JP4722681B2 - Compression processing device, compression processing device control method, compression processing control program, and computer-readable recording medium - Google Patents

Description

  The present invention relates to a compression processing apparatus, a compression processing method, a compression processing program, and a computer-readable recording medium for compressing data arranged two-dimensionally.

  When compressing two-dimensional image data (image data) in an image processing system such as a printer driver, there is a method of predicting a target pixel according to the state of its surrounding pixels and performing arithmetic coding according to the prediction result. As an algorithm for compressing such two-dimensional image data, for example, JBIG (Joint Bi-level Image coding experts Group) compression is known (for example, see Non-Patent Documents 1 and 2). In JBIG compression, two-dimensional image data is scanned from left to right one line at a time from the top, cells (pixels) that are units to be encoded are selected, and the cells are encoded one by one. Perform compression. When encoding each cell, refer to several neighboring cells (contexts) in the vicinity of the encoding target cell (target cell), and use the probability that the target cell appears for this neighboring cell. Encoding is performed. The context is a neighboring cell that is specified by a template having a predetermined shape and is used for encoding the target cell at a predetermined relative position with respect to the target cell. FIG. 11 is a diagram of a 2-line template (2-line template) in JBIG compression. In FIG. 11, a cell surrounded by a thick frame is a target cell, and a cell with a diagonal line around it is a neighboring cell (context). Note that, as shown in FIG. 11, the description method in which the target cell is surrounded by a thick line and the neighboring cells are hatched is similarly used in other drawings described in this specification.

  In JBIG compression, reading is performed based on a two-line template or a three-line template, but there are cases where the context protrudes outside the image data and a cell outside the image data must be referred to as a neighboring cell. When the 2-line template shown in FIG. 11 is used, the target cell is within 2 cells from the right end of the image data (FIG. 12A), or is within 4 cells from the left end (FIG. 12B). Or, if it is at the upper end of the image (FIG. 12C), the context protrudes outside the image data. In FIGS. 12A to 12C, a cell in which X is written is a cell in which data is stored, and a cell in which “?” Is written is a cell that protrudes from the image data. Further, a cell surrounded by a thick frame is a target cell, and a hatched cell is a neighboring cell forming a context.

  In this way, when the context protrudes outside the image data and a cell outside the image data must be referred to as a neighbor cell, JBIG compression leaves the neighbor cell outside the image data blank (0 , Zero), and compression is performed. For example, in encoding the target cell, when referring to (reading) the context, processing as shown in FIG. 13 is performed. It is detected whether or not the context protrudes from the image data (S101). If the context does not protrude (NO in S101), the neighboring cell that is the context with the target cell is read (S102). In the case of protruding (YES in S101), the target cell is read (S103), and the neighboring cells that are contexts that do not protrude from the image data are read (S104), and the neighboring cells that are contexts are read from the image data. Zero is substituted for the protruding portion (S105).

  As described above, when it is necessary to refer to a neighboring cell outside the image data, the above processing is performed. Therefore, when JBIG compression is performed on the computer, the neighboring cell is outside the image data. There is a problem in that the compression speed decreases because it is necessary to encode and compress while determining whether or not to protrude.

  As a method for solving the above problem, there is known a compression processing apparatus in which blank areas are provided at both ends of two-dimensional image data to be compressed (see, for example, Patent Document 1). In this apparatus, as shown in FIGS. 14A to 14C, memory for one horizontal line, four vertical lines on the left side, and two vertical lines on the right side is secured on the upper side of the image data. Insert. In other words, blank cells for one horizontal line, four vertical lines on the left side, and two vertical lines on the right side are provided above the image data. In this way, when the target cell is within 2 cells from the right end of the image data (FIG. 14A), within 4 cells from the left end (FIG. 14B), Even in the case of being at the upper end (FIG. 14C), the end cell of the image data can be encoded without performing special processing. If the neighborhood is referred to as it is, since blank cells are provided at both ends of the two-dimensional image data, encoding can be performed normally without performing special processing.

On the other hand, there is a method in which a three-dimensional cell structure is subjected to space-filling scanning with a single stroke, converted into one-dimensional data, and compressed by referring to other cells in the space (see, for example, Patent Document 2). ).
Patent No. 3622043 (Registered December 3, 2004) JP 2002-122537 A (published April 26, 2002) ITU-T (International Telecommunication Union Telecommunication standardization sector) WHITE BOOK Digital still image compression coding related recommendation (New Japan ITU Association) Recommendation T.82 WTSC-93 (WORLD TELECOMMUNICATION STANDARDIZATION CONFERENCE), Helsinki, 1-12 MARCH 1993

  However, when a blank area is provided at the end of the binary image data stored in the memory, a blank cell must be inserted so as to sandwich the line for each horizontal line of the image data stored in the memory space. For this purpose, a memory area different from the one storing the original two-dimensional image is secured, and one line of the image is stored in the memory area while sandwiching both ends of the horizontal line of the image data between blank cells. You must take a method such as copying one by one. This wastes time and memory. Further, when compressing an image, it is necessary to skip blank cells in the memory without being compressed, and this also wastes time.

  On the other hand, when space filling scanning is performed with a single stroke, scanning takes time, and a memory area for storing the data subjected to space filling scanning is required. In addition, on one-dimensional data created by space-filling scanning, complicated calculations are required to determine neighboring cells. More specifically, prior to cell coding, as shown in FIG. 15, when two-dimensional image data is provided (S111), the rearrangement of image data cells (spatial replenishment) is performed according to a space replenishment algorithm. Scanning) is performed (S112). In this space supplement scanning, for example, the cells are continuously arranged as shown by the dotted line in FIG. 16A so that all the cells of the image data are drawn with one stroke. That is, as shown in FIG. 16B, the odd-numbered horizontal lines in the two-dimensional image data shown in FIG. 16A (continuous cells from right to left in FIG. 16A, ie, white lines) Next to even lines (in the cell corresponding to the arrow), the horizontal lines (continuous cells from right to left in FIG. 16A) are stored so as to be reversed (continuous from left to right). Dimensional data. Note that each square cell in FIG. 16A is a cell. In FIG. 16A, the horizontal line consists of 4 cells, whereas in FIG. 16B, one line consists of 7 cells.

