JP2012134847A - Image compression apparatus, image compression method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve color change of a character on a preview screen.SOLUTION: An image compression apparatus which divides an image composed of multiple different colors into rectangle-shaped image areas to be compressed comprises: first image area specifying means for specifying a rectangle-shaped image area having first color information; and second image area specifying means for specifying a rectangle-shaped image area having second color information which is adjacent to the rectangle-shaped image area having the first color information. In the apparatus, a first compression code and a second compression code are generated, a data amount of the second compression code is subtracted from a total compression code which is a sum of data amounts of the first compaction code and the second compaction code, vector information which is information identifying the second color information is generated based on the second color information, and the vector information is stored in a part of the data amount of the subtracted second compression code, and a compression image is generated.

Description

  The present invention relates to an image compression apparatus, an image compression method, and an image compression program.

  There is a technique for visually generating one image by combining a plurality of images. For example, there is a technique for forming one composite image from a character image representing a character shape, a background image representing a pattern, a foreground image representing a character color, and a black character image representing the position of a black character. That is, information of one pixel on the synthesized image has one or more pieces of image information, and which image has information differs depending on the pixel position. A character image, a background image, a foreground image, and a black character image are generated from one scanner image, but a high-compression file can be generated by decomposing each image.

  Among the images forming the composite image, the foreground image is subjected to block-unit JPEG (Joint Photographic Experts Group) compression. If a block containing many different colors is compressed at a reduced resolution, the original color is often lost when the block is restored. In order to correct such a color change of a character, there is a technique in which a generated image is once displayed on a monitor (touch panel) provided in the image processing apparatus and a user can instruct correction on the monitor.

  In Patent Document 1, a character area is extracted for the purpose of setting the color of a character area as intended by a user as a means of correcting a color mixture or error that occurs when each component image is generated from a scanner image. The image corresponding to the character area is generated, a composite image having the generated image as one component is displayed as a preview, and the character area has a means for converting the color of the character area into the specified color. A technique for converting to the above is disclosed.

  However, the technology so far relates to a plurality of images representing the shape of the character area and a file structure for expressing the character color in one color for each character area. ) Has a problem that it cannot be applied as it is. Therefore, the problem that the legibility of characters cannot be improved has not been solved.

  Therefore, the present invention has been made in view of the above problems, and in a file creation method for expressing a character color as an image, an image compression device, an image compression method, and an image compression device capable of improving a character color change on a preview screen. And to provide a program.

  In order to solve the above-mentioned problem, the image compression apparatus according to the present invention described in claim 1 is an image compression apparatus having compression means for compressing an image composed of a plurality of different colors by dividing the image into rectangular image areas. A first image area specifying unit that specifies a rectangular image area having first color information from a plurality of character images; and the first image area specifying unit that is specified by the first image area specifying unit. Second image area specifying means for specifying a rectangular image area having second color information from among the plurality of character images adjacent to the rectangular image area having color information, and the compression The means includes a first compressed code generating means for generating a first compressed code, a second compressed code generating means for generating a second compressed code, a data amount of the first compressed code, and the second With the amount of compression code data The second color information is identified based on the second color information and a data amount reducing means for subtracting the data amount corresponding to the data amount of the second compressed code from the total compressed code data amount Vector information generating means for generating vector information that is information to be performed, and the vector information generated by the vector information generating means in a portion of the data amount corresponding to the reduced data amount of the second compressed code And a compressed image information generating means for generating compressed image information in which the vector information is stored in the data of the first compressed code.

  The image compression apparatus according to the present invention is the image compression apparatus according to claim 1, further comprising display means for displaying the compressed image information as a preview, and the second image area specifying means displays a preview on the display means. A rectangular image area having the second color information is specified by specifying an area for the image to be displayed.

  Furthermore, the image compression apparatus according to the present invention is the image compression apparatus according to claim 1 or 2, wherein the rectangular image area having the second color information is a rectangular image having the first color information. It is characterized by being specified from the upper, lower, left and right areas adjacent to the area.

  The image compression device according to the present invention is the image compression device according to claim 2 or 3, wherein a plurality of rectangular image regions each having a plurality of different color information, and the plurality of the plurality of image regions. A plurality of vector information that is information for identifying color information is prepared in advance as an image area information group, and when the area is designated, the second color information is selected from the image area information group. A rectangular image area is specified.

  The image compression method according to the fifth aspect of the present invention is an image compression method in an image compression apparatus having compression means for dividing and compressing an image composed of a plurality of different colors into rectangular image regions. A step of specifying a rectangular image region having first color information from a plurality of character images by a first image region specifying unit; and a step of specifying the first image by a second image region specifying unit. Identifying a rectangular image area having second color information from among the plurality of character images adjacent to the rectangular image area having the first color information specified by the area specifying means; The compression means includes a step of generating a first compression code by the first compression code generation means, a step of generating a second compression code by the second compression code generation means, and a data amount Reduction hand Accordingly, the data amount corresponding to the data amount of the second compression code is reduced from the data amount of the total compression code which is the sum of the data amount of the first compression code and the data amount of the second compression code. Reduced by the vector information generating means, the vector information generating means generating vector information as information for identifying the second color information based on the second color information, and the vector information storing means. Storing the vector information generated by the vector information generating means in a portion of the data amount corresponding to the data amount of the second compressed code, and the compressed image information generating means by the first compressed code. Generating compressed image information in which the vector information is stored in the data.

