JP2012015644A - Image processor and image processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To highly efficiently compress an image while minimizing image deterioration due to color reduction, when the image is compressed while suppressing a processing cost, especially, when image compression is repetitively performed.SOLUTION: An information processor divides image data by each block and performs compression processing multiple times with respect to the image data by sequentially defining the divided blocks as processing objects. The information processor determines whether or not to perform subtractive color processing for reducing the number of colors of the blocks based on subtractive color information. When it is determined to perform the subtractive color processing, the subtractive color processing is performed with respect to the blocks to be processed. When the number of colors is reduced as the result of the subtractive color processing, updating that the subtractive color processing has been performed is performed in the subtractive color information.

Description

  The present invention relates to an image processing apparatus and an image processing method. More specifically, the present invention relates to a technique for compressing image data in predetermined block units and a technique for performing image processing and decompression processing on image data compressed in block units.

  2. Description of the Related Art Conventionally, digital multifunction peripherals handle image data having a resolution of 1200 dpi or more with increasing demand for high-resolution color image data. In response to this, image processing apparatuses such as digital cameras and facsimile machines, not limited to digital multi-function peripherals, handle image data by compressing it in order to save memory and hard disk capacity and reduce writing time. As a result, low cost and high speed are realized.

  As an encoding method used for still image data compression processing, a method using discrete cosine transform (JPEG method) and a method using Wavelet transform are known. This type of encoding method encodes image data in units of a predetermined block (for example, 8 × 8 pixels or 16 × 16 pixels), and performs discrete cosine transform, quantization, and entropy encoding, thereby achieving high compression efficiency. Has achieved. In addition, this type of encoding method is called a variable-length encoding method, and has a feature that the code amount changes for each image data to be encoded.

  The problem of the compression processing described above is that random access for referencing a small area is difficult because a variable-length encoding method in units of blocks is used. That is, since the position (memory address) of the block to be accessed is indefinite, some method of knowing the position is necessary, and in addition, an expansion process in units of blocks is necessary for actual access. Another problem with the above-described compression processing is that the processing in units of blocks is complicated. For example, the discrete cosine transform in the JPEG method normally requires 8 × 8 pixels as a block size. In order to perform the discrete cosine transform processing in units of blocks at a high speed, a high-speed arithmetic unit, dedicated hardware, Buffer memory is required. As an approach to the problem of processing cost related to such compression processing, an image compression technique (for example, Patent Document 1) with a fixed compression ratio with a small block size is disclosed.

  A further problem with the above-described compression processing is that the number of pixels that require image processing increases dramatically as the resolution increases, and the processing load increases. For example, when the resolution is doubled from 600 dpi to 1200 dpi, the number of pixels to be processed is quadrupled. Here, in the case of using the above-described encoding method using the JPEG method or Wavelet conversion, in order to refer to or convert the image data, it is necessary to decompress the compressed image data as described above. Become. That is, since the image processing cannot be performed with the compressed image data as it is, the processing time is increased as a result of performing the image processing for every pixel in all the pixels in the high resolution image data.

  As an approach aimed at image processing on image data compressed without performing decompression processing, there is a known run-length compression method that stores pixel data and the number of consecutive data. In addition, an image compression technique (for example, Patent Document 2) is disclosed in which an edge is detected in units of blocks and compression is performed by storing two colors of the edge.

JP 2004-104621 A JP-A-10-257488

  As described above, in the case of a JPEG method using discrete cosine transform or a method using Wavelet transform, there is a problem of processing cost because the block size as a processing unit is relatively large. That is, there is a problem that a large amount of calculation is required for each block, and dedicated hardware for realizing processing becomes expensive.

  Here, even when an image compression technique with a small block size and a fixed compression rate as described in Patent Document 1 is used, when image processing is performed on compressed image data, the image data must be decompressed. Processing is required. Therefore, there is a problem in that an enormous processing time is required depending on the number of pixels to be processed. This is a problem that, for example, in order to create low resolution data from high resolution data compressed by the JPEG method or the like, a decompression process is always required after the decompression process.

  There is also a problem in a compression method that can perform image processing on compressed image data without performing decompression processing. That is, in the compression method (for example, the above-described run length compression method) that does not have pixel values of image data but retains only matching information with adjacent pixels and performs image compression, the number of colors is reduced (color reduction) and is irreversible. In the case of compression, there is a problem of image quality degradation. This is because, for example, in the above-described run length compression method, when colors are similar and run with the same color is counted and encoded, the color difference from the original image data is propagated to a plurality of pixels. This is a problem that causes image degradation. Note that this image degradation problem is not limited to the run-length compression method, and is a common problem when combining compression based on matching information with adjacent pixels and color reduction processing, and is also applicable to the technique of Patent Document 2.

  In order to solve the above-described problems, in the present invention, an image processing apparatus that realizes highly efficient image compression while minimizing image degradation due to color reduction when image compression is repeatedly performed on image data. An object is to provide an image processing method.

  An image processing apparatus according to the present invention is an image processing apparatus that divides image data into blocks, sequentially processes the divided blocks, and performs compression processing on the image data a plurality of times. A determination unit that determines whether or not to perform a color reduction process for reducing the number of colors based on color reduction information; and when the determination unit determines that the color reduction process is to be executed, the color reduction process is performed on the processing target block. Color data included in the block by comparing color data of each pixel of the block to be processed with a color reduction processing unit that executes processing, an update unit that updates the color reduction information according to a result of the color reduction processing Specifying means for specifying a pattern flag corresponding to the arrangement pattern of the pattern, the pattern flag specified by the specifying means, and color data for the number of colors included in the block Output data as compressed output data, and until the predetermined resolution is reached, the set data of the color data of the first color of the block is further divided into blocks, and again as a processing target block to the determination means Input means for inputting.

  According to the present invention, when the divided blocks are sequentially processed and the image data is subjected to compression processing a plurality of times, the image quality deterioration due to the color reduction processing is minimized while suppressing the processing cost. The compression rate of image data can be maximized.

  Further, if the image data compressed according to the present invention is used, the processing cost can be suppressed, so that the image processing and the expansion processing can be easily performed.

It is a block diagram which shows the structural example of an image processing apparatus. It is a block diagram which shows the structural example of a controller. It is a figure which shows the pattern of the block at the time of dividing | segmenting raster image data. It is a figure which shows the relationship between the identifier of a block and its pattern (pattern flag). It is a figure which shows the whole flow of the compression process in 1st Embodiment. It is a figure which shows the detailed flow of the compression process in 1st Embodiment. It is a figure which shows a mode that the pattern of a block is converted into the pattern flag. It is a figure which shows the relationship between the input data and output data in a compression process. It is a figure which shows the detailed flow of the color reduction compression process in 1st Embodiment. It is a figure which shows the mapping on the memory space of image data. It is a figure which shows the whole flow of the compression process in 2nd Embodiment. It is a figure which shows the relationship between a page, a tile, and a block. It is a figure which shows the data structure of the packed data (packet). It is a figure which shows the structural example of a packet management table. It is a figure which shows the address of each packet mapped on the memory space. It is a figure which shows the data structure of the packet at the time of performing a compression process in multiple steps. It is a figure which shows the detailed flow of the color reduction compression process in 3rd Embodiment. It is a figure which shows the specific example at the time of applying the multiple times compression process and the color reduction compression process in 3rd Embodiment.