  Thus, when encoding the data which has been subjected to the spatial supplement scanning and becomes one-dimensional data, the following must be performed. Here, it is assumed that the spatial supplement scanning is performed on the two-dimensional image data shown in FIG. 17A to become one-dimensional data shown in FIG. In FIGS. 17A to 17D, each square cell is a cell. In the one-dimensional data shown in FIG. 17B, the address of a neighboring cell to be referred to when the target cell at a certain address x is encoded is determined as follows. As shown in FIG. 17A, when the target cell is the rightmost cell of the horizontal line of the even row in the original image data, the addresses of neighboring cells forming the context are x + 1, x + 2, x + 3, and x + 4, respectively. , X + ((number of cells in original image data horizontal line) -1), x + ((number of cells in original image data horizontal line) -2), x-1, x-2, x-3, x-4. It becomes. Here, as shown in FIG. 7A, the number of cells (line width) of the horizontal lines of the original image data is 12.

  Further, it is assumed that the spatial supplement scanning is performed on the two-dimensional image data shown in FIG. 17C, and the one-dimensional data shown in FIG. In the one-dimensional data shown in FIG. 17D, the address of the neighboring cell to be referred to when the target cell at a certain address x is encoded is determined as follows. As shown in FIG. 17C, when the target cell is on the horizontal line of the even-numbered row of the original image data and is located inside the 2 cells from the right end and inside the 4 cells from the left end, the following is performed. become. That is, the addresses of neighboring cells forming the context are x + 1, x + 2, x + 3, x + 4, x− (in the original image data, the number of cells from the right end cell of the horizontal line where the target cell exists to the cell right next to the target cell ( 17 (c), the number of cells from the head cell of the white arrow described as line 2 to the cell to the right of the target cell, and the right edge thereafter. (Referred to as the number of cells from the cell to the right adjacent cell) × 2 + 1, x− (the number of cells from the rightmost cell to the right adjacent cell) × 2, x− (the number of cells from the rightmost cell to the right adjacent cell) × 2 −1, x− (the number of cells from the rightmost cell to the right neighbor cell) × 2-2, x− (the number of cells from the rightmost cell to the right neighbor cell) × 2−3, x− (from the rightmost cell) The number of cells up to the right adjacent cell) × 2−4. In FIG. 17C, the number of cells from the right end cell to the right adjacent cell is three.

  Thus, when trying to refer to a fixed context such as JBIG's two-line template on one-dimensional data created by space-filling scanning with a single stroke, where in the original two-dimensional image data? It is necessary to determine a neighboring cell by recognizing whether it is a certain cell (for example, the rightmost cell). Since the method of determining the address of the neighboring cell differs depending on the position of the target cell in the original two-dimensional image data, a complicated calculation must be performed. For these reasons, in the encoding compression processing using the one-stroke space filling scan, it takes time for the encoding processing, and the compression speed becomes slow.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide a compression processing apparatus, a compression processing method, a compression processing apparatus control program, and a computer reading that can perform high-speed encoding / compression processing with a small storage capacity. It is to realize a possible recording medium.

  In order to solve the above-described problem, a compression processing apparatus according to the present invention compresses compression target data in which cells are two-dimensionally arranged by encoding while sequentially selecting encoding target cells. The cell addresses are such that each row of cells in the data to be compressed is continuous in the same direction, and in each adjacent two rows, the last cell in one row is continuous with the first cell in the other row. Address management means, and encoding means for encoding the encoding target cell based on a plurality of neighboring cells at a predetermined relative address position determined by the address of the encoding target cell in the address order, It is characterized by having.

  In order to solve the above-described problem, the compression processing apparatus control method according to the present invention compresses the compression target data in which the cells are arranged in two dimensions by sequentially selecting the encoding target cells. In the control method of the compression processing apparatus, each row formed by the cells of the compression target data is continuous in the same direction, and in each adjacent two rows, the last cell of one row is continuous with the first cell of the other row. An address management step for managing the address of the cell, and encoding the encoding target cell based on a plurality of neighboring cells at a predetermined relative address position determined by the address of the encoding target cell in the address order And an encoding step.

  According to the above configuration and the above method, in compression target data in which cells are arranged in units of cells, the rows formed by the cells are continuous in the same direction, and each of the two adjacent rows has one row. The cell address is managed so that the last cell is continuous with the first cell in the other row. This is a module that receives data in which cells are arranged two-dimensionally in a compression processing device. In data created by storing data in a storage area in the order in which data arranged in two dimensions is received (such as page data) ) The original two-dimensionally arranged data cells are continuous in the same direction in each row, and in each adjacent two rows, the last cell in one row is continuous with the first cell in the other row. It uses the management of cell addresses, that is, management in a one-dimensional storage area.

  When the encoding target cell is compressed, the encoding target cell is encoded based on a plurality of neighboring cells at a predetermined relative address position determined by the address of the encoding target cell in the order of addresses. Here, the “predetermined relative address position” is an address position uniquely determined with respect to the address of the encoding target cell. Specifically, for example, it is assumed that the address of the encoding target cell is n. The addresses of neighboring cells are, for example, n-1, n-2, n-3, n-4, n- (number of cells in one row of compression target data) +2, n- (one row of compression target data). Number of cells) +1, n- (number of cells in one row of compression target data), n- (number of cells in one row of compression target data) -1, n- (number of cells in one row of compression target data)- 2, n− (number of cells in one row of compression target data) −3. The number of cells in one row of the compression target data is determined to be one for the compression target data. In this case, even if the encoding target cell has an address different from n, the address of a neighboring cell of the encoding target cell having another address can be obtained by applying the address to the above-described n. In this way, a cell that is always obtained by the same calculation regardless of which address the encoding target cell has is called a neighboring cell at a predetermined relative address position determined by the address of the encoding target cell.

  Conventionally, when encoding is performed using a neighboring cell at a relative position of the encoding target cell so as to use a two-line template in JBIG compression, when the data to be compressed is accessed beyond the right end, for example. Therefore, blank cells were necessary on the left and right of the image data. In this case, when trying to add a blank cell to the image data, the blank cell is added while copying the received image data, or when receiving, the data is received line by line and a blank cell is added. Must. In order to add a blank cell, it must be performed when image data is first stored in a storage area, and cannot be performed by a module that performs compression alone.

