if the indices are not separated by a predetermined distance or more. Associate the index of.

Further, the image processing apparatus integrates indexes indicating colors that are similar to each other in each index, and does not integrate indexes that indicate the same color.
As a result, unnecessary data included in the compressed data can be reduced. Therefore, it is possible to reduce the data amount of the compressed image data obtained by compressing the monochrome image and improve the compression efficiency of the monochrome image.

Embodiment 3. FIG.
In the second embodiment, the image forming apparatus of the present invention is an apparatus that receives a monochrome image and forms a monochrome image. However, in the third embodiment, the image forming apparatus of the present invention receives a color image. The form which is an apparatus which forms a monochrome image is shown.

FIG. 35 is a block diagram illustrating an internal functional configuration of the image forming apparatus according to the third embodiment.
The image forming apparatus includes a color image input device 11 that optically reads a color image. The color image input device 11 is connected to an image processing device 5 that is an image processing device according to an embodiment of the present invention. ing. A monochrome image output device 16 and a transmission device 14 are connected to the image processing device 5. An operation panel 12 is connected to the color image input device 11, the image processing device 5, the monochrome image output device 16, and the transmission device 14.
The color image input device 11 performs the same processing as in the first embodiment, and outputs RGB analog signals obtained by reading the color image to the image processing device 5.

The image processing device 5 performs image processing on the RGB analog signals input from the color image input device 11, generates image data composed of digital monochromatic K signals, and outputs the image data to the monochrome image output device 16. . Further, the image processing device 5 generates compressed image data obtained by compressing image data composed of digital RGB signals, and stores the generated compressed image data in the storage unit 29 or outputs it to the transmission device 14. The compressed image data stored in the storage unit 29 and the image represented by the compressed image data transmitted by the transmission device 14 are color images.
The monochrome image output device 16 outputs a monochrome image based on the image data input from the image processing device 5.
The configurations of the operation panel 12 and the transmission device 14 are the same as those in the first embodiment.

  The image processing device 5 converts the analog signal input from the color image input device 11 into a digital signal by the A / D conversion unit 20, and the shading correction unit 21, the input tone correction unit 22, the region separation processing unit 53, the color The correction unit 54, the black generation and under color removal unit 55, the spatial filter processing unit 46, the output gradation correction unit 47, and the gradation reproduction processing unit 48 are sent in this order, and image data composed of digital K signals is sent to the monochrome image output device 16. It is the composition to output. The image processing device 5 is configured to send a digital signal from the region separation processing unit 53 to the compression processing unit 3 and to output compressed image data to the transmission device 14.

The configurations of the A / D conversion unit 20, the shading correction unit 21, and the input tone correction unit 22 are the same as those in the first embodiment.
The region separation processing unit 53 separates each pixel in the image represented by the RGB signal input from the input tone correction unit 22 into one of a character region, a halftone dot region, and a photograph region, and each pixel is A region identification signal indicating whether the image belongs to the region is output to the black generation and under color removal unit 55, the spatial filter processing unit 46, and the gradation reproduction processing unit 48, respectively.
Further, the region separation processing unit 53 outputs the RGB signals input from the input tone correction unit 22 to the color correction unit 54 and the compression processing unit 3, respectively.

The color correction unit 54 extracts a G signal from the RGB signal input from the region separation processing unit 53, and outputs the extracted G signal to the black generation and under color removal unit 55.
The black generation and under color removal unit 55 outputs the G signal from the color correction unit 54 to the spatial filter processing unit 46.
The configurations of the spatial filter processing unit 46, the output tone correction unit 47, and the tone reproduction processing unit 48 are the same as those in the second embodiment.
The configuration of the compression processing unit 3 is the same as that of the first embodiment, and the compression processing unit 3 receives image data composed of RGB signals from the region separation processing unit 53, and implements the present invention as in the first embodiment. The compressed image data is generated by executing the image processing method according to the embodiment.

  The color correction unit 54 may be configured to perform a process of generating a luminance signal from the RGB signal instead of extracting the G signal from the RGB signal. For example, the color correcting unit 54 generates a luminance signal by calculating the luminance value Y from the color intensity RGB of each RGB color as Y = 0.30R + 0.59G + 0.11B. In the case of this form, the black generation and under color removal unit 55, the spatial filter processing unit 46, the output gradation correction unit 47, and the gradation reproduction processing unit 48 execute processing using the luminance signal as color information.

