JP5515825B2 - Image processing circuit and image forming apparatus - Google Patents

Description

  The present invention relates to an image processing circuit and an image forming apparatus.

  In recent years, image forming apparatuses such as color printers and color copying machines have been widely used. In the image forming apparatus, a plurality of storage devices such as HDD (Hard Disk Drive) and DRAM (Dynamic Random Access Memory) for storing data are operated in parallel for the purpose of speeding up the processing. In order to simplify the hardware configuration and reduce costs, color image data compression techniques are effective.

  For example, multi-value image data is divided into blocks of N × M pixels, pixel data for each block is fixed-length compressed at a constant compression rate, and a part of a plurality of pieces of fixed-length compressed information is variable-length compressed. An image encoding device has been proposed (see Patent Document 1).

  Here, an example of a conventional compression method will be described with reference to FIG. The conventional compression unit 100 performs compression processing on image data divided into blocks of 4 pixels × 4 pixels for each color of yellow (Y), magenta (M), cyan (C), and black (K). Do. The compression unit 100 includes BTC (Block Truncation Coding) compression units 120Y, 120M, 120C, and 120K, differential data generation units 130Y, 130M, 130C, and 130K, data concatenation units 140Y, 140M, 140C, and 140K, a Huffman encoding unit 150Y, 150M, 150C, 150K are provided. It is assumed that data of 8 bits per 1 CLK is input to the compression unit 100 for each color of YMCK.

  For each block, the BTC compression unit 120Y converts the yellow input data belonging to the block into a predetermined number of gradation data based on the minimum and maximum values of the yellow input data (8 bits) belonging to the block. The gradation data (3 bits / CLK), the minimum value data (8 bits / 16 CLK), and the maximum value data (8 bits / 16 CLK) corresponding to each input data are output.

  The difference data generation unit 130Y generates difference data indicating a difference from adjacent pixels with respect to the gradation data output from the BTC compression unit 120Y. The difference data generation unit 130Y outputs the minimum value data and the maximum value data as they are.

  The data concatenation unit 140Y bit-concatenates adjacent four pixel data for the difference data of the gradation data output from the difference data generation unit 130Y. The amount of difference data after bit concatenation is 12 bits per 4 CLK. Note that the data linking unit 140Y outputs the minimum value data and the maximum value data as they are.

  The Huffman encoding unit 150Y performs Huffman encoding processing on 12-bit concatenated gradation data per 4 CLK, minimum data of 8 bits per 16 CLK, and maximum data of 8 bits per 16 CLK.

  Regarding the BTC compression units 120M, 120C, and 120K, the difference data generation units 130M, 130C, and 130K, the data concatenation units 140M, 140C, and 140K, and the Huffman encoding units 150M, 150C, and 150K, the BTC compression unit 120Y and the difference data generation unit 130Y Since the configuration is the same as that of the data concatenation unit 140Y and the Huffman encoding unit 150Y, description thereof will be omitted.

JP 2001-94985 A

  However, the conventional compression technique may not be sufficient in the compression rate of image data. Therefore, further improvement of the compression rate is desired.

  The present invention has been made in view of the above problems in the prior art, and an object thereof is to improve the compression rate of image data.

In order to solve the above-mentioned problem, the invention according to claim 1 is the smallest of the multi-value data belonging to the block for each block obtained by dividing the image data of each color composed of multi-value data of a plurality of colors into a plurality of blocks. Based on the value and the maximum value, classify the multi-value data belonging to the block into a predetermined number of gradation data, and output the classified gradation data, the minimum value and the maximum value, A first compression unit that compresses image data of each color by a fixed length for each color, and bit data of pixel data at the same position of gradation data of each color output by the first compression unit , and the first compression unit the data between the blocks of the same location of the minimum and maximum values of the respective output color bit-connected by a connecting means for the image data of each of the plurality of color as one of the image data, the bit concatenated image Second compression means for variable-length compressing chromatography data is an image processing circuit comprising a.