  According to a sixth aspect of the present invention, there is provided a program executed by a computer of an image compression apparatus having a compression unit that compresses an image composed of a plurality of different colors by dividing the image into rectangular image areas. Then, the first image area specifying means specifies the rectangular image area having the first color information from the plurality of character images, and the second image area specifying means determines the first image. Identifying a rectangular image area having second color information from among the plurality of character images adjacent to the rectangular image area having the first color information specified by the area specifying means; The compression means includes: a process for generating a first compression code by the first compression code generation means; a process for generating a second compression code by the second compression code generation means; The data amount of the first compressed code is equivalent to the data amount of the second compressed code from the data amount of the total compressed code, which is the sum of the data amount of the first compressed code and the data amount of the second compressed code. A process for reducing the amount of data, a process for generating vector information, which is information for identifying the second color information, based on the second color information by the vector information generating means, and a vector information storing means, A process of storing the vector information generated by the vector information generating means in a portion of the data amount corresponding to the reduced data amount of the second compressed code, and a compressed image information generating means And processing for generating compressed image information in which the vector information is stored in the data of one compressed code.

  According to the present invention, by compressing a block on a foreground image corresponding to an area for which a character color change is designated on the preview screen while reducing the color included in the block, the compression code of the block is compressed. Reduce the amount and store the reduced color as vector information of the color adjacent to the block that is close to the reduced color in the portion of the reduced compressed code, and simultaneously handle the reduced color Since information for identifying pixels to be stored is stored, an image compression apparatus and an image compression method capable of improving a color change of a character on a preview screen in a technique for expressing a character color as an image (foreground image) And a program can be obtained.

1 is a block diagram illustrating a schematic configuration of an image processing apparatus according to an embodiment of the present invention. It is a figure explaining the format process of the high compression PDF in embodiment of this invention. It is a figure explaining the color change in embodiment of this invention. It is a figure explaining about the color change in embodiment of this invention. It is a block diagram which shows the flow of the data in embodiment of this invention. It is a conceptual diagram of the separation process of the original image in the embodiment of the present invention. It is a flowchart which shows the flow of operation | movement of the controller in embodiment of this invention. It is a figure which shows the image which acquires the process target image information corresponding to the block of the foreground image containing the character pixel in the designated range in this invention. It is a flowchart which shows the flow which determines the block to correct from the correction character pixel candidates in embodiment of this invention. It is a schematic diagram which shows that the block of the process target in embodiment of this invention can be substituted by the further previous block. It is a schematic diagram which shows the state of the pixel accumulate | stored in HDD in embodiment of this invention. It is a figure explaining putting correction pixel information into the part which deleted the compression code in the embodiment of the present invention. It is a figure which shows the process sequence of the basic system of JPEG compression in embodiment of this invention. It is a figure which shows the table of the zigzag scan of the JPEG image in embodiment of this invention. It is a figure which shows an example of the Huffman code table of AC component in JPEG compression in embodiment of this invention. In entropy coding in an embodiment of the present invention, if entropy decreases, the bit size after replacement decreases. It is a figure explaining an invalid area | region increasing by the part of the reduced bit number in the entropy encoding in embodiment of this invention. It is a figure which shows the vector to the block with the color to replace in other embodiment of this invention.

  Next, embodiments for carrying out the present invention will be described in detail with reference to the drawings. In addition, in each figure, the same code | symbol is attached | subjected to the part which is the same or it corresponds, The duplication description is simplified thru | or abbreviate | omitted suitably.

  The present invention relates to an image format in which a plurality of images are combined and displayed, and in a technique for expressing a character color as an image (foreground image), it corresponds to a designated area with respect to a color change of a character displayed on a preview screen. By compressing the block on the foreground image by reducing the data amount of the color included in the block, the compression code amount of the block is reduced, and the color whose data amount has been reduced is reduced. The vector information of the color adjacent to the block close to the reduced color is stored in the reduced compressed code portion, and at the same time, the information for identifying the pixel corresponding to the reduced color is stored. Yes.

  First, an image processing apparatus in which the image compression apparatus according to the present invention is mounted will be described with reference to FIG. FIG. 1 is a block diagram showing a schematic configuration of a digital image processing apparatus according to an embodiment of the present invention. When the color image processing apparatus 1 operates as a copying machine, the scanner 2 reads image data from the document 3, converts the image data (analog signal) into digital data, and outputs the digital data.

  As will be described later, the scanner correction unit 4 classifies the image area into characters, line drawings, photographs, etc. with respect to the image data (digital data) output from the scanner 2, or RGB (Red Green Blue) of the document image. Image processing such as filter processing is performed on the data.