  Hereinafter, embodiments for carrying out the invention will be described in detail with reference to the drawings. In each embodiment described below, a digital multi-function peripheral (MFP) having a plurality of functions such as scanning, printing, and copying will be described as an example of an image processing apparatus.

[First Embodiment]
FIG. 1 is a block diagram illustrating a configuration example of an image processing apparatus 100 according to the first embodiment. As illustrated in FIG. 1, the image processing apparatus 100 includes a controller 101, a scanner 102, a printer 103, a network 104, and an HDD storage unit 105. The controller 101 is connected to the scanner 102, printer 103, network 104, and HDD storage unit 105, and is a control unit that controls the entire image processing apparatus 100. The scanner 102 is a processing unit that inputs image data as an image input device. The printer 103 is a processing unit that outputs image data as an image output device. The network 104 includes a LAN, a public line (WAN), and the like, and is a communication unit that inputs and outputs image data and device information. The HDD storage unit 105 includes a hard disk drive and stores system software, programs, image data, and the like.

  FIG. 2 is a block diagram illustrating a configuration example of the controller 101 illustrated in FIG. Detailed processing of each unit of the controller 101 will be described with reference to FIG.

  First, a case where image data scanned by the scanner 102 is read will be described. The scanner image processing unit 211 receives RGB (red, green, blue) image data read by the scanner 102 and performs image processing such as shading processing and filter processing on the image data. The image compression unit 212 performs compression processing on the image data processed by the scanner image processing unit 211. Thereafter, a DMAC (direct memory access controller) 213 transfers the compressed data obtained by the compression processing by the image compression unit 212 via the image memory bus 203 and stores it in the memory 202.

  Next, a case where the scanned image data is printed will be described. The DMAC 261 transfers the compressed data stored in the memory 202 to the color processing unit 262 via the image memory bus 203. The color processing unit 262 converts the compressed data transferred by the DMAC 261 into a CMYK (cyan, magenta, yellow, black) color space. Thereafter, the color processing unit 262 further performs color processing such as density adjustment and printer gamma correction on each value of CMYK. The DMAC 261 stores the color-processed data in the memory 202 again via the image memory bus 203. Thereafter, the DMAC 221 reads compressed data stored in the memory 202 via the image memory bus 203 in order to perform image processing for printing. The image expansion unit 222 expands the color processed data into raster image data. The print image processing unit 223 performs area gradation processing on the CMYK raster image data using a dither method or an error diffusion method, and outputs the processed image to the printer 103.

  Next, a case where the scanned image data is transmitted to the network will be described. The DMAC 261 transfers the compressed data stored in the memory 202 to the color processing unit 262 via the image memory bus 203. Then, the color processing unit 262 performs display gamma adjustment, paper background color adjustment, and the like, and then converts to a YCbCr (luminance, BLUE color difference, RED color difference) color space. Thereafter, the DMAC 261 stores the data processed by the color processing unit 262 in the memory 202 via the image memory bus 203. Then, the DMAC 231 transfers the compressed data stored in the memory 202 to the image expansion unit 232 via the image memory bus 203 in order to perform image processing for transmission. Thereafter, the image expansion unit 232 expands the compressed data into raster image data. The transmission / reception processing unit 233 performs JPEG compression processing on YCbCr raster image data in the case of color image transmission, and performs JBIG compression processing after binarizing the Y data in the case of monochrome binary image transmission. Output to the network 104.

  Next, a case where scanned image data is stored will be described. The DMAC 241 transfers the compressed data stored in the memory 202 to the spool image compression / decompression unit 242 via the image memory bus 203. The spool image compression / decompression unit 242 performs JPEG compression with higher compression because the writing speed of the HDD is slower than the writing speed of the memory. Thereafter, the HDD controller 243 stores the compressed data in the HDD storage unit 105. Note that when the compressed data stored in the HDD storage unit 105 is transferred to the memory 202 again, the above-described processing may be performed in reverse.

  Next, a case where PDL data sent from another device connected via the network 104 shown in FIG. 1 is written to the memory 202 will be described. Although the PDL interpretation unit is not shown in FIG. 2, the CPU 201 functioning as the PDL interpretation unit interprets the PDL data and outputs the display list as a result to the memory 202. The rendering unit 251 renders the display list stored in the memory 202 as RGB raster image data, and the image compression unit 252 compresses the rendering result. Thereafter, the DMAC 253 stores the compressed data in the memory 202 via the image memory bus 203.

  Note that the process of printing, transmitting to a network, and storing the PDL data can be realized by performing the same process as that of the scanned image data.

  In the present embodiment, the raster image data generated from the PDL data is compressed by the image compression unit 252 or the raster image data obtained by scanning is compressed by the image compression unit 212. The form is not limited to such a configuration. For example, instead of separately providing the image compression units 212 and 252 as shown in FIG. 2, a common image compression unit may be provided.

  Next, raster image data compression processing, which is a feature of this embodiment, will be described in detail. In this embodiment, first, raster image data in units of pages is divided into blocks of 2 × 2 pixels, and image data compression processing is performed using the blocks extracted by division as processing units.

  FIG. 3 is a diagram showing a block pattern when the raster image data is divided. Before describing the compression processing, the number of combinations in accordance with the number of colors occupied in 2 × 2 4-pixel data will be described with reference to FIG. As shown in FIG. 3, since the number of 2 × 2 pixels is 4, the number of colors occupied there is a maximum of 4 colors, and there are only combinations of 1 to 4 colors at most in the block.

  First, when there is one color in the block, all four pixels are composed of the same color, so there are only one combination.

  Next, consider the case where the block has two colors. When two colors are laid out in four pixels, if the upper left pixel is considered as the first color, the first color or the second color enters the remaining three pixels other than the upper left. Seven combinations are possible except for.

  Next, consider the case where the block has three colors. The number of three colors laid out in four pixels can be rephrased as the number of only one of the three colors used twice. Of the four pixel coordinates, two pixels have the same color. What is necessary is just to obtain | require the number of cases. That is, the number in the case of three colors is a combination of taking two coordinates from four coordinates, and there are six ways in total.

  Finally, when there are four colors in the block, there is only one pattern as in the case of one color.

  As described above, when the numbers in all cases of one color to four colors are totaled, 15 patterns can be considered in total. Considering that a flag (identifier) is added to identify all these patterns, 4 bits are required as the data amount of the flag for identifying 0 to E (hexadecimal number). The relationship between the 15 block patterns and the flags is defined as shown in FIG. 4, and this flag is hereinafter referred to as a “pattern flag”.