Invention according to claim 2, the image processing circuit according to claim 1, for each color gradation data outputted by said first compression means, the first difference data indicating a difference between adjacent pixels generated, the minimum value and maximum value of each color output by the first compression means comprises a differential data generating means for generating a second difference data indicating the difference between the previous block, said coupling means, said the data between the pixels at the same position of the first differential data of each color that is generated by the difference data generating means and the bit coupling of the co-located block in the second difference data of each color that is generated by the difference data generating means Bit concatenates data .

According to a third aspect of the present invention, in the image processing circuit according to the first or second aspect , the variable length compression by the second compression means is a Huffman coding process.

A fourth aspect of the present invention is an image forming apparatus comprising the image processing circuit according to any one of the first to third aspects.

According to the present invention, it is possible by bit coupling data between the pixels at the same position of the gradation data of each fixed length compressed color, improve the compression rate of the image data.

1 is an overall configuration diagram of an image forming apparatus according to an embodiment of the present invention. It is a block diagram of a compression part. It is a block diagram of a BTC compression part. (A) is an example of yellow input data. (B) is an example of magenta input data. (C) is an example of cyan input data. (D) is an example of black input data. (A) is a figure which shows the correspondence of the yellow input data and gradation data in a BTC compression part. FIG. 6B is a diagram illustrating a correspondence relationship between magenta input data and gradation data in the BTC compression unit. (C) is a diagram showing a correspondence relationship between cyan input data and gradation data in the BTC compression unit. FIG. 6D is a diagram illustrating a correspondence relationship between black input data and gradation data in the BTC compression unit. (A) is a figure which shows the gradation data, minimum value data, and maximum value data of yellow output from a BTC compression part. FIG. 6B is a diagram illustrating magenta gradation data, minimum value data, and maximum value data output from the BTC compression unit. (C) is a diagram showing cyan gradation data, minimum value data, and maximum value data output from the BTC compression unit. (D) is a diagram showing black gradation data, minimum value data, and maximum value data output from the BTC compression unit. It is a block diagram of a difference data generation part. (A) is a figure which shows the difference data of yellow. (B) is a figure which shows the difference data of magenta. (C) is a figure which shows the difference data of cyan. (D) is a figure which shows the difference data of black. It is a figure for demonstrating the position of each pixel in a block. It is a figure which shows the connection gradation data, connection minimum value data, and connection maximum value data for 1 block. It is a block diagram of the conventional compression part.

  Embodiments of the present invention will be described below with reference to the drawings. The following is one embodiment of the present invention and does not limit the present invention.

  FIG. 1 shows an overall configuration of an image forming apparatus 1 according to the present embodiment. As shown in FIG. 1, the image forming apparatus 1 includes a scanner unit 2, an image processing unit 3, a line block conversion unit 4, a compression unit 5, a data distribution unit 6, HDDs 7a, 7b, 7c and 7d, a data integration unit 8, An expansion unit 9, a block line conversion unit 10, an image processing unit 11, and an image forming unit 12 are provided. The line block conversion unit 4, the compression unit 5, the data distribution unit 6, the data integration unit 8, the decompression unit 9, and the block line conversion unit 10 are each configured by an ASIC (Application Specific Integrated Circuit).

  The scanner unit 2 includes a light source, a CCD (Charge Coupled Device) image sensor, an A / D converter, and the like, and forms an image of a document by forming an image of reflected light of light scanned from the light source to the document and performing photoelectric conversion. Are read as R (red), G (green), and B (blue) signals, and the read image is A / D converted and output to the image processing unit 3.

  The image processing unit 3 converts the RGB data output from the scanner unit 2 into image data (YMCK data) including four colors of yellow, magenta, cyan, and black, and outputs the image data to the line block conversion unit 4. In the present embodiment, the image processing unit 3 converts RGB data into 256-level multi-value data (8 bits) for each color of YMCK.

  The line block conversion unit 4 divides the image data into 4 pixel × 4 pixel blocks for each color of YMCK. Specifically, the line block conversion unit 4 converts the image data arranged in the order along the line in the main scanning direction for each YMCK color in the order of the block unit.