  However, the present invention does not require a blank cell. In the configuration and the method of the present invention, when performing encoding based on neighboring cells, a plurality of neighboring cells at a predetermined relative address position determined based on the address of the encoding target cell are read. Neighboring cells do not protrude from the image data. In the present invention, for example, the address of the rightmost cell in the compression target data in which the cell is two-dimensionally arranged as a unit is continuous with the address of the leftmost cell that is the head of the next row. Managed. Therefore, in the data to be compressed in which cells are arranged two-dimensionally, if the part accessed as a neighbor cell protrudes from the rightmost cell of the data to be compressed, the left end of the line immediately below the line to be accessed by protruding. Can be read out as neighboring cells. For this reason, even if the storage capacity is small enough to store the compression target data, there is no problem such as illegal access or a compression result different from the original compression result.

  As described above, according to the configuration and the method of the present invention, it is possible to reduce the storage capacity used for encoding, and it is not necessary to skip blank cells so as not to encode blank cells. The speed can be increased.

  Conventionally, when encoding is performed after two-dimensional data is written in one stroke and space-filled scanning is performed and converted into one-dimensional data, scanning takes time, and space-filling scanning can be performed. In addition, a memory area for storing one-dimensional data is necessary. In addition, on one-dimensional data created by space-filling scanning, complicated calculations are required for determining neighboring cells. This is because, when one-dimensional data created by one-stroke space-filling scanning is to be referred to a fixed neighborhood shape such as JBIG's two-line template, the image in the original two-dimensional image data is used. It was because it was necessary to recognize that it was an end.

  However, according to the present invention, for encoding, only a plurality of neighboring cells at predetermined relative address positions determined based on the address of the encoding target cell are used for the encoding target cell as described above. There is no need to recognize the position of the encoding target cell in the compression target data in which the cell is two-dimensionally arranged in units of the original cell and change the calculation for obtaining the cell address for each position. . Therefore, the address of the neighboring cell can be easily obtained, and reading of the neighboring cell is facilitated. Also, no space filling scan is required. Therefore, the speed of the compression process can be increased by these amounts. In the present invention, it is not necessary to perform space-filling scanning with a single stroke, and it is not necessary to store one-dimensional data obtained by performing space-filling scanning.

  As described above, according to the configuration and the method described above, when compressing compression target data in which the cell is a unit and the cell is arranged two-dimensionally, neighboring cells are excluded from the compression target data at the left and right ends of the compression target data. It is not necessary to determine whether or not to protrude, and it is not necessary to provide blank cell regions on both the left and right ends of the image data. Therefore, there is no need for such an operation, and it is possible to compress the data in which the given cell is arranged in two dimensions in units of a given cell as it is. Therefore, compression processing can be performed at high speed with a small storage capacity.

  In addition to the above configuration, the compression processing apparatus according to the present invention further comprises data storage means for storing the compression object data in a recording area following the blank data consisting of blank cells, and the address management means includes each cell of blank data. And the address of the cell may be managed so that the last cell of the blank data follows the first cell of the compression target data.

  According to the above configuration, blank data composed of blank cells is stored in the storage area before the data to be compressed. At this time, the address of blank data is managed so that each cell is continuous and the last cell of the blank data is continuous with the first cell of the compression target data. For this reason, when the cell forming the uppermost row of the compression target data is encoded, the neighboring cell does not protrude beyond the blank cell. Therefore, even in the uppermost encoding target cell of the compression target data, a neighboring cell can be easily obtained and encoded based on this neighboring cell, and there is no need for special processing.

  In the compression processing apparatus according to the present invention, in addition to the above configuration, the data storage means can secure a neighboring cell used for encoding the encoding target cell to be encoded first in the compression target data. The blank cells may be stored in the number.

  According to the above configuration, the number of blank cells is the same as the number of neighboring cells that can be used for encoding the encoding target cell to be encoded first in the compression target data by the encoding means. Sometimes there are only blank cells that are read at least once. For this reason, only the storage capacity necessary for encoding can be stored in the storage area while minimizing the storage capacity. Therefore, the storage capacity used for encoding can be suppressed.

  In the compression processing apparatus according to the present invention, in addition to the above configuration, the compression target data may be image data, and the address management unit may manage the address of the pixel of the image data as the cell.

  According to the above configuration, the pixel address is such that image data consisting of pixels is continuous in the same direction in each row, and in each adjacent two rows, the last pixel in one row is continuous with the first pixel in the other row. The encoding target pixel is encoded with reference to a plurality of neighboring pixels at predetermined relative address positions determined based on the address of the encoding target pixel in the order of addresses, and the image data is compressed. Therefore, image data can be compressed at high speed with a small storage capacity.

  Here, the cell may not correspond to one pixel but may correspond to, for example, two pixels. Of course, the compression target data is not limited to image data, and any data may be used as long as the data is obtained by arranging the cells in units of cells in two dimensions. The cell may be any cell as long as the data to be compressed can be divided into units to be encoded.

  In addition, the compression processing apparatus according to the present invention may be realized by a computer. In this case, the compression processing apparatus is realized by the computer by causing the computer to operate as each unit in the compression processing apparatus. A processing apparatus control program and a computer-readable recording medium recording the compression processing apparatus control program also fall within the scope of the present invention.

  According to these configurations, the same operational effects as those of the compression processing device can be realized by causing the computer to read and execute the compression processing device control program.

  As described above, the compression processing apparatus according to the present invention is continuous in the same direction in each row formed by the cells of the compression target data, and the last cell in one row is the other row in each of two adjacent rows. Address management means for managing the address of the cell so as to be continuous with the first cell of the cell, and a plurality of encoding target cells in a predetermined relative address position determined based on the address of the encoding target cell in the address order Encoding means for encoding using the neighboring cells.

  According to the above configuration, in data to be compressed in which cells are arranged in two dimensions in units of cells, each row formed by the cells is continuous in the same direction, and each adjacent two rows are cells at the end of one row. Cell addresses are managed so that is continuous with the first cell in the other row. This is a module that receives data in which cells are arranged two-dimensionally in a compression processing device. In data created by storing data in a storage area in the order in which data arranged in two dimensions is received (such as page data) ) The original two-dimensionally arranged data cells are continuous in the same direction in each row, and in each adjacent two rows, the last cell in one row is continuous with the first cell in the other row. It uses the management of cell addresses, that is, management in a one-dimensional storage area.