As described above in detail, the image processing apparatus according to the present embodiment can form a monochrome image and generate color compressed image data.
Also in the present embodiment, when the foreground layer is generated, the image processing apparatus associates an index corresponding to the color with each pixel, and there is already a pixel associated with the same index as the target pixel. Determines whether these indexes are more than a predetermined distance apart. Further, the image processing apparatus associates indexes having different values while indicating the same color as the index of the target pixel if the indexes are separated by a predetermined distance or more, and the same if the indices are not separated by a predetermined distance or more. Associate the index of.

Further, the image processing apparatus integrates indexes indicating colors that are similar to each other in each index, and does not integrate indexes that indicate the same color.
As a result, unnecessary data included in the compressed data can be reduced. Therefore, it is possible to reduce the data amount of the compressed image data obtained by compressing the color image and improve the compression efficiency of the color image.

Embodiment 4 FIG.
In the first to third embodiments, the image processing apparatus of the present invention forms part of the image forming apparatus. However, in the fourth embodiment, the image processing apparatus of the present invention forms part of the scanner apparatus. The form is shown.

FIG. 36 is a block diagram illustrating an internal functional configuration of a scanner apparatus including the image processing apparatus according to the fourth embodiment.
The scanner device includes a color image input device 11 that optically reads a color image, and the color image input device 11 is connected to an image processing device 6 according to an embodiment of the present invention.
A host device (not shown) such as a personal computer (PC) is connected to the image processing device 6 via a communication cable or a communication network.
An operation panel 12 is connected to the color image input device 11 and the image processing device 6.

The color image input device 11 performs the same processing as in the first embodiment and outputs RGB analog signals obtained by reading the color image to the image processing device 6.
The image processing device 6 converts an analog signal input from the color image input device 11 into a digital signal by the A / D conversion unit 20, and converts the shading correction unit 21, the input tone correction unit 22, the region separation processing unit 63, and the compression It is configured to send in order of the processing unit 3.
The configurations of the A / D conversion unit 20, the shading correction unit 21, and the input tone correction unit 22 are the same as those in the first embodiment.
The region separation processing unit 63 outputs the RGB signal input from the input tone correction unit 22 to the compression processing unit 3.

The configuration of the compression processing unit 3 is the same as that of the first embodiment, and the compression processing unit 3 is inputted with image data composed of RGB signals from the region separation processing unit 63, and implements the present invention as in the first embodiment. By executing the image processing method according to the embodiment, compressed image data obtained by compressing the input image is generated. The compression processing unit 3 outputs the generated compressed image data to a host device (not shown).
The host device receives the compressed image data output from the image processing device 6 and executes processing such as storing the compressed image data, transmitting the compressed image data to the outside, or outputting the image based on the compressed image data.

As described above, also in the present embodiment, as in the first to third embodiments, when generating the foreground layer, the image processing apparatus associates an index corresponding to a color with each pixel, If pixels that are associated with the same index already exist, it is determined whether those indexes are separated by a predetermined distance or more. Further, the image processing apparatus associates indexes having different values while indicating the same color as the index of the target pixel if the indexes are separated by a predetermined distance or more, and the same if the indices are not separated by a predetermined distance or more. Associate the index of.
Further, the image processing apparatus integrates indexes indicating colors that are similar to each other in each index, and does not integrate indexes that indicate the same color.

  As a result, unnecessary data included in the compressed data can be reduced. Accordingly, the amount of compressed image data obtained by compressing a color image can be reduced, and the compression efficiency of the image can be improved. When transmitting / receiving compressed image data, the amount of data to be transmitted / received is reduced, It is possible to reduce the time required for transmission / reception.

Embodiment 5 FIG.
In the fifth embodiment, an embodiment in which the image processing apparatus of the present invention is realized using a general-purpose computer will be described.
FIG. 37 is a block diagram showing an internal configuration of the image processing apparatus 7 according to the fifth embodiment.
The image processing apparatus 7 according to the present embodiment is configured using a general-purpose computer such as a PC, and includes a CPU 71 that performs a calculation, a RAM 72 that stores temporary information generated along with the calculation, an optical disk, and the like. A drive unit 73 such as a CD-ROM drive for reading information from a recording medium and a storage unit 74 such as a hard disk are provided.

  The CPU 71 causes the drive unit 73 to read the computer program 81 according to the embodiment of the present invention from the recording medium 8 according to the embodiment of the present invention, and causes the storage unit 74 to store the read computer program 81. The computer program 81 is loaded from the storage unit 74 to the RAM 72 as necessary, and the CPU 71 executes processing necessary for the image processing device 7 based on the loaded computer program 81.