  FIG. 5 is a flowchart illustrating the entire compression process according to the first embodiment.

  First, the raster image data shown in FIG. 3 is input in units of pages (S501). Then, one page background pixel is set for each page (S502). This is pixel data to be used as an initial storage block at the time of compression processing, and normally white (255 for 8-bit RGB color images, 0 for CMYK images) is used. Here, the reason why white is used as the background pixel is that if the background pixel is set in advance, pixel data having the same color as the background pixel may not be output. In a normal document image, the background pixel is white. This is because it is most likely. Next, a “color reduction flag”, which is color reduction information for holding the application history of the color reduction processing for the block of 2 × 2 pixels shown in FIG. 3, is initialized to 0 (S503). Next, the color reduction flag is referred to in the first page compression process (S504). At this time, if the predetermined upper limit of the number of color reductions is 1 with respect to 0 in which the color reduction flag referred to in S504 indicates that the color reduction processing is not applied, it is determined that the color reduction is performed because the number of color reductions has not reached the upper limit (S505). Yes), the subtractive color compression process is executed (S506). On the other hand, if the predetermined upper limit of the number of color reductions is set to 0 (color reduction processing is prohibited) with respect to 0 indicating that the color reduction is not applied, the color reduction flag referred to in S504, the color reduction is not performed because the number of color reductions has reached the upper limit. (No in S505) and the compression process is executed (S507). A detailed description of the color reduction compression processing in S506 will be described later with reference to FIG. A detailed description of the compression processing in S505 will be described later with reference to FIG. Here, after the first page compression process is completed, it is determined whether to recompress, and if it is determined not to recompress (No in S508), the page compression process ends. On the other hand, if it is determined to recompress (Yes in S508), the process proceeds to S504, and the second page compression process is started. In the second page compression process, the color reduction flag is referred again (S504). At this time, if the color reduction flag referred to in step S504 indicates that the color reduction processing has been applied, and if the upper limit of the number of color reductions determined is 1, the color reduction number has reached the upper limit and it is determined that no color reduction will occur ( No in S505), the compression process is executed (S507). In accordance with the processing flow described above, the page compression processing is repeated until the input page image data has a desired resolution (S508). For example, when the number of compressions is set to two stages, the resolution of the page image data is converted from 1200 dpi to 600 dpi in the first compression process, and from 600 dpi to 300 dpi in the second compression process. . In the example of FIG. 5, it is determined whether the color reduction process is performed for each page. In other words, it is determined whether the color reduction process is performed on the processing target block in the page. I can say that.

  FIG. 6 is a flowchart for explaining in detail the compression processing of S505 executed by the image compression units 212 and 252. Based on the possible combinations of the 2 × 2 pixels shown in FIG. 3 and FIG. 4, processing executed by the image compression units 212 and 252 will be described using FIG. 6. As an input, for example, each of RGB has 256 gradations of 8 bits, and the data is described as an image of 24 bits per pixel in a dot-sequential manner of 8-bit data.

  First, a 2 × 2 pixel block is input (S601), and 24 bits are compared with all combinations of two pixels in the block (S602). As a result of the comparison, 1 is output when all bits match, and 0 is output when they do not match.

  FIG. 7 is a diagram for supplementarily explaining the flowchart of FIG. 6. As shown in FIG. 7, the coordinates are 1, 2, 3, and 4 in the order of the upper left, upper right, lower left, and lower right pixel positions in the 2 × 2 pixels. Since there are six coordinates of combinations of any two pixels in one block, including 1-2, 1-3, 1-4, 2-3, 2-4, and 3-4, 6 comparisons are made in S602. The comparison result is output as 6 bits. As shown in the comparison result of FIG. 7, if all the pixels in one block are the same color, all 1s are output as the comparison result. Conversely, if all the four pixels have different pixel values, the comparison results are all 0 is output.

  In the example using this 2 × 2 pixel block, there are 15 patterns that can appear from color matching in 4 pixels, and therefore, as shown in FIG. A flag is specified (S603). Next, the number of colors appearing in the four pixels and the color data are extracted (S604). As shown in FIG. 7, since each pattern indicating the arrangement of each color in a block is associated with a 4-bit pattern flag (that is, a 6-bit comparison result), the number of colors and colors in each block Data can be identified. In FIG. 4 of the present embodiment, it is defined that the color (pixel value) of the upper left pixel is the first color (first color data) in all patterns. Accordingly, when the pattern flag is 0, the color (pixel value) of the upper left pixel is extracted as the first color with the number of colors of 1. If the pattern flag is 1 to 7, the number of colors is 2 and the color (pixel value) of the upper left pixel is extracted as the first color, and further, the second color defined according to each pattern flag ( The color (pixel value) of the pixel at the position where the (second color data) exists is extracted. For example, when the pattern flag is 1, the color of the upper right pixel is extracted as the second color. When the pattern flag is 8 to D, the number of colors is 3 and the color (pixel value) of the upper left pixel is extracted as the first color, and further, the second color defined according to each pattern flag and The color (pixel value) of the pixel at the position where the third color (second and third color data) exists is extracted. For example, when the pattern flag is 8, the color (pixel value) of the upper right pixel is extracted as the second color, and the color (pixel value) of the lower right pixel is extracted as the third color. When the pattern flag is E, the color of the upper left pixel is the first color, the color of the upper right pixel is the second color, the color of the lower left pixel is the third color, and the color of the lower right pixel is the fourth color. Extract as That is, based on the pattern flag (or comparison result), the number of colors of the block is specified (S605, S607, S609), and the pattern flag and color data corresponding to each are output (S606, S608, S610, S611). .

  FIG. 8 is a diagram for supplementarily explaining the flowchart of FIG. 6. The data output here will be described with reference to FIG. As shown in FIG. 8, for example, when the pattern flag is 0 (that is, the inside of the four pixels is composed of one color) (YES in S605), the second and subsequent colors do not exist. The pixel value of the first color (24-bit color data) is output (S606). If the pattern flag is 1 to 7 (that is, two colors) (YES in S607), the coordinates of the pixel where the second color exists are calculated from the pattern flag, and the four bits of the pattern flag and the pixels for two colors A value (color data for 48 bits) is output (S608). The output data is configured in the order of the pixel value of the first color and the pixel value of the second color following the pattern flag. If the pattern flag is 8 to D (that is, three colors) (YES in S609), 4 bits of the pattern flag and pixel values for three colors (72-bit color data) are output (S610). If the pattern flag is E (that is, 4 colors) (NO in S609), 4 bits of the pattern flag and pixel values for 4 colors (96-bit color data) are output (S611). In other words, the color data output from each block did not appear in the block when scanned in the order of the coordinates in the block (upper left, upper right, lower left, lower right in the order 1, 2, 3, 4). This corresponds to output of color data in order from the first color.