  When the encoding target cell is compressed, the encoding target cell is encoded based on a plurality of neighboring cells at predetermined relative address positions determined based on the address of the encoding target cell in the order of addresses. . Here, the “predetermined relative address position” is an address position uniquely determined with respect to the address of the encoding target cell. In this way, a cell that is always obtained by the same calculation regardless of which address the encoding target cell has is called a neighboring cell at a predetermined relative address position determined by the address of the encoding target cell.

  In the above configuration, no blank cell is required. In the above configuration, when encoding is performed based on the neighboring cell, a plurality of neighboring cells at a predetermined relative address position determined based on the address of the encoding target cell are read, so the neighboring cell is read from the image data. There is no such thing as protruding. This is because, in the above configuration of the present invention, for example, the address of the rightmost cell in the compression target data in which the cell is arranged in two dimensions is the address of the leftmost cell that is the head of the next row. Managed continuously. For this reason, in the compression target data in which the cells are arranged in two dimensions in units of cells, if the part accessed as a neighboring cell protrudes from the rightmost cell of the compression target data, it is one more than the line that is to be accessed. The cell at the left end of the next row can be accessed and read as a neighbor cell. For this reason, even if the storage capacity is small enough to store the compression target data, there is no problem such as illegal access or a compression result different from the original compression result.

  As described above, according to the above configuration of the present invention, the storage capacity used for encoding can be reduced, and the process of skipping blank cells so as not to encode blank cells is not required. Can be planned.

  Further, according to the above configuration of the present invention, for the encoding, a plurality of neighboring cells at a predetermined relative address position determined based on the address of the encoding target cell as described above are encoded. It is only necessary to use it, and it is necessary to recognize the position of the encoding target cell in the compression target data in which the cell is arranged in two dimensions with the original cell as a unit, and to change the calculation for obtaining the cell address for each position. There is no. Therefore, the address of the neighboring cell can be easily obtained, and reading of the neighboring cell is facilitated. Also, no space filling scan is required. Therefore, the speed of the compression process can be increased by these amounts. In the present invention, it is not necessary to perform space-filling scanning with a single stroke, and it is not necessary to store one-dimensional data obtained by performing space-filling scanning.

  As described above, according to the above configuration of the present invention, when compressing the compression target data in which the cell is arranged in units of two cells, neighboring cells are excluded from the compression target data at the left and right ends of the compression target data. It is not necessary to determine whether or not to protrude, and it is not necessary to provide blank cell regions on both the left and right ends of the image data. Therefore, there is no need for such an operation, and it is possible to compress the data in which the given cell is arranged in two dimensions in units of a given cell as it is. Therefore, compression processing can be performed at high speed with a small storage capacity.

  One embodiment of the present invention is described below with reference to FIGS.

  The compression processing apparatus 1 according to the present embodiment compresses each cell (pixel) by arithmetically encoding each cell (pixel) with respect to image data (compression target data) shown in two dimensions, and compressed image data (code) obtained thereby. Output). Here, the cell may not correspond to one pixel but may correspond to, for example, two pixels. Of course, the compression target data is not limited to two-dimensional image data, and any data may be used as long as the data is obtained by arranging the cells in units of cells. The cell may be any cell as long as the data to be compressed can be divided into units to be encoded.

  The compression processing apparatus 1 encodes (reads out) the encoding target cell with reference to the neighboring cells forming the context. The determination of neighboring cells to be read will be described in detail later. In the compression processing in the compression processing apparatus 1 of the present embodiment, arithmetic coding is performed. In arithmetic encoding, the encoding result is represented by a small number from 0 to 1. The dominant probability symbol is a symbol predicted from the vicinity of the encoding target cell, and the inferior probability symbol is a symbol different from the prediction result. As shown in FIG. 8, the dominant probability symbol has appeared by dividing the region from 0 to 1 into the probability of appearance of the dominant probability symbol (MPS appearance probability) and the appearance probability of the inferior probability symbol (LPS appearance probability) as the first interval. In this case, the operation is repeated with the appearance probability of MPS as the next interval, and when the inferior probability symbol appears, the appearance probability of LPS as the next interval. The compression result is the lower end of the appearance probability of the dominant probability symbol when this operation is completed for all input bits.

  As shown in FIG. 1, the compression processing apparatus 1 includes a data storage unit (address management unit, data storage unit) 2, a memory (storage area) 3, a context generation unit 22, an acquisition unit 23, an arithmetic coding unit (encoding unit). ) 24, a recording unit 25, and a register 26.

  When image data (two-dimensional image data) is input to the compression processing device 1, the data storage unit 2 stores the two-dimensional image data in the memory 3. At this time, in each row formed by the cells of the original two-dimensional image data, cells are continuous in the same direction, and in each adjacent two rows, the cell at the end of one row is continuous with the head cell of the other row. Manage addresses. This means that the two-dimensional image data is stored in a one-dimensional memory having continuous addresses. In the present embodiment, the cell corresponds to the pixel of the two-dimensional image data. A cell does not correspond to one pixel, but may correspond to, for example, two pixels. Further, the data storage unit 2 manages the addresses not only of the input data stored in the memory 3 but also the compression target data originally in the compression processing apparatus 1 as described above.

  The data storage unit 2 stores blank data composed of blank cells in the memory 3 before the image data. Then, the cell addresses are managed so that the cells of the blank data are continuous and the last cell of the blank data follows the first cell of the compression target data.

  When blank cells are stored in this way, when the cells forming the uppermost row of the two-dimensional image are encoded, neighboring cells do not protrude beyond the blank cells. Therefore, even in the uppermost part of the two-dimensional image, neighboring cells can be easily obtained, and encoding can be performed without the need for special processing. Note that it is not necessary to prepare a blank cell for the first line of the horizontal line of a two-dimensional image, if the process of considering a neighboring cell as a blank is performed as in the conventional case.

  The state of the image data stored in this way is shown in FIG. In FIG. 2, image data is stored in a cell written as X, and a cell written as 0 is a blank cell. The blank cell area added at this time does not affect the compression result. In FIG. 2, a blank cell may or may not be prepared for “?”. Without preparing blank cells in “?” In FIG. 2, the following arithmetic coding unit 24 can secure the number of neighboring cells used for coding the coding target cell to be coded first in the two-dimensional image data, When blank cells are stored in the memory, only blank cells that are read at least once during encoding are prepared. Therefore, only the storage capacity necessary for encoding can be kept to a minimum. Therefore, the storage capacity used for encoding can be suppressed.