  The compression unit 5 performs compression processing on the image data output from the line block conversion unit 4. Since the writing speed to the HDDs 7a, 7b, 7c, and 7d is limited, it is necessary to compress the image data so that the amount of data that can be written to the HDDs 7a, 7b, 7c, and 7d is obtained. In this embodiment, it is assumed that the write limit is 4 bits per 1 CLK for each HDD 7 a, 7 b, 7 c, 7 d and 16 bits per 1 CLK in the entire HDD 7 a, 7 b, 7 c, 7 d.

  The data distribution unit 6 distributes and writes the image data compressed by the compression unit 5 to each of the HDDs 7a, 7b, 7c, and 7d.

  The HDDs 7a, 7b, 7c, and 7d store the compressed image data distributed by the data distribution unit 6.

  The data integration unit 8 reads the image data stored in the HDDs 7a, 7b, 7c, and 7d, and integrates the image data distributed to the HDDs 7a, 7b, 7c, and 7d.

  The decompression unit 9 performs decompression processing on the image data integrated by the data integration unit 8 and outputs the image data to the block line conversion unit 10.

  The block line conversion unit 10 converts the image data arranged in the order of blocks of 4 pixels × 4 pixels for each YMCK color, in the order along the lines in the main scanning direction.

  The image processing unit 11 performs image processing such as screen processing on the YMCK data output from the block line conversion unit 10 and outputs the processed image data to the image forming unit 12.

  The image forming unit 12 includes a photosensitive drum, a charging unit that charges the photosensitive drum, an exposure unit that exposes the surface of the photosensitive drum based on YMCK data output from the image processing unit 11, a developing unit that attaches toner to the photosensitive drum, The image forming apparatus includes a transfer unit that transfers a toner image formed on the photosensitive drum to the printing paper, a fixing unit that fixes the toner image formed on the printing paper, and the like.

  FIG. 2 shows the configuration of the compression unit 5. As shown in FIG. 2, the compression unit 5 includes BTC (Block Truncation Coding) compression units 20Y, 20M, 20C, and 20K, differential data generation units 30Y, 30M, 30C, and 30K, a data concatenation unit 40, and a Huffman encoding unit 50. Is provided. It is assumed that 8 bits of data per 1 CLK are input to the compression unit 5 for each color of YMCK. In other words, the compression unit 5 receives 32 bits of data per 1 CLK for YMCK4 colors.

  The BTC compression units 20Y, 20M, 20C, and 20K are first compression units that perform fixed-length compression (primary compression) on image data composed of multi-value data of yellow, magenta, cyan, and black, respectively. Fixed length compression refers to data compression performed at a constant compression rate.

  The BTC compression unit 20Y, for each block obtained by dividing the yellow image data into a plurality of blocks, based on the minimum value and the maximum value of the yellow input data (8 bits) belonging to the block, the yellow input data belonging to the block Are classified into a predetermined number of gradation data, and the classified gradation data, minimum value data, and maximum value data are output. In the present embodiment, the “predetermined number” is “8”, and the BTC compression unit 20Y compresses 8-bit input data into 3-bit gradation data (0 to 7). The BTC compression unit 20Y performs BTC compression on the yellow input data (8 bits / CLK), the gradation data (3 bits / CLK) corresponding to each input data, the minimum value data (8 bits / 16 CLK), and the maximum Value data (8bit / 16CLK) is output.

  Here, the configuration of the BTC compression unit 20Y will be described with reference to FIG. As shown in FIG. 3, the BTC compression unit 20 </ b> Y includes a shift register 21, a minimum / maximum value calculation unit 22, a gradation data calculation unit 23, and a shift register 24.

The shift register 21 holds 16 pieces of 8-bit input data (one block of 4 pixels × 4 pixels) input once in 1 CLK.
The minimum value / maximum value calculation unit 22 calculates a minimum value and a maximum value from 16 pieces of input data for one block held in the shift register 21. That is, the minimum value / maximum value calculation unit 22 outputs the minimum value data of 8 bits and the maximum value data of 8 bits once every 16 CLK.