In addition, the image processing apparatus 7 includes an input unit 75 such as a keyboard or a pointing device for inputting information such as various processing instructions operated by a user, and a display unit 76 such as a liquid crystal display for displaying various information. And. Furthermore, the image processing apparatus 7 includes a transmission unit 77 that can be connected to an external communication network (not shown), and a reception unit 78 that is connected to an external input device 82 such as a scanner device that inputs image data.
The input device 82 optically reads an image to generate image data, and transmits the generated image data to the image processing device 7. The receiving unit 78 receives image data transmitted from the input device 82. The receiving unit 78 functions as an accepting unit in the present invention. The transmission unit 77 can transmit data to the outside by a communication method such as facsimile or e-mail via a communication network (not shown).

  The CPU 71 loads the computer program 81 into the RAM 72, and executes the image processing method according to the embodiment of the present invention in accordance with the loaded computer program 81. That is, when image data is input from the input device 82 by the receiving unit 78, the CPU 71 performs the foreground mask generation processing unit 31, the foreground color indexing processing unit 32, the background layer generation processing unit 33, 2 in the first embodiment. The foreground mask generation process, the foreground color indexing process, the background layer generation process, the binary image generation process, and the image compression process similar to the processes performed by the value image generation processing unit 34 and the image compression processing unit 35 are executed. Thereby, the CPU 71 performs a process of generating compressed image data obtained by compressing the received image data.

The CPU 71 stores the generated compressed image data in the storage unit 74. Further, the CPU 71 performs processing for causing the transmission unit 77 to transmit the generated compressed image data or the compressed image data read from the storage unit 74 to the outside in accordance with the loaded computer program 81.
Note that the CPU 71 may be configured to receive image data created by the user using the input unit 75 and to generate compressed image data obtained by compressing the received image data. In this case, the CPU 71 functions as an accepting unit in the present invention.

As described above, also in the present embodiment, as in the first to fourth embodiments, when generating the foreground layer, the image processing apparatus associates an index corresponding to the color with each pixel, If pixels that are associated with the same index already exist, it is determined whether the indexes are separated by a predetermined distance or more. Further, the image processing apparatus associates indexes having different values while indicating the same color as the index of the target pixel if the indexes are separated by a predetermined distance or more, and the same if the indices are not separated by a predetermined distance or more. Associate the index of.
Further, the image processing apparatus integrates indexes indicating colors that are similar to each other in each index, and does not integrate indexes that indicate the same color.

  As a result, unnecessary data included in the compressed data can be reduced. Therefore, it is possible to reduce the data amount of the compressed image data obtained by compressing the color image and improve the compression efficiency of the image.

  The recording medium 8 on which the computer program 81 is recorded is a magnetic tape, a magnetic disk, a portable hard disk, an optical disk such as a CD-ROM / MO / MD / DVD, or an IC card (including a memory card) / light. Any form of a card-type recording medium such as a card may be used. Further, the recording medium 8 of the present invention is mounted on the image processing apparatus 7, and a semiconductor memory from which the CPU 71 can read out the recorded contents of the recording medium 8, that is, mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM ( Electrically Erasable Programmable Read Only Memory), flash ROM, or the like may be used.

  Furthermore, the computer program 81 of the present invention is downloaded to the image processing apparatus 7 from an external server device (not shown) connected to the image processing apparatus 7 via a communication network such as the Internet or a LAN and stored in the storage unit 74. Form may be sufficient. In the case of this form, a program necessary for downloading the computer program 81 is stored in the storage unit 74 in advance, or is read from the predetermined recording medium using the drive unit 73 and stored in the storage unit 74. What is stored is only required to be loaded into the RAM 72 as necessary.

  In Embodiments 1 to 5 described above, the mode of extracting a pixel corresponding to a character as a pixel included in the foreground layer has been described. However, the present invention is not limited to this, and the present invention is included in the foreground layer. Alternatively, a process for extracting a pixel corresponding to a line drawing may be performed. Further, the present invention may be configured to perform processing for extracting pixels corresponding to characters and line drawings as pixels included in the foreground layer.