  Next, management of the address of the data storage unit 2 will be described. In the two-dimensional image data, addresses are managed so that each row is continuous in the same direction, and in each adjacent two rows, the last cell of one row is continuous with the first cell of the other row. This will be described using a specific example.

  For example, assume that there is two-dimensional image data shown in FIG. Here, each square cell in FIG. 3A is a cell. One row is a horizontal line in FIG. 3A and corresponds to continuous cells from right to left, that is, cells represented by white arrows. As shown in FIG. 3B, the data storage unit 2 is arranged so that the cells in the first row (line 1) in the two-dimensional image data shown in FIG. Then, the addresses are managed so that the cells in the next row (line 2) are arranged in order from left to right. The same applies to the third and subsequent rows. In this way, addresses are managed so that each row continues in the same direction, and the leftmost cell of the next row continues after the rightmost cell of a certain row. In FIG. 3A, the horizontal line consists of 4 cells, whereas in FIG. 3B, one line consists of 7 cells. The two-dimensionally arranged data is continuous in the same direction in each row, and in each adjacent two rows, the cell address is such that the last cell in one row is continuous with the first cell in the other row. Yes.

  The memory 3 is a storage means for storing input image data.

  The context generation unit 22 extracts, from the image data stored in the memory 3, a pixel (encoding target cell) to be encoded and a context (neighboring cell) that is an array of a plurality of pixels in the vicinity (neighboring). . In the compression processing apparatus 1, not only the input bit string up to the previous input signal but also the state of neighboring cells is used for prediction. That is, the compression processing apparatus 1 has a function of estimating the appearance probability of the dominant probability symbol and the inferior symbol for each combination of states that can be taken by a plurality of neighboring cells in the prediction.

  The context generation unit 22 selects and reads the encoding target cell. The selection of the encoding target cell is sequentially performed in the order of the address of each cell of the image data stored in the memory 3. Further, the context generation unit 22 reads a context that is a neighboring cell of the target cell from the input image data. When reading this context, in the image data in which the addresses are managed and stored as described above, a plurality of neighboring cells at a predetermined relative address position determined by the address of the encoding target cell are read.

  Here, the “predetermined relative address position” is an address position uniquely determined with respect to the address of the encoding target cell. Specifically, for example, it is assumed that the address of the encoding target cell is n. The addresses of neighboring cells are n-1, n-2, n-3, n-4, n- (number of cells in one row of compression target data) +2, n- (number of cells in one row of compression target data). +1, n- (number of cells in one row of compression target data), n- (number of cells in one row of compression target data) -1, n- (number of cells in one row of compression target data) -2, n- (Number of cells in one row of data to be compressed) -3. Here, the number of cells in one row of the two-dimensional image data is determined as one in the two-dimensional image data. In this case, even if the encoding target cell has an address different from n, the address of a neighboring cell of the encoding target cell having another address can be obtained by applying the address to the above-described n. In this way, a cell that is always obtained by the same calculation regardless of which address the encoding target cell has is called a neighboring cell at a predetermined relative address position with respect to the encoding target cell.

  Therefore, as shown in FIG. 5, for example, even when the target cell is at the right end of the original two-dimensional image data (FIG. 5A), it is in the fourth cell from the left end (FIG. 5B). Or even at the top of the image (FIG. 5C), if the address of the target cell is n, the addresses of neighboring cells are n-1, n-2, n-3, n-4, n- (Number of cells in one row of compression target data = 8) +2, n-8 + 1, n-8, n-8-1, n-8-2, n-8-3, all are obtained.

  Then, the context generation unit 22 outputs the extracted encoding target cell and neighboring cells forming the context to the acquisition unit 23. Further, when receiving the cell increment instruction from the code processing unit 29 of the arithmetic coding unit 24, the context generation unit 22 extracts the next encoding target cell and context from the image data stored in the memory.

  Next, the acquisition unit 23 will be described. The acquisition unit 23 acquires a dominance probability symbol (MPS) as a predicted value from the probability estimation table 27 of the recording unit 25 that records the dominance probability symbol for the arrangement (context) of surrounding pixels.

  Here, the probability estimation table 27 includes the relationship between the context and the MPS value expected from the context, as shown in FIG. The probability estimation table 27 also includes the relationship between the context and the index ST of the LPS interval estimated value (LSZ) corresponding to this context. The LSZ will be described later. Using the input context, the acquisition unit 23 accesses the probability estimation table 27 of the recording unit 25 to obtain the index ST. When obtaining the MPS and the index ST, the acquiring unit 23 outputs the pixel value (input bit value; input symbol), MPS, and index ST to the arithmetic coding unit 24. As described above, the LPS interval of the encoding target bit (symbol) from the value of the neighboring pixel (neighboring cell, context) of the encoding target pixel (encoding target cell, pixel corresponding to the encoding target bit) (this) In this case, an estimation step for estimating the estimated value of the LPS interval is executed by the acquisition unit 23 and the probability estimation table 27. That is, the acquisition unit 23 and the probability estimation table 7 function as an estimation unit that executes the estimation step.

  The arithmetic code unit 24 generates a real number x as encoded data from the input pixel value, MPS, and index ST. The arithmetic code unit 24 includes a code processing unit 29 and a preprocessing unit 30.

  The code processing unit 29 selects an appropriate processing unit from among the processing units 29 a to 29 c in accordance with the selection signal input from the preprocessing unit 30. Also, encoded data is generated by the selected processing unit using the input pixel value, MPS, and ST. The code processing unit 29 includes an MPS processing unit 9a that encodes a dominant probability symbol, an LPS processing unit 9b that encodes an inferior probability symbol, and a RENORM processing unit 9c that performs a normalization process.

  The preprocessing unit 30 generates a selection signal according to the pixel value, MPS, and ST input from the acquisition unit 23 and outputs the selection signal to the code processing unit 29. The preprocessing unit 30 includes a selection unit 30a.