  Thus, relatively simple processing is possible by outputting the input data of 4 colors (96 bits) in the block of 2 × 2 pixels to the 4-bit pattern flag and the pixel values corresponding to the number of colors existing there. This makes it possible to reduce the amount of output data. In particular, in the case of a raster image having a large number of blocks having the same color pixel in a 2 × 2 pixel block (that is, a block having a small number of colors in each block), the compression rate of the output data amount is also increased. Also, by referring to the pattern flag, the number of colors in the block can be specified. By performing such processing for all image blocks, the entire image data can be compressed.

  In the processing flow of FIG. 6, finally, it is confirmed whether or not the next block of 2 × 2 pixels exists in the page-unit raster image data shown in FIG. 3. Here, if it exists (Yes in S612), the conversion from the input data to the output data shown in FIG. 8 is repeated for the next block (S601 to S611), and if it does not exist (No in S612), Ends the compression process for each page.

  Next, raster image data color reduction compression processing, which is a feature of this embodiment, will be described in detail. In subtractive color compression processing, if there are only a few levels of pixels in a block, it is regarded as the same color, and by performing subtractive color processing in each block, the number of blocks judged to be four colors is reduced, and the compression rate is improved. .

  FIG. 9 is a flowchart for explaining in detail the color-reduction compression processing of S506 executed by the image compression units 212 and 252. Using FIG. 9, after performing the color reduction processing for each block, the processing for image compression will be described based on the possible combinations of 2 × 2 pixels as described above.

  First, a 2 × 2 pixel block to be processed is input (S901), and a color reduction process is performed from a pixel value of 4 pixels to one color (S902). Here, for example, a color reduction to one color is realized by calculating an average pixel value of four pixels. That is, the difference between the pixel value after the color reduction and the pixel values of the four input pixels is calculated (S903), and the magnitude of the error is determined (S904). For example, the absolute value of the difference between the input pixel value of each pixel and the RGB value after color reduction is taken, and if the value is less than or equal to the threshold value, it can be determined that the difference is small. If the difference is small, the pixels in the block can be regarded as the same color because there are only a few levels of difference. Therefore, if it is determined that the difference is small (Yes in S904), it is determined that the color is reduced to one color, and a color reduction flag indicating that the color reduction processing is applied is generated (S905). Thereafter, the pattern flag is specified as 0 (S906), and the pixel value and pattern flag of the reduced color are output (S907).

  If it is determined that the error is large (No in S904), then color reduction processing is performed to two colors (S908). Here, for example, two pixels A and B having the largest RGB value difference among the four pixels are extracted, clustered according to whether the remaining two pixels are closer to A or B, and an average value is obtained in each cluster. Two colors. Then, the difference between the two-colored pixel value and the input four pixel values is calculated (S909), and the magnitude of the error is determined (S910). That is, the difference between the average value of each cluster and each pixel used for the clustering is calculated to calculate the sum of the absolute values. If the value is equal to or smaller than the threshold value, it is determined that the difference is small. If it is determined that the difference is small (Yes in S910), it is determined that the color is reduced to two colors, and a color reduction flag indicating that the color reduction processing is applied is generated (S911). Thereafter, a pattern flag is specified according to the position of the pixel value when the color is reduced to two colors (S912), and the pixel value of the two colors and the pattern flag are output (S913).

If it is determined that the error is also large here (No in S910), then a color reduction process to three colors is performed (S914). Here, for example, two pixels having the smallest difference in RGB values are extracted from the four pixels, an average value of the values of the two pixels is obtained, and the colors are reduced to three colors together with the pixel values of the other two pixels. Then, the difference between the pixel values of the three colors and the pixel values of the input four pixels is calculated (S915), and the magnitude of the error is determined (S916). Since two pixels other than the two pixels used for the average value have the same pixel value as the input pixel, an error is caused by the difference between the average value of the two pixels having a small RGB value difference and the respective pixel values of the two pixels. Determine the size of.
If it is determined that the difference is small (Yes in S916), it is determined that the color is reduced to three colors, and a color reduction flag indicating that the color reduction processing is applied is generated (S917). After that, the pattern flag is specified according to the position of the pixel value when the color is reduced to three colors (S918), and the pixel value and the pattern flag of the reduced color are output (S919).

  If it is determined that the error is large again (No in S916), it is determined that the block is visually problematic when the color is reduced, and the pattern flag is specified as E without generating a color reduction flag (S920). The pixel values and pattern flags of all four pixels are output (S921).

  In the processing flow of FIG. 9, finally, it is confirmed whether or not the next block of 2 × 2 pixels exists in the page-unit raster image data shown in FIG. Here, if it exists (Yes in S922), the conversion from the input data to the output data shown in FIG. 8 is repeated (S901 to S921). If it does not exist (No in S922), the color-reduction compression processing in units of pages. Exit. As described above, in the process of FIG. 9, a color reduction flag is generated when color reduction processing is performed in any block in image data in a page unit with a certain resolution.

  Next, a process (memory storing process) in which the DMACs 213 and 253 write the pattern flag and color data obtained in this way into the memory is performed. At this time, the DMAC changes the writing positions of the pattern flag, the first color data, and the second color, third color, and fourth color data. The DMAC designates three addresses: a memory start address for writing a pattern flag, a memory start address for writing first color data, and a memory start address for writing data for the second and subsequent colors. That is, the pattern flags of each block are stored together in a pattern flag storage unit (memory area for storing pattern flags) on the memory. Also, the color data (color data set data) of the first color (first color) of each block is stored in a first color storage unit (memory area for storing the first color data of each block) on the memory. They are stored together. Further, the color data of the second to fourth colors of each block are collectively stored in the second, third and fourth color storage units (memory areas for storing the second to fourth color data of each block) on the memory. Will be. In addition to the above, as a color reduction information associated with the data, a color reduction flag is separately stored in a memory space or as a hardware register.

  FIG. 10 is a diagram showing an example of writing image data in the memory space by the DMAC. Note that when an 8-bit RGB image having a size of M × N pixels is input to the compression unit, the data size of the pattern flag storage unit in which pattern flag data is stored is (M × N × 4/8) / 4 bytes. Further, the data size of the first color storage unit in which the first color data is stored is (M × N × 24/8) / 4 bytes. The reason for dividing by 4 is that the pattern flag and the first color storage unit need only be prepared for each block of 2 × 2 pixels. The data sizes of the second, third, and fourth color storage units that store the second, third, and fourth color data differ depending on the raster image data to be processed. This is because the number of blocks in which the second, third, and fourth colors exist differs depending on the image.