The gradation data calculation unit 23 classifies values from the minimum value to the maximum value into gradation data of 0 to 7 based on the minimum value and the maximum value calculated by the minimum value / maximum value calculation unit 22, and shifts them. According to which gradation data each of the 16 input data for one block output from the register 21 belongs to, 3-bit gradation data is output.
The shift register 24 holds 16 pieces of 3-bit gradation data output by the gradation data calculation unit 23 and outputs 3 bits of gradation data per 1 CLK.

  Since the BTC compression units 20M, 20C, and 20K have the same configuration as that of the BTC compression unit 20Y, the description thereof is omitted with reference to FIG. In addition, the same code | symbol as BTC compression part 20Y is used also about each part which comprises BTC compression part 20M, 20C, 20K.

  4A to 4D, yellow, magenta, cyan, and black are input to the BTC compression units 20Y, 20M, 20C, and 20K of the compression unit 5 in units of blocks of 4 pixels × 4 pixels, respectively. An example of input data (image data) is shown.

  In the example of FIG. 4A, since the minimum value of the yellow input data in the block is 10 and the maximum value is 66, the gradation data calculation unit 23 of the BTC compression unit 20Y is between 10 and 66. Is divided into seven equal parts, and the input data is divided into seven equal parts, according to which one of 18, 26, 34, 42, 50, 58, minimum value 10, maximum value 66 is closest to 0 7 gradation data is output. Specifically, the gradation data calculation unit 23 of the BTC compression unit 20Y, as shown in FIG. 5A, when the input data is closest to 10, that is, when the input data is 10 to 13, The gradation data “0” is output. When the input data is closest to 18, that is, when the input data is 14 to 21, gradation data “1” is output. When the input data is closest to 26, that is, when the input data is 22 to 29, the gradation data “2” is output. When the input data is closest to 34, that is, when the input data is 30 to 37, the gradation data “3” is output. When the input data is closest to 42, that is, when the input data is 38 to 45, the gradation data “4” is output. When the input data is closest to 50, that is, when the input data is 46 to 53, gradation data “5” is output. When the input data is closest to 58, that is, when the input data is 54 to 61, the gradation data “6” is output. When the input data is closest to 66, that is, when the input data is 62 to 66, gradation data “7” is output.

  FIG. 6A shows gradation data, minimum value data, and maximum value data that are output when the yellow input data shown in FIG. 4A is input to the BTC compression unit 20Y.

  Similarly, in the example of magenta, cyan, and black input data shown in FIGS. 4B to 4D, the gradation data calculation unit 23 of the BTC compression units 20M, 20C, and 20K is connected to FIGS. As shown in d), gradation data corresponding to the input data is output. FIGS. 6B to 6D show gradations that are output when the magenta, cyan, and black input data shown in FIGS. 4B to 4D are input to the BTC compression units 20M, 20C, and 20K. Indicates data, minimum value data, and maximum value data.

  Each of the BTC compression units 20Y, 20M, 20C, and 20K outputs, for each color, gradation data, minimum value data, maximum value data, and 4 bits of data per 1 CLK. That is, for YMCK4 colors, 16-bit data is output per 1 CLK. As described above, since the data input to the compression unit 5 is 32 bits per 1 CLK for YMCK 4 colors, the BTC compression units 20Y, 20M, 20C, and 20K perform fixed length compression with a compression rate of 50%. Become. The compression rate is obtained by (data amount after compression) / (data amount before compression) × 100 (%).

  The difference data generation units 30Y, 30M, 30C, and 30K differ from the previous data for each of the gradation data, minimum value data, and maximum value data output from the BTC compression units 20Y, 20M, 20C, and 20K. Is generated and output to the data concatenation unit 40. The difference data generation units 30Y, 30M, 30C, and 30K generate difference data indicating differences from adjacent pixels for the gradation data of each color that has been fixed-length compressed by the BTC compression units 20Y, 20M, 20C, and 20K. In addition, the difference data generation units 30Y, 30M, 30C, and 30K are the minimum value data and maximum value data in the previous block for the minimum value data and the maximum value data output from the BTC compression units 20Y, 20M, 20C, and 20K. The difference data indicating the difference is generated.

  Here, a configuration of the difference data generation unit 30Y will be described with reference to FIG. As illustrated in FIG. 7, the difference data generation unit 30Y includes a register 31 and a difference calculation unit 32.