  In order to generate a selection signal, the selection unit 30a compares the input pixel value (encoding target symbol) with the MPS expected from the context, and determines whether the encoding target symbol is a dominant probability symbol. Determine whether it is an inferior probability symbol. The selection unit 30 a generates a selection signal according to the input and outputs the selection signal to the code processing unit 29.

  In the recording unit 25, a probability estimation table 27 and a probability estimation table 28 are recorded. Note that the probability estimation table 27 and the probability estimation table 28 are completely different. The table can also be translated into a table, but in the Japanese translation of ITU-T T.82 (ITU-T WHITE BOOK Digital Still Image Compression Encoding Recommendation Recommendation T.82). It will be used as it is.

  The probability estimation table 27 is a table for storing the dominant probability symbol MPS for each context and the index ST of the LSZ.

  The dominant probability symbol takes a value of 0 or 1. The index ST of LSZ takes any value from 0 to 112 in JBIG. In addition, since the context is 10 bits while the index is 7 bits, there are different contexts having the same index. For example, the LSZ index ST of the context “00000000” and the LSZ index ST of “0000000001” may both be 13. This LSZ index ST is a value that has appeared so far at the position of the pixel of interest for each combination status (context) of 0 or 1 values (symbols) that appear in neighboring cells in the JBIG probability estimation table 8 described later ( This is used to represent the prediction of the appearance probability of a value (symbol) that will appear at the target pixel from the history of the symbol.

  The probability estimation table 28 is a table for obtaining the LSZ, the next ST, and the like from the LSZ index ST as shown in Table 24 of Non-Patent Document 1 shown in FIGS. In order to obtain LSZ from the probability estimation table 8, the index ST of the LSZ obtained from the probability estimation table 7 is used. This probability estimation table 8 is not rewritten even during the encoding process. That is, the probability estimation table 8 is only referred to and is not rewritten.

  Further, the compression processing apparatus 1 includes a register 26 as a working memory. The register 26 is a memory that can be accessed at high speed. The register 26 includes an Areg 26a and a Creg 26b.

  The Areg 26a is a storage area for holding (accumulating) the value Areg of the interval size. Further, Creg 26b is a storage area for holding a low value Creg of the interval. Areg 26a and Creg 26b are integers and have fixed precision.

  Note that the context generation unit 22 and the acquisition unit 23 may be included in the arithmetic code unit 24. The encoding process in the arithmetic coding unit 24 is not limited to the above, and any process may be used as long as the encoding target cell is encoded using a neighboring cell. .

  Next, compression processing in the compression processing apparatus 1 will be described based on the flowchart of FIG. When the two-dimensional image data (page data) is input to the compression processing device 1, the data storage unit 2 stores the input image data in the memory, as shown in FIG. Two-dimensional image data is stored (S149). At this time, the cell addresses are managed so that each row of the two-dimensional image data is continuous in the same direction, and in each adjacent two rows, the last cell of one row is continuous with the first cell of the other row. The In FIG. 2, a cell described as X is an area in which image data is stored, and a cell written as 0 is a blank cell in which image data is not stored. The blank cell added at this time does not affect the compression result. In the compression processing of the present embodiment, as shown in FIG. 10A, only two-dimensional image data is given in S149. However, in the conventional compression processing using blank cells, data in which blank cells are inserted on the left and right sides of image data is input. Alternatively, in the case of performing the space filling scan in the conventional one-stroke drawing, as shown in FIG. 15, when the two-dimensional image data is given, the rearrangement of the cells of the image data (the space supplement scanning is performed according to the space supplement algorithm. ) Had to be done.

  Next, the image data stored in the memory 3 is divided into at least one stripe (S150). This stripe is a unit for performing compression processing, and consists of one or more horizontal lines (rows) in the original two-dimensional image data.

  Next, individual stripe processing is performed inside the stripe loop (LOOP_A) of S150 to S164. The processing for one stripe is started from S150, and when the stripe processing (processing for repeating cell encoding) in S150-S164 stripe loop (LOOP_A) is completed, the stripe loop (LOOP_A) in S150-S164 is completed. Returning to step S150, the processing of the next stripe starts from step S151. When the processing for all stripes is completed by the loop from S150 to S164, encoding of the entire image data is completed. Here, in the conventional compression processing, it was necessary to divide into stripes and perform processing in a stripe loop. However, in the compression processing apparatus 1 of this embodiment, it can be regarded as one-dimensional data with continuous addresses. So it is not necessary to divide into stripes. Of course, it can be divided.

  There is a cell loop (LOOP_B) in the stripe loop (LOOP_A). In the cell loop (LOOP_B), the encoding process of each cell is performed in S151 to S162. When the processing of one cell is completed, the process returns to the beginning S151 of the cell loop (LOOP_B), and the next cell is encoded from S152.

  Then, encoding is performed for the encoding target cell in the cell loop (LOOP_B) of S151 to S162, and when processing of all the pixels in one stripe is completed, processing of one stripe is completed. Next, the processing of S152 to S164 will be described.

  In step S152, the context generation unit 22 selects an encoding target cell (target cell) from the image data managed in the cell address and stored in the memory 3. The selection of the encoding target cell is sequentially performed according to the address. In S <b> 152, the context generation unit 22 reads neighboring cells (contexts) from the image data whose cell addresses are managed and stored in the memory 3. This neighboring cell is located at a predetermined relative address position with respect to the encoding target cell. Therefore, regardless of which address the encoding target cell has, the address of a neighboring cell with respect to the encoding cell is always determined by the same calculation. And it transmits to the acquisition part 23 (S152). In addition, the context generation unit 22 transmits the read target cell and neighboring cells to the acquisition unit 23. In the compression processing of this embodiment, in S152, as shown in FIG. However, in the conventional compression processing, as shown in FIG. 13, it is necessary to perform each processing after determining whether or not a neighboring cell protrudes from the image.

  In S153, the acquisition unit 23 accesses the probability estimation table 27 of the recording unit 25 according to the target cell and context transmitted from the context generation unit 22, and estimates the probability of the dominant probability symbol and the inferior probability symbol for the context. Gets the value index ST.