  Here, as shown in FIG. 10, regarding the memory area (first color storage unit, and second, third, fourth, and fourth color storage units) after the first color write start address, the number of bits in which the color data of the pixel remains as it is. It is possible to perform image processing on the compressed data by using the stored data. That is, unlike compressed data such as JPEG, the color (pixel value) of each pixel can be specified without decoding the compressed data and returning it to raster image data. Therefore, when color processing that is completed with one pixel input and one pixel output, for example, color conversion using LUT, gamma correction processing, color space conversion processing using matrix operation, etc., is performed without returning to the original raster image data. The image data stored as shown in FIG. 8 can be directly processed. When the color processing unit 262 shown in FIG. 2 performs image processing in units of pixels, pixel data after the first color write start address on the memory 202 is read via the DMAC 261, and processing in units of pixels is completed. Write back to the memory 202. At this time, if the number of bits of the pixel does not change due to some pixel unit processing, the memory 202 can be saved by overwriting the same location on the memory 202. As described above, by directly using the compressed data, the transfer efficiency on the memory bus is improved, and processing is performed on a smaller number of color data than the number of pixels of the original raster image data. Is possible.

  Further, as shown in FIG. 10, the image data is divided into a pattern flag, a first color, and other colors and is stored as a first color by using the fact that they are discretely stored in the memory. Low resolution image data can be used as it is. This can be used, for example, when realizing functions such as preview display of PDL image data and scan image data stored in the MFP, network transmission, and the like. Specifically, the first color storage unit stores the pixel value (color data) in the memory when the raster image data is divided into blocks of 2 × 2 pixels and the pixel at the upper left coordinate of each block is sampled. Will be present continuously. Therefore, for example, even if the print resolution is 600 dpi, 300 dpi image data can be used as it is when previewing or transmitting at 300 dpi or less is sufficient. As described above, when it is necessary to obtain reduced raster image data, only the color data of the first color stored in the first color storage unit is extracted at a time, so that half the size can be easily obtained. Raster image data can be obtained.

  Furthermore, as shown in FIG. 10, the image data is divided into pattern flags, first colors, and other colors, and is stored discretely on the memory. It is possible to generate image data having a resolution (image size) of. As an example, a description will be given of reduced transmission when 600 dpi raster image data is converted into data as shown in FIG. 8 and stored. When 300 dpi is specified as the resolution of the image to be transmitted, the data stored in the first color storage unit may be extracted as it is and transmitted. If a resolution greater than 300 dpi sampled in the first color storage unit, such as 400 dpi, is designated, the following processing is performed. That is, based on the data stored in the pattern flag storage unit, the first color storage unit, and the second, third, and fourth color storage units, the compressed data is expanded once and scaled using a known scaling process. Then, the scaled image is transmitted. If transmission resolution less than 300 dpi is specified, scaling processing is performed to the specified resolution using only the data in the first color storage unit. In this way, data can be read while switching according to the required image size.

  In this way, if an original half-size image can be obtained, it is also possible to perform compression again using the image as an input. Further, the higher the resolution of the input image data, the more redundant the sampled reduced image remains, and the compression process can be performed in multiple stages, and the compression rate at that time can be expected sufficiently.

  Next, the image expansion units 222 and 232 that are paired with the image compression units 212 and 252 will be described. Note that the image decompression units 222 and 232 perform processing for returning the pattern flag and color data as described above to raster image data. The compressed data pattern flag write start address, first color write start address, and second, third and fourth color write start addresses of the compressed data arranged on the memory 202 as shown in FIG. specify. The DMACs 221 and 231 read data from three addresses and transfer them to the image decompression units 222 and 232.

  The image expansion units 222 and 232 first interpret the 4-bit pattern flag and calculate the number of colors in the block. In addition to the first color data, the second, third, and fourth color data are read according to the number of colors, and the first color is determined according to the color data arrangement pattern defined in advance for each pattern flag shown in FIG. In addition, the color data of the second to fourth colors are rearranged. In this way, the 2 × 2 pixel block is expanded and decoded.

  In addition, when the image size is reduced to 1/2 by the image expansion units 222 and 232, the pattern flag and the 2, 3, and 4 color data are not required as described above. Specify only the write start address. As a result, only the first color data is read from the memory 202 to form an image. By processing in this way, it becomes possible to save the bandwidth of the memory bus.

  As described above, according to the first embodiment, not only can a memory capacity and a memory bus band be saved by a relatively simple compression method, but also image processing and reduction scaling are performed on a pixel basis. In this case, the processing load can be reduced.

  In the first embodiment, the block pattern indicating the arrangement of the color data and the pattern flag are associated as shown in FIG. 7, but the present invention is not limited to this. For example, a pattern indicating the arrangement of color data and a pattern flag may be defined in advance so that the pixel value of the lower right pixel of the 2 × 2 pixel block is the first color. Further, in the first embodiment, the description has been given using the 2 × 2 pixel size as the block size, but the present invention is not limited to this. In the description of compression, the description has been given by taking RGB 8 bits as an example of image data. However, image data in CMYK color space, GRAY scale image data, or image data in which each color component has a bit number other than 8 bits. There may be.

  Further, the irreversible compression method by the color reduction process described in the first embodiment is an example, and can be implemented by using another known color reduction process. For example, it can be replaced by another method for obtaining a known color difference.

  Further, the number of compression stages can be repeatedly compressed a plurality of times, for example, in two stages and three stages.

  As described so far, MFPs have a wide variety of image input means, such as those using a scanner and those from a network, with various resolutions. There are also more types depending on the application. For example, even if the input is from the same scanner, the resolution is relatively high at 600 dpi for scanning for copying, but it may be 300 dpi or less for network transmission or FAX. In addition, if the input is PDL, a plurality of users can select from the driver according to the printer resolution, and normally 600 dpi, 1200 dpi, 2400 dpi, and the like can be selected.

  As described above, it is possible to optimize the processing speed, the data size, and the image quality by switching and optimizing the number of compression stages at the time of compression according to various resolutions.

  For example, in the case of PDL, when a document created with word processing software is printed at 1200 dpi, the image size is quadrupled when printed at 2400 dpi. However, the actual contents are the same, and the 2400 dpi image data is more redundant than the 1200 dpi image data. Therefore, when a 2400 dpi image is input, a three-stage compression process is performed to compress it to 300 dpi, and for a 1200 dpi image, a two-stage compression process is performed to compress it to 300 dpi. Although there are differences in image quality degradation and performance due to the large number of compressions, high-resolution image data such as 2400 dpi is usually highly redundant as described above. No difference is born.

  Similarly, in the case of scanning, the data size and performance are optimized by switching the number of compression stages according to the input resolution, such as compressing up to 300 dpi with 2 stages for 1200 dpi and 1 stage for 600 dpi. Can be performed.

[Second Embodiment]
Next, the second embodiment will be described in detail with reference to the drawings. In the second embodiment, the raster image data compression process in the above-described embodiment improves the random accessibility to the image data. Note that in the second embodiment as well, the image processing apparatus will be described by taking an MFP as an example, as in the first embodiment.