  The compression processing unit 5 compresses the RGB corrected 8-bit image data, the edge character area signal (1 bit), and the color area signal (1 bit), and sends the data to the general-purpose bus 6. The compressed image data is sent to the controller 7 through the general-purpose bus 6. The controller 7 has a semiconductor memory (not shown) and accumulates sent data. In the accumulated data, the image size and the type of the read original are recorded as bibliographic information.

  Although the compression processing is performed on the image data here, if the bandwidth of the general-purpose bus 6 is sufficiently wide and the capacity of the HDD (Hard Disk Drive) to be stored is large, the image data is handled in an uncompressed state. May be.

  Next, the controller 7 sends the image data of the HDD to the decompression processing unit 8 via the general-purpose bus 6. The decompression processing unit 8 decompresses the compressed image data into the original RGB 8-bit data, edge character region signal (1 bit), and color region signal (1 bit), and sends them to the printer correction unit 9.

  In the printer correction unit 9, the RGB image data is converted into YMCBk (Yellow Magenta Cyan Black) data by the color correction processing unit, and the portion which is an edge character region signal and not a color region signal is replaced with Bk single color data as a black character. . Further, γ correction processing, halftone processing, and the like are performed, and the light / dark characteristic correction processing and tone number conversion processing of the plotter 10 are performed.

  In the gradation number conversion process here, image data is converted from 8 bits to 2 bits for each color using error diffusion and dither processing. The plotter 10 is a transfer paper printing unit using a laser beam writing process, draws 2-bit image data as a latent image on a photoconductor, and forms a copy image 11 on the transfer paper after image formation / transfer processing with toner.

  When operating as a distribution scanner that distributes image data to the external PC terminal 13 via the network 12, the controller 7 performs color conversion processing, gradation processing, format processing, and the like on the image data stored in the HDD. .

  In the gradation processing, gradation conversion processing is performed according to the mode during the distribution scanner operation. In the format processing, general-purpose image format conversion to JPEG, TIFF (Tagged Image File Format) or PDF (Portable Document Format) format is performed. Depending on the setting, image data of a highly compressed PDF image is generated, and a PDF header is added to those images. Thereafter, the image data is distributed to the external PC terminal 13 via a NIC (Network Interface Controller) 15.

  The delivered image data can be preview-checked by a preview image display unit provided in the operation unit 16. If preview display is set, an image for preview display is generated inside the controller 7 based on the image data formed by the formatting process.

  Since the preview screen is generally small, reduction processing is performed for preview display. The image data generated for preview display is transferred to the operation unit 16. The operation unit 16 displays the transferred image data. Reduction processing is not essential. Since image data loss occurs due to the reduction, the reduction process is not performed when the image quality is important.

  If it is determined by checking the preview image that the image remains as it is, the image is delivered to the external PC terminal 13 by pressing a predetermined button. As a result of checking the preview image, when it is desired to start again from reading because the document is wrong, the reading is started again by pressing a predetermined button.

  When it is determined that a part of the preview image is to be corrected, the correction can be made by designating the correction part on the screen. The preview image display unit provided in the operation unit 16 has a touch panel format, and the image processing apparatus can recognize the position information of the pressure applied to the touch panel. For this reason, the image processing apparatus recognizes a location designated by the user for correction. Since the processing related to this correction is a core part of the present invention, it will be described later.

  Further, when the image processing apparatus operates as a printer that prints out from the external PC terminal 13 via the network 12, an image and a command for performing a print instruction are analyzed from data received via the NIC 15, and as image data, The bitmap is expanded into a printable state, the expanded data is compressed, and the data is accumulated. The accumulated data is written to the HDD 14 which is a large capacity storage device as needed. When the image data is stored, bibliographic information, which will be described later, is also written in the HDD 14.

  Next, the controller 7 sends the image data of the HDD 14 to the decompression processing unit 8 via the general-purpose bus 6. The decompression processing unit 8 decompresses the compressed image data to the original 8-bit data and sends it to the printer correction unit 9. In the printer correction unit 9, if the input is RGB, the color correction processing unit converts it into YMCBk data.

  Next, γ correction processing, halftone processing, and the like are performed independently for each YMCBk, and lightness and darkness correction processing and gradation number conversion processing of the plotter 10 are performed. In the tone number conversion process here, the image data is converted from 8 bits to 2 bits using error diffusion or dither processing. The plotter 10 is a transfer paper printing unit using a laser beam writing process. The plotter 10 draws 2-bit image data as a latent image on a photoconductor, and forms a copy image on the transfer paper after image formation / transfer processing with toner.

  In a digital image processing apparatus, generally, an original is read by a scanner, image data is converted into digital data, and an image area (image area) of the original is classified into areas having different characteristics (image area separation). Various image processing is performed on the image data according to the result of determining which region of the classified image region (image region) the target pixel belongs to. Thereby, the image quality of the output image can be greatly improved. The operation unit 16 sets a mode related to the user's operation and gives a start instruction. By performing the setting and instructing the start, the operation described above is started.

  Next, a high-compression PDF image compression apparatus according to an embodiment of the present invention will be described. FIG. 2 is a diagram for explaining the format processing of the high-compression PDF in the embodiment of the present invention. The high-compression PDF is composed of a character image, a foreground image, a background image, and a black character image.