In S154, in the preprocessing unit 30 of the arithmetic coding unit 24, the selection unit 30a determines whether or not the coding target cell is an MPS according to the input coding target cell value and the MPS. When it is determined in S154 that the target cell is MPS, the MPS processing unit 29a performs MPS processing (CODEMPS: S155 to S158) in the code processing unit 29 of the arithmetic coding unit 24. On the other hand, when it is determined in S154 that the target cell is LPS, in the code processing unit 29 of the arithmetic coding unit 24, the LPS processing unit 29b performs LPS processing (CODELPS: S159 to S161).

  In S155, the MPS processing unit 29a accesses the probability estimation table 28 of the recording unit 25, acquires the probability estimation value of the inferior probability symbol using the index ST of the probability estimation value of the inferior probability symbol, and according to this Then, Areg 26a and Creg 26b of the register 26 are rewritten. In S156, the MPS processing unit 29a determines whether or not the value of Areg 26a is less than 0.5. If it is less than 0.5, the process proceeds to S157, and if it is 0.5 or more, the process proceeds to S162. In S157, the MPS processing unit 29a instructs the RENORM processing unit 29c, and the RENORM processing unit 29c normalizes the Areg 26a and Creg 26b of the register 26, and outputs the obtained encoded data as necessary. . In S158, the MPS processing unit 29a updates the probability estimation table 28 of the recording unit 25, and proceeds to S162.

On the other hand, in S159, the LPS processing unit 29b accesses the probability estimation table 28 of the recording unit 25, acquires the probability estimation value of the inferior probability symbol using the index ST of the probability estimation value of the inferior probability symbol, In response, Areg 26a and Creg 26b of the register 26 are rewritten. In S160, the LPS processing unit 29b performs LPS and MPS switching and code output for Areg 26a and Creg 26b of the register 26 according to the value of SWITCH in the probability estimation table. That is, the LPS processing unit 29b determines whether or not to perform the replacement depending on whether or not the SWITCH value of the probability estimation table corresponding to the index of the LSZ is 1. In S161, the LPS processing unit 29b updates the probability estimation table 28 of the recording unit 25, and proceeds to S162.

  In S162, the code processing unit 29 instructs the context generation unit 22 to increment the cell, and the loop for the cell ends. By performing such a procedure on all the cells of the stripe, the stripe is compressed. In S163, it is determined whether or not the processing for the stripe has been completed. If the processing has been completed, the encoded data is output from the register 26 (flash processing: S163). In this way, before proceeding to compression of the next stripe, a portion of the compression result so far that has not been output is output. In S164, the context generation unit 22 selects a new stripe. If compression of all stripes is complete, compression of the entire image is complete. If all of the input image data has been encoded, the process ends.

  The compression processing apparatus 1 has the above-described configuration, and the data storage unit 2 is continuous in the same direction in each row formed by cells (pixels) of the original two-dimensional image data, and one in each of the two adjacent rows. The address of the cell (pixel) is managed so that the cell (pixel) at the end of the row is continuous with the cell (pixel) at the beginning of the other row. The arithmetic encoding unit 24 encodes the encoding target cell based on a plurality of neighboring cells at predetermined relative address positions determined by the encoding target cell in the order of the addresses.

  In the compression processing apparatus 1, the memory 3 includes a row of original two-dimensional image data cells that are continuous in the same direction, and in each adjacent two rows, the last cell of one row is the head of the other row. The address of the cell is managed so as to be continuous with the other cell, and the compression target data is stored. This is a module for receiving two-dimensional image data in a compression processing device. In data created by storing in a storage area in the order of reception of two-dimensional image data (such as page data), a cell of two-dimensional image data Each row is continuous in the same direction, and in each adjacent two rows, the cell address is such that the last cell of one row is continuous with the first cell of the other row, that is, one-dimensionally It is stored in the memory 3.

  And when compressing an encoding object cell, with respect to the encoding object cell of the two-dimensional image data in which the address is managed, a predetermined relative address position determined by the address of the encoding object cell in the address order Are encoded based on a plurality of neighboring cells. Here, the “predetermined relative address position” is an address position uniquely determined with respect to the address of the encoding target cell. That is, for any encoding target cell at any address, a cell whose address is determined by the same calculation is called a neighboring cell at a predetermined relative address position determined by the address of the encoding target cell.

  Further, the compression processing apparatus 1 does not require a blank cell as described above. In the compression processing apparatus 1, when encoding is performed based on neighboring cells, a plurality of neighboring cells at a predetermined relative address position are read, so that the neighboring cells do not protrude from the image data. In the compression processing apparatus 1, for example, the address of the rightmost cell in the two-dimensional image data in which the cell is two-dimensionally arranged in units of cells is the address of the leftmost cell that is the head of the next row. Managed continuously. Therefore, in the two-dimensional image data that is arranged in units of cells and is secondarily arranged by the cells, if the part accessed as a neighboring cell extends beyond the rightmost cell of the data to be compressed, the line that is to be accessed by protruding is used. The cell at the left end of the next lower row can be accessed and read as a neighbor cell. For this reason, even if the storage capacity is small enough to store the compression target data, there is no problem that the storage area is illegally accessed or the compression result is different from the original compression result. As described above, according to the configuration and the method of the present invention, it is possible to reduce the storage capacity and to eliminate the process of skipping blank cells so as not to encode blank cells. Can do.

  Further, in the compression processing apparatus 1, for encoding, it is only necessary to use a plurality of neighboring cells at a predetermined relative address position determined by the address of the encoding target cell as described above with respect to the encoding target cell. There is no need to recognize the position of the encoding target cell in the compression target data arranged two-dimensionally in the original cell unit and change the calculation for obtaining the cell address for each position. Therefore, the address of the neighboring cell can be easily obtained, and reading of the neighboring cell is easy. Also, no space filling scan is required. Therefore, the speed of the compression process can be increased by these amounts. In the present invention, it is not necessary to perform space filling scanning with a single stroke, and it is not necessary to perform space filling scanning, so it is not necessary to store one-dimensional data obtained by performing space filling scanning, and a compression module. It can be dealt with only.

  As described above, according to the compression processing apparatus 1, when compressing two-dimensional image data in which the cells are arranged in units of cells, neighboring cells are outside the image data at the left and right ends of the two-dimensional image data. It is not necessary to determine whether or not to protrude, and it is not necessary to provide blank cell regions on either the left or right end of the two-dimensional image data. Therefore, there is no need for such an operation, and the given two-dimensional image data can be compressed as it is. Therefore, compression processing can be performed at high speed with a small storage capacity.