  As described above, the compression process described in the first embodiment is a variable-length compression process that depends on image data and changes its compressed data size. Therefore, there is no random accessibility to the compressed data, and it is not suitable for the purpose of cutting out and processing only a part (for example, some objects) of the compressed image data.

  Therefore, a compression method that focuses on improving random accessibility as compared with the method described in the first embodiment will be described in detail with reference to the flowchart of FIG. First, raster image data is input in units of pages shown in FIG. 12 (S1101). Then, one page background pixel is set for each page of image data (S1102). This is pixel data to be used as an initial storage block at the time of compression processing, and normally white (255 for 8-bit RGB color images, 0 for CMYK images) is used. Next, the image data to be input to the compression unit is divided into blocks of a predetermined size (S1103). In the second embodiment, the pixel size is 32 × 32 pixels. In the following description, in order to distinguish from the conventional 2 × 2 pixel block, the rectangular area composed of the plurality of blocks is referred to as a “tile”. As shown in FIG. 12, there are 16 × 16 2 × 2 pixel blocks in one tile.

  Next, predetermined fixed-length header information is assigned to the tile (S1104). This header information is information obtained by adding a color reduction flag to a page ID, tile coordinates, color space, number of bits of pixel data, tile data size, presence / absence of attribute information, compression flag, and the like. Here, the page ID is a unique ID number assigned to each page. The tile coordinate is coordinate information indicating where the tile is located on the raster image in page units. Here, coordinates are described in two dimensions, an X coordinate and a Y coordinate. The color space is information indicating an identifier for identifying whether the tile is an RGB image, a CMYK image, or a GRAY-SCALE image. The number of bits of pixel data is information indicating the bit length per pixel in the tile. The data size is information in byte units indicating the data size of the first color of the tile and the data size of the second, third, and fourth color data. The presence / absence of attribute information is information indicating whether or not attribute information such as characters and photographs is added to the image data in units of pixels. The compression flag is flag (or counter) information indicating whether the tile is compressed data, non-compressed data, and how many times compression is performed at that time. The subtractive color flag is flag information indicating whether the tile is applied with the subtractive color process, is not applied data, or how many times (for example, how many times) the tile is applied. This color reduction flag will be described later in the following embodiments.

  Next, with reference to the color reduction flag, which is the color reduction information in tile units, the tile is subjected to switching between color reduction compression processing and compression processing based on the color reduction flag, and this is executed for each tile in a desired resolution (image size). ) Is repeated until (S1105 to S1109). This series of compression processing is the same as the compression processing (S504 to S508) described with reference to FIG. 5 in the first embodiment, and the description thereof is omitted.

  Then, the tile data compressed in response to the above result is packed into one data together with the header information of the tile (S1110). The data structure of the packed data is shown in FIG. Hereinafter, a data unit including the above-described header portion is referred to as a “packet”. In order to make such a packet, after the compression process is completed for each tile and the data size is determined, the data is packed with the space between the first color storage unit and the second, third, and fourth color storage units packed. To do. That is, the header, the pattern flag storage unit, the first color storage unit, and the second, third, and fourth color storage units are data-formed as tiles. Thereafter, the data is output to the memory via the DMAC. Next, the coordinates and sizes of the packets are listed as a list and created (updated) as a packet management table (S1111). An example of this packet management table is shown in FIG. By repeating the above process up to the last tile (YES in S1112), the raster image compression process in units of pages is completed.

  When data is written to the memory in units of tiles as described above, as shown in FIG. 14, the size differs for each packet, and the top address of each packet is skipped. Therefore, the start address of the packet at an arbitrary coordinate is searched using the packet management table. Therefore, if the write address of the leading packet is known, the leading address of any packet can be obtained with the data size up to the coordinates described in the packet management table as an offset. For example, when the third packet is read in the memory space shown in FIG. 15, the total size of the first and second packets is obtained from the packet management table, and the third packet is added by adding an offset to the address of the first packet. Calculate the packet address. Then, the data of the third packet can be acquired by reading the data from the calculated third packet address.

  As described above, since arbitrary data can be accessed in tile units, partial processing of an image becomes possible. For example, when it is desired to extract and process a partial area of an image, it is only necessary to acquire and process packet data corresponding to the area. Even when a partial region of an image is extracted and processed as described above, processing can be performed using compressed data as described in the first embodiment.

  It is also possible to perform multistage compression on a tile basis. The data structure of the packet at this time is as shown in FIG. 16, and data is nested in the first color storage unit. When performing multi-stage compression, as described in the first embodiment, the first color data in the output data is further compressed. Therefore, the data of the first color storage unit is nested in the data structure of the packet.

  In addition, since the expansion process is provided with a header for each packet, the process is performed using information described in each header. The decompression process is a process for decompressing the packetized data to image data for each tile. First, when the compression flag indicates non-compression, the data from which the header is removed is output, and when not, decompression processing is performed. In the expansion process, the pattern flag storage position, the first color data storage position, and the second, third, and fourth color storage positions are obtained from the header, and the subsequent processing is sequentially expanded into tile image data as in the above-described embodiment. .

  For example, since the position of the pattern flag is after the fixed-length header, it can be obtained by adding an offset. If the tile size is 32 × 32 pixels, the pattern flag size is fixed at 32 × 32 bits, and the first color data storage position is obtained by adding an offset from the pattern flag position. Finally, the data storage position for the second, third, and fourth colors has an indefinite length, so the offset is added to the data position for the first color with reference to the first color size described in the header. Ask for it.

  When the first color data is further compressed, the next data size described in the header for each number of stages is referred to, the data is read while adding an offset for that size, and the expansion process Do it later.

  As described above, according to the second embodiment, even in the variable length compression process in which the compressed data size changes depending on the image data, the random accessibility to the compressed data is improved, and the image is tiled. It is possible to cut out and process only a part of.

[Third Embodiment]
Next, the third embodiment will be described in detail with reference to the drawings. In the third embodiment, in the compression processing of raster image data in the first and second embodiments described above, the subtractive color processing is adaptively controlled with respect to input image data. Note that, in the third embodiment, as an image processing apparatus, an MFP will be described as an example, as in the first embodiment.

  In the process shown in the flowchart of FIG. 9 described above, when the color reduction process is possible, the color reduction process is immediately executed to generate the color reduction flag. In this case, for example, when information indicating that one color reduction is allowed is set as the color reduction information, the color reduction processing is always performed in the first color reduction compression processing, and a color reduction flag is generated. However, in the compression process in page units or tile units, the color reduction process does not work effectively in the first compression process, and the color reduction process may work effectively only in the second and subsequent compression processes. In the third embodiment, an example will be described in which color reduction processing is adaptively controlled by performing processing in consideration of whether the color reduction processing has worked effectively.

  Therefore, a color reduction compression method focusing on controlling the color reduction processing adaptively for input image data will be described in detail with reference to the flowchart of FIG. However, since the flow of the color reduction compression process shown in FIG. 17 is substantially the same as the flow of the description of the color reduction compression process shown in FIG. 9 described above, only the difference will be described below.