  The character image is composed of binary values of characters / non-characters and indicates the shape and position of the characters. The foreground image has color information of pixels that are characters in the character image. The background image has information about a portion that is a non-character in the character image. The black character image indicates the shape and position of the black character of the processing target image.

  A portion described as an invalid area in the background image and the foreground image indicates an area that does not need to have information. For example, since the foreground image has the color of a pixel that is a character in a character image, the corresponding region in the background image is an invalid region. In the format process when generating a high-compression PDF, the above four images are generated, and each is individually compressed.

  Next, the color change of the foreground image will be described with reference to FIG. FIG. 3 is a diagram for explaining the color change in the embodiment of the present invention. Although not explicitly shown in FIG. 2, the foreground image is subjected to block compression after the resolution is lowered. When the resolution is lowered, if a character of a different color is included in the block, information is lost. As a result, the user may perceive the color change of the character.

  Specifically, in FIG. 3, as an image before the resolution is reduced, the resolution is reduced for an image composed of an image part A with only a red character string and an image part B on the boundary surface between the red character string and the blue character string. If the block compression is performed after that, even if the image portion A with only the red character string is reduced, the image with only the red character string is reproduced, but the image portion B at the boundary surface between the red character string and the blue character string is reproduced. If the reduction processing is performed, red and blue are mixed, and may be recognized as a color change by the user.

  Similarly, the color change of the foreground image will be described with reference to FIG. FIG. 4 is a diagram illustrating the color change in the embodiment of the present invention. In FIG. 4, when the processing target image is a character image composed of red and blue, the foreground image having the color information of the pixel that is the character of the character image is a portion represented by diagonal lines. When decompressed after reducing the resolution and then performing block compression, red and blue are mixed at the boundary between red information and blue information, and when combined with a character image and displayed, it is perceived as a color change by the user. End up.

  Next, a data flow between the controller 7 (FIG. 1) and the HDD 14 (FIG. 1) in the embodiment of the present invention will be described with reference to FIG. FIG. 5 is a block diagram showing a data flow in the embodiment of the present invention. In FIG. 5, an output image generating unit 51 and an output image assembling unit 52 are in the format process described above.

  The output image generation unit 51 generates a background image, a character image, a foreground image, and a black character image from the processing target image. It does not matter what algorithm is used, but here, separation processing based on the algorithm shown in FIG. 6 is used. FIG. 6 is a conceptual diagram of original image separation processing in the embodiment of the present invention.

  In FIG. 6, first, in step (hereinafter referred to as “S”) 10, each pixel of the acquired original image may be substituted with, for example, brightness (in the case where the image is expressed in RGB, the G component). The pixels brighter than the threshold are classified into the background (third attribute), and pixels darker than the threshold are classified into the foreground (first attribute).

  In S11, a pixel having a pixel value very close to black (for example, all pixels of R, G, B are all below a certain threshold) is selected from the pixels classified as the foreground (first attribute), and the pixel Are reclassified as black characters (second attribute).

  A first image generated based on each attribute, that is, based on the first attribute is, for example, A in FIG. 6, and a second image generated based on the second attribute is, for example, in C in FIG. 6. A third image generated based on the third attribute is, for example, B in FIG.

  In S12, the position of the pixel belonging to the generated first image (the pixel in which the pixel value is set in the first image) is referred to, and the pixel value of the pixel at the same position as the belonging pixel is turned ON (for example, 1) is generated, and selection data in which pixel values of pixels other than the pixel are set to OFF (for example, 0) is generated. The generated selection data is, for example, D in FIG.

  In S13, the pixel value of the invisible area of the first image is set, and the pixel value of the invisible area of the third image is set. The first image in which the pixel value of the non-visible region is set is, for example, E in FIG. 6 and, for example, F in FIG.

  As shown in FIG. 6, by separating an area (for example, black characters) relating to a specific color or lightness included in the original image from the original image as an independent image (layer), the specific color or lightness is obtained. Such single color or lightness (black or zero lightness) can be reproduced (output).

  Returning to FIG. 5, in the present invention, the selection data D shown in FIG. 6 is assumed to be a character image. The four images C, D, E, and F generated using the algorithm of FIG. 6 are once stored in the HDD 14 (FIG. 1).

  Next, the preview image generation unit 53 generates an image for preview. The four images stored in the HDD 14 (FIG. 1) are developed, reduction processing and enlargement processing are performed on each image as necessary, and a preview image is generated by combining them. There is no limitation on the reduction process or the enlargement process, but the correspondence between each pixel on the four generation images and each pixel on the generated preview image is retained.

  The output image correction unit 54 determines a correction pixel from the correction area designated by the user, and performs a process of editing the output image so as to correct the display color of the correction pixel. Correction is performed from the image data of the character image corresponding to the correction pixel, the image data of the block including the pixel on the foreground image corresponding to the correction pixel, the foreground image data near the block, and the accumulated processing target image data. A black character image for identifying the added foreground image and the corrected pixel is generated. The processing of the output image correction unit 54 is a core part of the present invention, and will be described later.