DESCRIPTION OF SYMBOLS 11 Color image input device 12 Operation panel 13 Color image output device 14 Transmission device 15 Gray image input device 16 Monochrome image output device 2 Color image processing device 3,45 Compression processing unit 31 Foreground mask generation processing unit 32 Foreground color indexing processing unit 321 Index generation unit 322 Index integration unit 323 Index correction unit 33 Background layer generation processing unit 34 Binary image generation processing unit 35 Image compression processing unit 4 Monochrome image processing device 5, 6, 7 Image processing device 71 CPU
8 Recording medium 81 Computer program

Claims (10)

Receiving means for receiving an image composed of a plurality of pixels;
Extraction means for extracting pixels corresponding to characters and / or line drawings out of the pixels constituting the image received by the reception means;
Foreground layer generation means for generating a foreground layer representing the color of each pixel extracted by the extraction means by any of a plurality of identifiers;
Background layer generation means for generating a background layer from which characters and / or line drawings are omitted from the image received by the reception means;
Binary image generating means for generating a binary image composed of a pixel represented by a specific identifier and other pixels for each identifier included in the foreground layer generated by the foreground layer generating means;
Compression means for compressing the binary image generated by the binary image generation means and the background layer generated by the background layer generation means,
In an image processing apparatus that performs processing for compressing an image,
The foreground layer generation means includes
An identifier corresponding means for associating an identifier associated with a specific color with each pixel according to the color of each pixel;
An identifier separating unit that separates identifiers by associating different identifiers between pixels that are separated from a predetermined distance on the image among pixels associated with the same identifier by association by the identifier association unit;
An identifier integration unit that integrates, among different identifiers, identifiers whose colors associated with the identifiers are approximated within a predetermined range after association by the identifier association unit and separation by the identifier separation unit; An image processing apparatus comprising: The identifier integrating means includes
After the association by the identifier correspondence unit and the separation by the identifier separation unit, the colors associated with the identifiers among the different identifiers are predetermined except for identifiers in which the associated colors are the same. The image processing apparatus according to claim 1, wherein identifiers that are approximated within the range are integrated. The identifier separating means includes
Selection means for sequentially selecting pixels according to a predetermined selection order;
A minimum calculation means for calculating a minimum distance between a pixel associated with the same identifier as the pixel being selected among the selected pixels and the pixel being selected;
Means for changing an identifier associated with the selected pixel to a new identifier associated with the same color as the identifier when the minimum distance calculated by the minimum operation means is greater than a predetermined threshold; The image processing apparatus according to claim 1, wherein the image processing apparatus is provided. The selection means is configured to sequentially select pixels, with one direction on the image as a main scanning direction and a direction orthogonal to the one direction as a sub-scanning direction,
4. The apparatus according to claim 3, further comprising means for switching a distance in the main scanning direction, a distance in the sub scanning direction, and a distance in both the main scanning direction and the sub scanning direction as the distance calculated by the minimum calculating means. An image processing apparatus according to 1.   The image processing apparatus according to claim 1, wherein the color of each pixel includes information indicating the color intensities of a plurality of primary colors.   The image processing apparatus according to claim 1, wherein the color of each pixel includes information indicating a color intensity of a single color.   An image forming apparatus comprising the image processing apparatus according to claim 1. Receiving an image composed of a plurality of pixels;
Extracting a pixel corresponding to a character and / or a line drawing from among pixels constituting the received image;
A foreground layer generation step of generating a foreground layer representing the color of each extracted pixel by any of a plurality of identifiers;
Generating a background layer without characters and / or line drawings from the received image;
Generating a binary image composed of a pixel represented by a specific identifier and other pixels for each identifier included in the foreground layer;
Compressing the generated binary image and the background layer, respectively,
The foreground layer generation step includes
Associating each pixel with an identifier associated with a particular color, depending on the color of each pixel;
Separating the identifiers by associating different identifiers between pixels that are separated from the predetermined distance on the image among pixels associated with the same identifier;
And after integrating and separating the identifiers, integrating the identifiers of the different identifiers whose colors associated with the identifiers are approximated within a predetermined range. . On the computer,
A procedure for extracting pixels corresponding to characters and / or line drawings from among pixels constituting the image;
A foreground layer generation procedure for generating a foreground layer representing the color of each extracted pixel by any of a plurality of identifiers;
Generating a background layer from which characters and / or line drawings are omitted from the image;
For each identifier included in the foreground layer, a procedure for generating a binary image composed of a pixel represented by a specific identifier and other pixels;
A computer program for executing a process of compressing an image by executing a process including a procedure of compressing a generated binary image and a background layer respectively.
The foreground layer generation procedure includes:
Depending on the color of each pixel, a procedure for associating an identifier associated with a specific color with each pixel;
A procedure of separating identifiers by associating different identifiers among pixels that are separated from a predetermined distance on the image among pixels associated with the same identifier;
And a step of integrating, after the association and separation of the identifiers, the identifiers of the different identifiers whose colors are similar to each other within a predetermined range.   A computer-readable recording medium on which the computer program according to claim 9 is recorded.

Images