The register 31 holds one piece of data output from the BTC compression unit 20Y.
The difference calculation unit 32 calculates a difference between the data output from the BTC compression unit 20Y and the previous data held in the register 31. Specifically, the difference calculation unit 32 generates difference data by subtracting the previous data from the processing target data.

  Since the difference data generation units 30M, 30C, and 30K have the same configuration as the difference data generation unit 30Y, illustration and description thereof are omitted.

  FIG. 8A shows the difference data of the yellow gradation data, minimum value data, and maximum value data shown in FIG. Specifically, for the pixel of P1 in FIG. 9 in the block, the difference from the pixel of P16 of the previous block is taken. That is, the difference data of the gradation data at the position P1 (hereinafter referred to as gradation difference data) is (gradation data of the pixel of P1) − (gradation data of the pixel of P16 of the previous block). Similarly, for the pixel P2 in FIG. 9, the difference data is (gradation data for the pixel P2) − (gradation data for the pixel P1), and for the pixel P3 in FIG. The gradation data of the pixel P3) − (the gradation data of the pixel P2). For the pixel P4 in FIG. 9, the difference data is (the gradation data of the pixel P4) − (the gradation of the pixel P3). The difference data for the pixel P5 in FIG. 9 is (gradation data for the pixel P5) − (gradation data for the pixel P4).

  For the difference data between the minimum value data and the maximum value data, the difference between the minimum value data and the maximum value data in the previous block is taken. For example, difference data of minimum value data (hereinafter referred to as minimum value difference data) is (minimum value data in a processing target block) − (minimum value data in a previous block). Further, the difference data of the maximum value data (hereinafter referred to as the maximum value difference data) is (maximum value data in the processing target block) − (maximum value data in the previous block).

  Similarly, FIGS. 8B to 8D show difference data of magenta, cyan, and black gradation data, minimum value data, and maximum value data shown in FIGS. 6B to 6D.

In addition, even when the gradation difference data is a negative value, it is represented by 3 bits. For example, the magenta tone difference data in the pixel at P13 in FIG. 8B (see FIG. 9) is “−1”, and this difference value is the same as “7”. That is, the difference value “7” means “7” or “−1”. The gradation difference data “−1” in the pixel at the position P13 in FIG. 8B has gradation data “2” at the position P13 in FIG. 6B and the position P12 in FIG. 6B. Although the gradation data (previous value) is “3”, it is calculated, but when the previous data is “3”, the difference value is “7”. Therefore, the difference value means “−1”.
Similarly, the minimum value difference data and the maximum value difference data are also represented by 8 bits even when they are negative values.

  The data concatenation unit 40 bit-concatenates the data at the same position of the data of each color for the YMCK four-color gradation difference data output from the difference data generation units 30Y, 30M, 30C, and 30K. It is a connecting means that uses image data as one image data. Specifically, the data concatenation unit 40 concatenates 3 bits of gradation difference data per 1 CLK for 4 colors and outputs 12 bits of concatenated gradation data per 1 CLK.

  Further, the data linking unit 40 bit-links the data of the same block of each color for each of the minimum value difference data and the maximum value difference data for YMCK4 colors output from the difference data generation units 30Y, 30M, 30C, and 30K. The image data for each color of YMCK is set as one image data. Specifically, the data linking unit 40 concatenates 4 bits each of the minimum value difference data of 8 bits per 16 CLK and the maximum value difference data of 8 bits per 16 CLK for each of the four colors, and the combined minimum value data of 16 bits per 16 CLK. Outputs 32-bit concatenated maximum value data.

FIG. 10 shows an example of one block of connected gradation data, connected minimum value data, and connected maximum value data generated based on the difference data in FIGS. For example, for the pixel at position P4 in FIG. 9, the yellow tone difference data is 1, the magenta tone difference data is 2, the cyan tone difference data is 0, and the black tone difference data is 1. In FIG. 10, the connected gradation data obtained by bit-concatenating these is expressed as {1 2 0 1}. That is, the symbol {} indicates that four pieces of 3 bit data (1, 2, 0, 1) are bit-connected to form one piece of 12 bit data.
{1 2 0 1} is the data
1 * (2 ^ 9) + 2 * (2 ^ 6) + 0 * (2 ^ 3) + 1 * (2 ^ 0)
It becomes.