  The output image assembling unit 52 combines the foreground image output from the output image correcting unit 54 and the accumulated character image, black character image, and background image into one PDF file. If the user does not instruct correction, the image output by the output image generation unit 53 flows to the output image assembly unit 52 without passing through the output image correction unit 54, and a PDF file is generated. .

  Next, processing that is performed by the controller 7 (FIG. 1) immediately after displaying the preview image will be described with reference to FIG. FIG. 7 is a flowchart showing a flow of operation of the controller in the embodiment of the present invention. As described above, by a predetermined operation of the preview image, the user can select the resumption of the processing of the preview image, the redoing of the processing, and the partial correction. Here, a flow in the case where partial correction is selected is shown.

  First, in S71, a dissatisfied part of the user is specified. That is, on the preview screen, designation of where to modify is accepted. The user designates the range of the correction portion by a predetermined method such as tracing the touch panel with a pen. From the specified range, the image processing apparatus grasps where the correction range is on the image.

  Next, in S72, it is determined whether or not a character pixel is included in the designated range. That is, it is checked whether or not a character pixel is included in the correction range. A character pixel means a pixel that is a character in a character image. The present invention is intended to correct the color of the character pixel. Therefore, if the character pixel is not included in the correction range, the correction cannot be made, and a display to that effect is displayed (S72: N).

  Since the correspondence between the preview image and the character image is stored when the preview image is generated as described above, the character image corresponding to the specified correction range is read from the HDD 14 (FIG. 1) to be within the specified correction range. It can be checked whether or not a character pixel is included.

  If a character pixel is included (S72: Y), the corresponding foreground image block is also read from the HDD 14 (FIG. 1). Foreground image blocks are 8 × 8 units in the case of JPEG. Since the correspondence between the preview image and the foreground image is stored when the preview image is generated as described above, information can be acquired from these pieces of information.

  Next, in S73, image information to be processed is acquired. That is, information on the processing target image corresponding to the block of the foreground image including the character pixels within the designated range is acquired. An image for acquiring the image information to be processed is shown in FIG. FIG. 8 is a diagram illustrating an image for acquiring processing target image information corresponding to a block of a foreground image including character pixels within a designated range in the embodiment of the present invention.

  In FIG. 8, when an image range desired to be corrected is designated on the preview screen, the foreground image block including the designated range corresponds to the four rectangular blocks 1 to 4 which are areas of the corresponding processing target image. However, these blocks include a character pixel of “that”.

  Returning to FIG. 7, in S74, the character pixel to be corrected is determined. The difference between the processing target image and the foreground image is calculated for the pixels corresponding to the character pixels within the specified range. A pixel having a large difference is set as a correction pixel candidate within the specified range.

  Even a character within the correction range does not necessarily have the intention of correcting all the pixels in the character. Since the specification of the range is input by the user, there is a possibility that characters that do not need to be corrected are included. In addition, when it is desired to specify the color of the outline of a character included in the character string, it is natural for the user to specify an area so as to surround the character string. In such a case, the middle part of the character that does not need to be corrected is also included in the correction range. Therefore, it is necessary to determine the corrected character pixel based on the designated area. Actually, the reason why the color change of characters is worrisome is that there are many characters existing in the same block as the characters of different colors, so there are many outline portions of the characters.

In this case, the correction pixel candidate is used instead of the correction pixel for the following reason. That is,
Due to the nature of the present invention, there are some restrictions on the number of corrected character pixels in a block and the pixel position in the block. For this reason, not all pixels having a large difference between the processing target image and the foreground image can be corrected. Therefore, the character pixel to be finally corrected is determined in the flow described below.

  Here, restrictions on correction in the present invention will be described. In the present invention, the color of the correction pixel is a certain color of a neighboring block. As will be described later, the color of the correction pixel is replaced with a color other than the correction pixel in the same block, the block is compressed, and the color of the original correction pixel is encoded in the region of the high frequency component of the compressed block. Since the block includes pixels that are not correction pixels, it is necessary to know which one is the correction pixel in the block. The pixels that are characters in the character image are non-characters in the black character image, and therefore can be identified by changing the pixels of the black character image of the correction pixel to black characters. That is, when assembling the output image, the character image is a character and the black character image is a non-black character. Is possible.

  Since the black character image is binary data, it can basically have only binary information as to whether it is a correction pixel. The high-frequency component of the compressed block needs to have vector information for a block having a color that replaces the corrected pixel. In order to have corrected pixels to be replaced with different colors in the same block, it is necessary to have vector information for multiple blocks and the position of the corrected pixels on the block, which is not realistic from the viewpoint of file size. . Therefore, basically, a correction pixel to be replaced with one color is determined for one block.

  In addition, since the block including the replacement color is a small rectangular shape corresponding to several tens of pixels in the vertical and horizontal directions, it is not assumed that characters of greatly different colors are included in the block.