  In the first and second embodiments, as shown in FIG. 9, the color reduction flag as the color reduction information is unconditionally added when the color reduction processing is applied (S905, S911, S917). On the other hand, in the third embodiment, as shown in FIG. 17, the color reduction flag, which is color reduction information, detects whether or not the color reduction is effective when the color reduction processing is applied (S1705, S1712, S1719). That is, when the difference from the input is greater than 0 (Yes in S1705, S1712, and S1719), a color reduction flag as color reduction information is added for the first time (S1706, S1713, and S1720), and the difference from the input is 0. If there is, the color reduction flag is not added. In other words, the case where the difference from the input is 0 means that there is no pixel value reduction effect due to the color reduction processing for the input pixel. Therefore, if color reduction information indicating that the color reduction processing has been performed in such a case is generated, the color reduction is actually performed even though the pixel value can be reduced by the color reduction processing in the subsequent compression processing. It will not be done. For this reason, there may be a case where a sufficient reduction effect cannot be obtained. Therefore, by adding the conditions shown in FIG. 17, when the color reduction process is applied in units of pages or tiles, it is possible to realize color reduction control only when the color reduction in the area is effectively acting.

  The above-mentioned condition that the difference from the input is 0 is that effective color reduction control is performed when a part or all of the input high-resolution image data is composed of low-resolution image objects. It is for realizing. Taking the same image size as an example, the same pixel value is continuous in the low-resolution image object portion as compared with the high-resolution image object portion. Therefore, the color reduction processing may not work effectively on the low-resolution image object portion. Specifically, for example, when compression processing is performed in units of tiles as in the second embodiment, only 300 dpi image data is included in a part of the tile unit area of 1200 dpi image data. At this time, the color reduction flag is added to the tile in the control of the flowchart of FIG. 9, but the color reduction flag is not added to the tile in the control of the flowchart of FIG. As described above, by correctly detecting the color reduction information for each region, it is possible to more effectively realize the color reduction control performed for the purpose of increasing the compression efficiency by irreversible compression while minimizing image quality degradation.

  Next, as a specific example, the processing described in the third embodiment is performed on page image data in which 8 × 8 pixel images A, B, and C having different resolutions shown in FIG. 18 are input as three pages. A case will be described in which the compression processing is performed with the upper limit of the number of color reductions and the decompression processing is performed again. The pixel values shown in FIG. 18 are highlighted with a gradation having a large color difference so as to be easily understood visually, but the image data to be actually processed is a gradation having a small color difference at a level that cannot be visually confirmed. Therefore, even if 8 × 8 pixels shown in FIG. 18 are output in one color, there is no problem with image quality.

  As shown in FIG. 18, the first compression process (from 600 dpi to 300 dpi) is executed for the page of image A. In the image A, as a pattern when the compression process is performed with 2 × 2 blocks, all the patterns are only the first color. That is, since the difference between the average value of the input pixels and the pixel value of each input pixel is 0, only the first color is output as output data without applying the color reduction process in the first compression process. Become. Next, a second compression process (from 300 dpi to 150 dpi) is performed on the compressed image A data (that is, data of the pixel values of the first color). In this case as well, since the block pattern is the pattern of only the first color, only the first color is output without applying the color reduction process. Further, a third compression process (150 dpi to 75 dpi) is performed on the compressed image A data. In this case, the second color pattern appears for the first time. Here, since the difference between the pixel value of the first color and the pixel value of the second color is not 0, the color reduction process is applied for the first time. As a result of the color reduction processing, only the first color averaged with the second color is output. As a result of the above, a total of one color value (the first color output by the third compression process) is stored in the memory as a result of the image compression (the resolution is compressed from 600 dpi to 75 dpi) for the image A equivalent to 600 dpi. . Note that, as a result of the color reduction process, the first color averaged with the second color is output, but the first color may be output as it is. When this compressed image is expanded, an image A equivalent to 300 dpi shown in FIG. 18 is output. That is, since the color reduction process, which is an irreversible compression process, is performed, the expanded image cannot reproduce the image before compression completely, but can reproduce image data having half the original resolution.

  Similarly, for the page of image B, the first compression process (1200 dpi to 600 dpi) is executed. In this case, since the block pattern is a pattern of only the first color, only the first color is output without applying the color reduction process. As already described, it should be noted that in the present embodiment, the upper left pixel of the 2 × 2 block is the first color. Next, a second compression process (from 600 dpi to 300 dpi) is performed on the compressed image B data. In this case, the second color pattern appears for the first time. Here, since the difference between the pixel value of the first color and the pixel value of the second color is not 0, only the first color averaged with the second color is output by applying the color reduction process for the first time. Further, a third compression process is executed for the compressed image B data, and since the color reduction process has been completed, the first color and the second color are output without applying the color reduction process. As a result of the above, a total of two color values (first color and second color output in the third compression process) are stored in the memory as a result of image compression on the image B equivalent to 1200 dpi. When this is expanded, an image B equivalent to 600 dpi is output.

  Similarly, the first compression process is executed for the image C, and the first color that is averaged with the second color is output by applying the color reduction process for the first time. Next, the second compression process is executed for the image C, and since the color reduction process has been completed, the first color and the second color are output without applying the color reduction process. Furthermore, since the image C has been subjected to the third compression process and has been subjected to the color reduction process, the first color and the second color are output without applying the color reduction process. As a result of the above, a total of six color values (the second color output in the second compression process and the first color output in the third compression process) as a result of the image compression on the image C corresponding to 2400 dpi, (Second color) is stored in the memory. When this is expanded, an image B equivalent to 1200 dpi is output.

  As described above, in the example illustrated in FIG. 18, when the color reduction process is applied to the page of the image A with only the first compression process as in the first embodiment, the color reduction is performed. Does not work effectively. However, according to the present embodiment, it is possible to achieve color reduction for one pixel at the time of the third compression. In addition, for example, when the color reduction process is applied to the page of the image B with only the first compression process, the color is not reduced, but according to the present embodiment, four pixels at the time of the second compression are used. Can achieve minute color reduction. For the page of image C, for example, in the same manner as when the color reduction process is applied only in the first compression process, the color reduction for 16 pixels at the time of the first compression is achieved also in this embodiment. it can. In this way, by applying color reduction only at the most effective resolution in each page image data of image A, image B, and image C, the compression rate by color reduction is maximized while minimizing the problem of image quality degradation. Color reduction control that can be improved greatly can be realized.

  In the example of FIG. 18 described above, it has been described that the image A, the image B, and the image C are each one page of image data. The example of FIG. 18 can be similarly described even when the tiled data described in the second embodiment is handled. In the second embodiment, since the color reduction flag is provided for each tile, the color reduction processing can be controlled particularly appropriately when objects having different resolutions are included in one page. In the second embodiment, the tile is described as 32 × 32 pixels, but it should be noted that in FIG. 18, the tile is an 8 × 8 pixel for the sake of brevity. Even when the image A, the image B, and the image C in FIG. 18 are included in one page as tiles, the image A, the image B, and the image C as described above each form a page. Subtractive color compression processing is performed.