  The Huffman encoding unit 50 is a second compression unit that performs variable length compression (secondary compression) on the image data bit-connected by the data connection unit 40. Variable length compression refers to compression with different compression rates depending on the data configuration. Specifically, the Huffman encoding unit 50 includes the concatenated gradation data (12 bits / CLK), the concatenated minimum value data (32 bits / 16 CLK), and the concatenated maximum value data (32 bits / 16 CLK) that are bit-concatenated by the data concatenation unit 40. On the other hand, a Huffman encoding process is performed. The Huffman encoding process is a lossless compression process in which data with a high appearance rate is converted into a code with a small number of bits, and data with a low appearance rate is converted into a code with a large number of bits. In addition, before performing the Huffman encoding process in the Huffman encoding unit 50, it is desirable to optimize the Huffman table used for the Huffman encoding process based on the input image data.

[Experimental result A: photographic image]
As an example of a photographic image, a SCID (Standard Color Image Data: high-definition color digital standard image data) chart is used, and a compression method in this embodiment (a compression method performed by the compression unit 5 shown in FIG. 2) and a conventional one are used. The compression method (compression method performed by the compression unit 100 shown in FIG. 11) was compared. The SCID chart is standard image data that has been internationally standardized for the purpose of standardizing printing operations and evaluating encoding characteristics. In this experiment, a chart called “Portrait” was used.

  In the conventional compression method, the compression ratio by BTC compression is (1 / 2.000) × 100 (%), the compression ratio by Huffman coding is (1 / 1.089) × 100 (%), and the final compression ratio is ( 1 / 2.000) × (1 / 1.089) × 100 = 45.9 (%). On the other hand, in the compression method in the present embodiment, the compression rate by BTC compression is (1 / 2.000) × 100 (%), and the compression rate by Huffman coding is (1 / 1.226) × 100 (%). The final compression ratio was (1 / 2.000) × (1 / 1.226) × 100 = 40.8 (%). According to the experimental result A, the amount of data could be reduced by using the compression method in the present embodiment compared to the conventional compression method.

[Experiment B: Character image]
As an example of a character image, using a Spencer chart, the compression method in this embodiment (the compression method performed by the compression unit 5 shown in FIG. 2) and the conventional compression method (the compression performed by the compression unit 100 shown in FIG. 11). Method).

  In the conventional compression method, the compression ratio by BTC compression is (1 / 2.000) × 100 (%), the compression ratio by Huffman coding is (1 / 4.776) × 100 (%), and the final compression ratio is ( 1 / 2.000) × (1 / 4.776) × 100 = 10.5 (%). On the other hand, in the compression method in the present embodiment, the compression rate by BTC compression is (1 / 2.000) × 100 (%), and the compression rate by Huffman coding is (1 / 6.917) × 100 (%). The final compression ratio was (1 / 2.000) × (1 / 6.917) × 100 = 7.23 (%). According to the experimental result B, it was possible to reduce the amount of data by using the compression method in the present embodiment, compared to the conventional compression method.

  However, according to the compression method in the present embodiment, the compression rate is not improved for any image as compared with the conventional compression method. However, as a result of performing experiments on many images using a method of bit-connecting data at the same position of BTC-compressed YMCK data and performing Huffman coding, the compression ratio is improved (the amount of compressed data is reduced). It turns out that there is a tendency to decrease. In the conventional example, secondary compression is performed in units of four adjacent pixels, but in the present embodiment, YMCK data at the same position is bit-connected to perform secondary compression. The data after bit concatenation is all 12-bit data, but the data of one pixel unit in which YMCK data is bit concatenated tends to have continuous data having a similar value compared to the data of four pixel units of YMCK alone. Therefore, it is considered that the compression rate is improved.

  As described above, according to the image forming apparatus 1 of the present embodiment, the compression rate of image data can be improved by bit-connecting the same-position image data of each color that has been fixed-length compressed. Can do. Therefore, the capacity of the HDD or DRAM can be reduced, and the hardware configuration can be simplified and the cost can be reduced.