  Returning to FIG. 7, in S75, a similar color block is searched. That is, this is a step of determining a corrected character pixel from the corrected character pixel candidate. The corrected character pixel candidates are classified into clusters by the method described below, and determined as a block that replaces the color of the corrected character pixel of the cluster determined to be corrected. A flow for determining a block to be corrected from the corrected character pixel candidates will be described with reference to FIG.

  FIG. 9 is a flowchart showing a flow for determining a block to be corrected from the corrected character pixel candidates in the embodiment of the present invention. As described above, the color that replaces the pixels included in one block is basically one color. Therefore, the corrected pixel candidates are clustered and divided into one or more clusters, and the corrected pixel candidate in the cluster to be corrected most is set as a corrected pixel.

  In FIG. 9, in S91, the corrected pixel candidates are clustered. The clustering method does not matter and will not be described in detail here. Similar color groups can be grouped together.

  In S92, a cluster representative value is calculated. The average value of the foreground image data corresponding to the corrected pixel is calculated as the representative value of the corrected pixel candidate of the cluster.

  In S93, the representative value of the neighboring block is calculated. For each cluster, look for whether there is a color close to neighboring blocks. An average value of foreground image data corresponding to character pixels of neighboring blocks is calculated.

  In S94, a neighboring block is selected. The values calculated in S92 and S93 are compared, and the neighboring block with the smallest difference is selected. However, when the difference is larger than the predetermined threshold, it is assumed that there is no block having a color similar to the neighborhood and is not output. Here, blocks within one block are treated as neighboring blocks, but the present invention is not necessarily limited to this. However, it is not preferable to go to blocks far away because the amount of vector information to be stored later increases.

  In S95, it is determined whether or not there is a cluster. However, if there is no output in S94, it means that no correctable color was found (S95: N), and the process proceeds to S98.

  In S96, a cluster is selected. Among the clusters output from S94, the cluster with the largest number of correction pixel candidates is selected. The corrected character pixel candidate included in this cluster is set as a corrected character pixel.

  In S97, vector calculation is performed. A vector from the block including the corrected pixel to the cluster selected in S96 is calculated.

  In S98, continuous determination is performed. That is, as described in S94, since the neighboring block is one block, if there is a color to be replaced at a location two or more blocks away, it cannot be designated as a vector. Here, if there is a block having a corrected pixel in the upper or left neighboring block, it is determined whether or not replacement with a block further ahead of the vector of the block is possible. The schematic diagram is shown in FIG. FIG. 10 is a schematic diagram showing that the block to be processed in the embodiment of the present invention can be replaced with a further block. If the previous block can be replaced, the upper or left neighboring block is also considered to have the same vector to indicate that the upper or left neighboring block is similar.

  Here, the reason for the top or left is that since the raster processing is generally performed from the upper left to the lower right, correction pixels are already determined in the left and upper blocks. Needless to say, when raster processing is performed from the right or bottom, the raster processing is not limited to top or left.

  Returning to FIG. 9, in S99, it is determined whether or not vector detection is possible. That is, as a result of investigating whether or not the replacement is possible in S98, if the replacement is possible in either the upper or left (right or lower) direction (S99: Y), the compression code is calculated in and after S96. It will be. If the vector cannot be detected (S99: N), the processing for this block ends in S90. That is, in S90, there is no correctable character in the designated area, that is, an impossible display is displayed on the operation unit.

  Returning to FIG. 7, in S76, the compressed code is corrected. That is, the compression code of the foreground image of the block including the correction pixel is generated. FIG. 11 is a schematic diagram illustrating a state of pixels accumulated in the HDD according to the embodiment of the present invention. Since the foreground image is subjected to block unit compression, a compression code is generated for each block. In the foreground image stored in the HDD (FIG. 1) 14, as shown in the upper diagram of FIG. 11, the invalid pixel portion contains the average value of the other pixels in order to reduce the file size. .

  In S76 of FIG. 7, both the invalid pixel and the correction pixel are replaced with the average value of the character pixels other than the correction pixel (a in FIG. 11). By doing so, the inside of the block approaches uniformly, and the amount of compressed code corresponding to the block decreases as shown in FIG. Color information that replaces the correction pixel is put in the reduced portion. FIG. 12 is a diagram for explaining that the corrected pixel information is inserted into the portion where the compressed code is deleted in the embodiment of the present invention. The method of entering the corrected pixel information will be described later. The pixels in the correction pixel block are the average of the character pixels other than the correction pixels.

  At the same time, in order to distinguish the correction pixel of the character pixel in the block from the other pixels, the pixel of the black character image corresponding to the correction pixel is made a black character. This is because a pixel that is a character in a character pixel is not a black character in a black character image. If a pixel that is a character in a character pixel is usually a black character in a black character image, the pixel in the black character image corresponding to the correction pixel needs to be a non-black character.

  Next, how to insert color information for replacing the corrected pixels of the foreground image will be described. It is assumed that the foreground image is JPEG compressed. In the case of JPEG, an 8 × 8 block is Huffman-encoded as shown in FIG. FIG. 13 is a diagram showing a processing procedure of the basic method of JPEG compression in the embodiment of the present invention.