  When the image A, the image B, and the image C shown in FIG. 18 constitute one page of image data as tiles, the result of the color reduction compression process is as follows. First, out of (8 × 8 pixels) × 3 images = 192 pixels, for example, when the color reduction process is applied only in the first compression process, it corresponds to the second color of the first compression of the image C The reduction of 16 pixels can be achieved. On the other hand, according to the present embodiment that adaptively controls the color reduction processing, the second color pixel of the third compression of the image A and the second color pixel of the second compression of the image B are added. In addition, a reduction of 21 pixels can be achieved. In this way, by applying the color reduction only at the most effective resolution for each object area of each tile of the tiled image A, image B, and image C, the color reduction is performed while minimizing the problem of image quality degradation. Color reduction control that can improve the compression ratio to the maximum can be realized.

  As described above, according to the third embodiment, the condition that the color reduction flag is added only when the color reduction process in which the color reduction effectively works is applied to the input image data. Furthermore, it is possible to effectively control the color reduction process.

[Fourth Embodiment]
Next, a fourth embodiment will be described in detail with reference to the drawings. In the fourth embodiment, raster image data compression processing in the above-described embodiment is performed by controlling the color reduction processing in consideration of the degree of color reduction for input image data. Note that, in the fourth embodiment, as an image processing apparatus, an MFP will be described as an example as in the first embodiment.

  As described above, the color reduction flag as the color reduction information is a simple flag that holds 0 when the color reduction process is not applied in page units or tile units, and holds 1 when it is applied. The color reduction information is not necessarily sufficient to control the number of applications.

  9 and 17, the color reduction compression method focused on controlling the color reduction processing in consideration of the degree of color reduction for the input image data will be described in detail. However, since the flow of the color reduction compression process is the same as the flow of the description of the color reduction compression process shown in FIGS. 9 and 17 described above, only the difference will be described below.

  In the fourth embodiment, a color reduction counter is held as color reduction information instead of the color reduction flag in the color reduction flag generation (S905, S911, S917) in FIG. 9 and the color reduction flag generation (S1706, S1713, S1720) in FIG. To do. This color reduction counter is a counter that holds the degree of color reduction, and holds a counter value, for example, the cumulative number of times that color reduction has been applied, the cumulative error of color values during color reduction, and the like. That is, when the cumulative number of times that color reduction is applied is held, the color reduction process can be controlled by giving a threshold value N so that the image quality is maintained and the compression rate is improved, and the color reduction process is set to the upper limit N times. In other words, when the number of times that the color process is applied is less than the predetermined number, it is possible to control so as to execute the color reduction process. Further, when the accumulated error of the color value at the time of color reduction is held, a threshold value N is given so that both the image quality maintenance and the compression rate are improved, and the upper limit of the accumulated error for measuring the degree of color reduction is set to the N level. Can be controlled. That is, it is possible to perform control so that the color reduction process is executed only when the cumulative error is less than a predetermined threshold.

  As described above, according to the fourth embodiment, it is possible to effectively control the color reduction processing by setting the condition that the color reduction counter is counted in consideration of the degree of color reduction with respect to the input image data. It becomes possible to do.

[Other Embodiments]
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments may be supplied to a system or apparatus via a network or various storage media. In this case, the computer (or CPU, MPU, etc.) of the system or apparatus reads the program and executes the process.

Claims (10)

An image processing apparatus that divides image data into blocks, sequentially processes the divided blocks, and performs compression processing on the image data a plurality of times.
Determining means for determining whether or not to execute a color reduction process for reducing the number of colors in a block based on the color reduction information;
A color reduction processing means for executing the color reduction processing on the processing target block when it is determined by the determination means to execute the color reduction processing;
Updating means for updating the color reduction information in accordance with the result of the color reduction processing;
A specifying unit for specifying a pattern flag corresponding to an arrangement pattern of color data included in the block by comparing the color data of each pixel of the block to be processed;
Output means for outputting the pattern flag specified by the specifying means and the color data for the number of colors included in the block as compressed output data;
And an input unit that further divides the aggregate data of the first color data of the block for each block until the predetermined resolution is reached, and inputs the divided data again to the determination unit as a block to be processed. Processing equipment.   The image processing apparatus according to claim 1, wherein the update unit updates that the color reduction process has been performed only when the number of colors is reduced by the color reduction process for the processing target block.   The image processing apparatus according to claim 1, wherein the color reduction information is set for each tile that is a rectangular area including a plurality of blocks.   The image processing apparatus according to claim 3, wherein the output unit outputs the output data as a packet for each tile, and the color reduction information is included in a header portion of the packet. The update unit updates a counter value that holds the number of times the color reduction process is applied to the processing target block as the color reduction information.
The image processing apparatus according to claim 1, wherein the determination unit determines to perform the color reduction process only when the number of times the color reduction process is applied is less than a predetermined number with reference to the counter value. The update means updates a counter value that holds the number of times the number of colors has been reduced by the color reduction processing for the processing target block as the color reduction information,
2. The determination unit according to claim 1, wherein the determination unit determines to perform the color reduction process only when the number of times the number of colors is reduced by the color reduction process is less than a predetermined number with reference to the counter value. Image processing apparatus. The update means updates the counter value that holds the error of the color value when the number of colors is reduced by the color reduction process for the block to be processed as a cumulative error as the color reduction information,
2. The determination unit according to claim 1, wherein the determination unit refers to the counter value and determines to execute the color reduction process only when a cumulative error caused by the color reduction process by the color reduction process is less than a predetermined threshold. An image processing apparatus according to 1.   The image processing apparatus according to claim 1, further comprising an image processing unit that performs color processing that is completed by one pixel input and one pixel output on the set data of the color data of the first color. An image processing method in which image data is divided into blocks, the divided blocks are sequentially processed, and image data is compressed a plurality of times.
A determination step of determining whether or not to execute a color reduction process for reducing the number of colors of the block based on the color reduction information;
A subtractive color processing step for executing the subtractive color processing on the processing target block when it is determined in the determination step that the subtractive color processing is to be executed;
An update step of updating the color reduction information according to a result of the color reduction processing;
A specifying step of specifying a pattern flag corresponding to an arrangement pattern of color data included in the block by comparing color data of each pixel of the block to be processed;
An output step of outputting the pattern flag specified in the specifying step and the color data for the number of colors included in the block as compressed output data;
And an input step of further dividing the set data of the color data of the first color of the block for each block until the predetermined resolution is reached and inputting the data as a processing target block to the determination step again Processing method.   A program for causing a computer to execute the image processing method according to claim 9.

Images