  Further, by handling the difference data of the image data, the data generated by bit concatenation by the data concatenation unit 40 tends to be closer to each other.

  Further, by performing Huffman encoding processing as secondary compression, the amount of data after compression can be reduced as the same value in the block increases.

  The description in the above embodiment is an example of the image processing circuit and the image forming apparatus according to the present invention, and the present invention is not limited to this. The detailed configuration and detailed operation of each part constituting the apparatus can be changed as appropriate without departing from the spirit of the present invention.

For example, in the above embodiment, the case where BTC compression is used as the fixed length compression method (primary compression) has been described as an example. However, the present invention is not limited to BTC compression, and other fixed length compression methods may be used.
In the above embodiment, the case where Huffman coding is used as the variable-length compression method (secondary compression) has been described as an example. However, the present invention is not limited to Huffman coding, and includes arithmetic coding, Shannon coding, and the like. The variable length compression method may be used. However, in order to prevent deterioration in image quality, it is desirable that the secondary compression is a reversible compression.

  Moreover, although the said embodiment demonstrated the case where the line block conversion part 4, the compression part 5, the data distribution part 6, the data integration part 8, the expansion | extension part 9, and the block line conversion part 10 were comprised by ASIC, Each unit may be configured by software.

  Moreover, it is good also as a structure remove | excluding the difference data production | generation part 30Y, 30M, 30C, 30K from the compression part 5. FIG. In that case, the data concatenation unit 40 may bit-concatenate the gradation data for the four colors at the same position, the minimum value data for the four colors of the same block, and the maximum value data for the four colors of the same block. .

DESCRIPTION OF SYMBOLS 1 Image forming apparatus 2 Scanner part 3 Image processing part 4 Line block conversion part 5 Compression part 6 Data distribution part 7a, 7b, 7c, 7d HDD
8 Data Integration Unit 9 Expansion Unit 10 Block Line Conversion Unit 11 Image Processing Unit 12 Image Forming Units 20Y, 20M, 20C, 20K BTC Compression Unit 21 Shift Register 22 Minimum Value / Maximum Value Calculation Unit 23 Gradation Data Calculation Unit 24 Shift Register 30Y, 30M, 30C, 30K Difference data generation unit 31 Register 32 Difference calculation unit 40 Data concatenation unit 50 Huffman coding unit 100 Compression unit 120Y, 120M, 120C, 120K BTC compression unit 130Y, 130M, 130C, 130K Difference data generation unit 140Y, 140M, 140C, 140K Data concatenation unit 150Y, 150M, 150C, 150K Huffman coding unit

Claims (4)

For each block obtained by dividing image data of each color composed of multi-value data of a plurality of colors into a plurality of blocks, the multi-value data belonging to the block is preliminarily determined based on the minimum value and the maximum value of the multi-value data belonging to the block. First compression means for classifying the image data of each color into a fixed length for each color by classifying the data into a predetermined number of gradation data and outputting the classified gradation data and the minimum and maximum values ; ,
Bit data of pixel data at the same position in the gradation data of each color output by the first compression means, and block data at the same position of the minimum value and the maximum value of each color output by the first compression means A connecting means for bit-connecting each other, and making the image data for each of the plurality of colors one image data;
Second compression means for variable-length compressing the bit-linked image data;
An image processing circuit comprising: For the gradation data of each color output by the first compression unit, first difference data indicating a difference from an adjacent pixel is generated , and the minimum value and the maximum value of each color output by the first compression unit are generated. , Comprising difference data generation means for generating second difference data indicating a difference from the previous block ,
It said connecting means, the difference data between the pixels at the same position of the first differential data of each color that is generated by the data generating means is bit-connected second differential data of each color that is generated by the difference data generating means Bit-concatenate data of blocks in the same position of
The image processing circuit according to claim 1 . The variable length compression by the second compression means is a Huffman encoding process.
The image processing circuit according to claim 1 or 2. An image forming apparatus including an image processing circuit according to any one of claims 1 to 3.

Images