  As shown in FIG. 13, JPEG compression is performed after Huffman coding of quantized data after discrete cosine transform (DCT). The DC component and the AC component are separately Huffman encoded. For the DC component, the difference value from the previous block is encoded, and for the AC component, the code of the component read by the zigzag scan (FIG. 14) is Huffman encoded.

  In Huffman coding, a bit code string is determined according to a Huffman code table. In order to create bit code adjustment data, it is necessary to select a suitable bit code string from the Huffman code table.

  FIG. 14 is a view showing a zigzag scan table of JPEG images in the embodiment of the present invention. The AC component is based on the Huffman code table as shown in FIG. 15 based on the number of bits of the coefficient and the run length of 0 before the block to be encoded in the order read in the zigzag scan table shown in FIG. Encoded. FIG. 15 is a diagram showing an example of a Huffman code table of AC components in JPEG compression according to the embodiment of the present invention.

  For example, when the coefficient value of the value of the highest frequency component (lower right in FIG. 14) is 5 and the run length of 0 is 0, 5 is “101” in binary representation, so FIG. “100” described in the place of rw = 3, run = 0 is selected. As the code, “101” representing 5 is added to “100” selected based on the table, and finally the code for this coefficient becomes “100101”.

  By the way, by replacing the pixel corresponding to the correction pixel with the average value of the other character pixels, the compressed code of the block including the correction pixel should have a low high-frequency component value. Even if the value of the highest high-frequency component is increased or decreased, the reproducibility is higher than that of the original image. In the present invention, information is embedded in the high-frequency component region.

Since it is necessary to identify the four values of the top, bottom, left, and right from the code embedded in the high frequency component region, the highest high frequency component value is changed according to the rules shown in Table 1. For example, when the corrected character pixel of a block is replaced with the representative value of the upper block adjacent to the block, when the coefficient of the highest high-frequency component of the foreground image block is 5, the multiple of 4 closest to 5 is 4. The coefficient is 4 and compression is performed. In the case of 4, since the binary representation representing 4 is “100”, the compression code for this coefficient is “100100”.

  Note that the compressed code regenerated in this way does not necessarily have the same number of bits as the original compressed code. In other words, when the bit code at the edit location is replaced with a bit code corresponding to the color to be rewritten, the bit code amount changes. However, if the bit code corresponding to the color to be rewritten is a file compressed using entropy coding, the size is smaller than the code before rewriting, except for exceptions.

  This is because the color of the character to be rewritten is a fixed value for the color to be changed, and all the pixel values in the block are the value. In many cases, the value of data to be rewritten is basically a value read from a document. In the original to be read, even for pixels of the same color, the values vary somewhat depending on the reading characteristics. In entropy coding, the larger the entropy, the larger the code amount, so the bit code after replacement is smaller (FIG. 16).

  As a result of the replacement, the reduced number of bits is absorbed by increasing the bit code of the invalid area following the area. The invalid area has no problem because data corresponding to any color does not affect the image quality. As a result, only the area to be replaced and the subsequent invalid area data are affected. Subsequent data bit shifts are not required (FIG. 17). It should be noted that if the bit code increases after editing, it can be dealt with if the bit code in the invalid area can be reduced.

  Thus, in the Huffman code table, codes having the same number of bits can exist. By assigning a meaning to each code in advance and outputting a code corresponding to the editing content, information can be embedded without changing the file size. If there is no code having the same number of bits, a code having a larger number of bits than that and having a plurality of codes is output. Although the number of bits is larger than the number of bits that should be output, it can be regarded as an editing area and absorbed in the subsequent invalid area. In this way, it is possible to prevent bit shift movement loss.

  In general, since the compression code is reduced in each block of the foreground image, the file size of the foreground image does not increase. The black character image will increase, but since it is binary data, a slight increase is sufficient.

  In the embodiment described above, the corrected pixel could only be replaced with a single color. In many cases, there is still no problem, but in another embodiment of the present invention, a method for having a corrected pixel to be replaced with a different color in the same block will be described.

  The difference from the above-described embodiment is S96 in FIG. 9 and S76 in FIG. That is, in S96 of FIG. 9, the correction pixel is highly likely to have a color to be replaced with the closest block of the block including the correction pixel. Therefore, several possible patterns are prepared in advance, and it is confirmed whether or not all the correction pixel candidates in the block match the pattern.

  FIG. 18 is a diagram showing vectors to blocks having replacement colors in another embodiment of the present invention. As pattern 1, the correction pixel is replaced with the upper, lower, left, and right blocks closest to each other. As pattern 2, the right and left blocks with the closest correction pixels are replaced. As pattern 3, replacement is performed with the upper and lower blocks closest to the correction pixel. In the case where there are a plurality of closest blocks in pattern 1, it is determined in advance which pattern is used. If it matches the pattern, it outputs which pattern it matches.

Only the difference from the embodiment of the present invention is shown in S76 in FIG. Since it is necessary to output information on the pattern to the compressed code, the information on the pattern is added in addition to Table 1.

なお、ブロックの代表値は、図９におけるＳ９３で用いた計算方法で算出する。 Hereinafter, decoding of a file generated using the present invention will be described.
About the embodiment

The representative value of the block is calculated by the calculation method used in S93 in FIG.

Regarding other embodiments, since the difference from the decoding with respect to the embodiment is Table 4, the corresponding table is shown below.

  As described above, unlike the conventional technique, in the technique for expressing the color of a character as an image (foreground image), the block on the foreground image corresponding to the designated area of the character color change is displayed on the preview screen. By reducing the color contained in the block and performing compression, the amount of compressed code of the block is reduced, and the vector information of the color adjacent to the block close to the reduced color is the reduced color. As the information for identifying the pixel corresponding to the reduced color is stored at the same time as storing in the reduced compressed code portion, in the technology for expressing the color of the character as an image (foreground image), It is possible to improve the color change of characters on the preview screen.

  The present invention has been described above by the preferred embodiments of the present invention. While the invention has been described with reference to specific embodiments thereof, various modifications and changes can be made to these embodiments without departing from the broader spirit and scope of the invention as defined in the claims. is there.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 10 Plotter 11 Copy image 12 Network 13 External PC terminal 14 HDD
15 NIC
16 Operation unit 2 Scanner 3 Document 4 Scanner correction unit 5 Compression processing unit 6 General-purpose bus 7 Controller 8 Decompression processing unit 9 Printer correction unit

JP 2008-236204 A

Claims (6)

An image compression apparatus having compression means for dividing and compressing an image composed of a plurality of different colors into rectangular image regions,
First image area specifying means for specifying a rectangular image area having first color information from among a plurality of character images;
A rectangular image area having second color information is specified from among the plurality of character images adjacent to the rectangular image area having the first color information specified by the first image area specifying means. Second image area specifying means for performing,
The compression means includes
First compressed code generation means for generating a first compressed code;
Second compressed code generation means for generating a second compressed code;
A data amount that subtracts a data amount corresponding to the data amount of the second compression code from a data amount of the total compression code that is the sum of the data amount of the first compression code and the data amount of the second compression code Reduction measures,
Vector information generating means for generating vector information which is information for identifying the second color information based on the second color information;
Vector information storage means for storing the vector information generated by the vector information generation means in a portion of the data amount corresponding to the reduced data amount of the second compression code;
Compressed image information generating means for generating compressed image information in which the vector information is stored in the data of the first compressed code;
An image compression apparatus comprising:   Display means for displaying the compressed image information as a preview is provided, and the second image area specifying means designates an area for an image preview-displayed on the display means, thereby having a rectangular shape having the second color information. The image compression apparatus according to claim 1, wherein an image area is specified.   The rectangular image area having the second color information is specified from upper, lower, left, and right areas adjacent to the rectangular image area having the first color information. 2. The image compression apparatus according to 2.   A plurality of rectangular image areas each having a plurality of different color information, and a plurality of vector information, which is information for identifying the plurality of color information of the plurality of image areas, are prepared in advance as an image area information group. 4. The rectangular image area having the second color information is specified from the image area information group when the area is designated. Image compression device. An image compression method in an image compression device having compression means for dividing and compressing an image composed of a plurality of different colors into rectangular image regions,
Identifying a rectangular image area having first color information from among a plurality of character images by a first image area specifying means;
Second color information from among the plurality of character images adjacent to the rectangular image area having the first color information specified by the first image area specifying means by the second image area specifying means. Identifying a rectangular image area having
The compression means includes
Generating a first compressed code by first compressed code generating means;
Generating a second compressed code by a second compressed code generating means;
The data amount reduction means corresponds to the data amount of the second compression code from the data amount of the total compression code, which is the sum of the data amount of the first compression code and the data amount of the second compression code. Reducing the amount of data,
A step of generating vector information, which is information for identifying the second color information, based on the second color information by a vector information generation unit;
Storing the vector information generated by the vector information generating unit in a portion of the data amount corresponding to the reduced data amount of the second compressed code by the vector information storing unit;
Generating compressed image information in which the vector information is stored in the data of the first compressed code by compressed image information generating means;
An image compression method comprising: A program to be executed by a computer of an image compression apparatus having compression means for compressing an image composed of a plurality of different colors by dividing the image into rectangular image regions,
A process of specifying a rectangular image area having first color information from among a plurality of character images by a first image area specifying means;
Second color information from among the plurality of character images adjacent to the rectangular image area having the first color information specified by the first image area specifying means by the second image area specifying means. Identifying a rectangular image area having
The compression means includes
Processing for generating the first compressed code by the first compressed code generating means;
A process of generating a second compressed code by the second compressed code generating means;
The data amount reduction means corresponds to the data amount of the second compression code from the data amount of the total compression code, which is the sum of the data amount of the first compression code and the data amount of the second compression code. Processing to reduce the amount of data,
Processing for generating vector information, which is information for identifying the second color information, based on the second color information by the vector information generating means;
Processing for storing the vector information generated by the vector information generating means in a portion of the data amount corresponding to the reduced data amount of the second compressed code by the vector information storing means;
Processing for generating compressed image information in which the vector information is stored in the data of the first compressed code by compressed image information generating means;
A program characterized by comprising:

Images