  In the present embodiment, as described above, the compression processing apparatus 1 models binary image data with a Markov information source, predicts an encoded symbol according to the state of its surrounding pixels, and performs arithmetic encoding according to the prediction result. It was described as using a method. However, the present invention is not limited to any compression format that compresses the difference between the value predicted from the vicinity of the encoding target cell and the value of the encoding target cell for data arranged in two dimensions in units of cells. Can be applied to. As described above, the present invention is not limited to the above-described embodiment, and various modifications can be made within the scope of the claims. That is, embodiments obtained by combining technical means appropriately changed within the scope of the claims are also included in the technical scope of the present invention.

  Each block of the compression processing apparatus 1 may be configured by hardware logic, or may be realized by software using a CPU as follows.

  That is, the compression processing apparatus 1 includes a CPU (central processing unit) that executes instructions of a control program that realizes each function, a ROM (read only memory) that stores the program, and a RAM (random access memory) that expands the program. And a storage device (recording medium) such as a memory for storing the program and various data. An object of the present invention is a recording medium on which a program code (execution format program, intermediate code program, source program) of a control program of the compression processing apparatus 1 which is software that realizes the functions described above is recorded so as to be readable by a computer. This can also be achieved by supplying to the compression processing apparatus 1 and reading and executing the program code recorded on the recording medium by the computer (or CPU or MPU).

  Examples of the recording medium include a tape system such as a magnetic tape and a cassette tape, a magnetic disk such as a floppy (registered trademark) disk / hard disk, and an optical disk such as a CD-ROM / MO / MD / DVD / CD-R. Card system such as IC card, IC card (including memory card) / optical card, or semiconductor memory system such as mask ROM / EPROM / EEPROM / flash ROM.

  The compression processing apparatus 1 may be configured to be connectable to a communication network, and the program code may be supplied via the communication network. The communication network is not particularly limited. For example, the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication. A net or the like is available. Further, the transmission medium constituting the communication network is not particularly limited. For example, even in the case of wired such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL line, etc., infrared rays such as IrDA and remote control, Bluetooth ( (Registered trademark), 802.11 wireless, HDR, mobile phone network, satellite line, terrestrial digital network, and the like can also be used. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  The present invention may be expressed as an image compression method described below. That is, a method of compressing a two-dimensional image while referring to a neighboring pixel of a pixel to be compressed, wherein the data of the two-dimensional image is stored in a one-dimensional storage area having continuous addresses, and the left end of the image It may be expressed as an image compression method in which compression is performed by referring to pixels in the opposite end portion of the two-dimensional image as the neighboring pixels when compressing the pixels in the portion or the right end portion.

  Further, in the image compression method, when storing the two-dimensional image data in a one-dimensional storage area having continuous addresses, the two-dimensional image may be stored following the blank data.

  In the image compression method, when the data of the two-dimensional image is stored in a one-dimensional storage area having continuous addresses, the size of the blank data area secured before the two-dimensional image is Data obtained by adding three pixels of data to one horizontal line of a two-dimensional image may be used.

  The present invention can be used for compression of data arranged two-dimensionally, and is effective for compression of image data in, for example, a facsimile machine (FAX), a printer, a printer driver of a personal computer, and the like.

It is a block diagram which shows the principal part structure of the compression processing apparatus of this embodiment. It is a figure which shows the state of the storage area in the said compression processing apparatus. (A) is a figure of two-dimensional image data, (b) is a figure for demonstrating management of the address of each cell in the said compression processing apparatus. (A) is a diagram showing the positions of the encoding target cell and neighboring cells in the two-dimensional image data, and (b) is one of the encoding target cell and neighboring cells of FIG. (A) managed by the compression processing apparatus. It is a figure explaining the address in a dimension. (A)-(c) is a figure explaining the neighbor cell read in the said compression processing apparatus. It is a figure which shows an example of the probability estimation table used with the said compression processing apparatus. (A) is a figure which shows a part of example of the probability estimation table used with the said compression processing apparatus, (b) is a figure which shows another part of probability estimation table. It is a conceptual diagram for demonstrating the arithmetic encoding performed with the said compression processing apparatus. It is a flowchart which shows the operation | movement of the encoding process performed with the said compression processing apparatus. (A), (b) is a figure explaining a part of process in the follow chart of FIG. It is a figure for demonstrating a template. It is a figure which shows the area | region which the template designates in the conventional normal compression processing apparatus. It is a figure which shows the flow of a read-out process of the neighbor cell in the conventional normal compression processing apparatus. It is a figure which shows the area | region which the template designates in the conventional compression processing apparatus which provided the blank cell area | region on the both sides of data. It is a figure which shows the flow of the process of the space filling scan in the compression process using the conventional space filling scan. (A) is a figure of two-dimensional image data, (b) is a figure for demonstrating management of the address of each cell in the compression process using the conventional space filling scanning. (A) is a diagram showing the positions of the encoding target cell and neighboring cells in the two-dimensional image data, and (b) is the encoding target cell of FIG. (A) in the compression processing using the conventional space filling scan. It is a figure explaining the one-dimensional address of a neighboring cell, (c) is a figure showing the position of the encoding object cell and neighboring cell in two-dimensional image data, (d) uses the conventional space filling scanning. It is a figure explaining the address in the one dimension of the encoding object cell of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Compression processing apparatus 2 Data storage part (address management means, data storage means)
3 Memory (storage area)
22 Context Generation Unit 23 Acquisition Unit 24 Arithmetic Code Unit (Encoding Unit)
25 Recording unit 26 Register 26a Creg
26b Creg
27 Probability estimation table 28 Probability estimation table 29 Encoding processing unit 29a MPS processing unit 29b LPS processing unit 30 Preprocessing unit 30a Selection unit

Claims (7)

In a compression processing apparatus that compresses compression target data in which cells are arranged in two dimensions in units of cells by sequentially selecting the encoding target cells,
The cell addresses are set to 1 so that each row of cells in the compression target data is continuous in the same direction, and in each adjacent two rows, the last cell of one row is continuous with the first cell of the other row. Address management means to manage by dimension ,
Encoding target cells, in order of address, for the address of the coding target cell, encoding based on the plurality of neighboring cells in a predetermined relative address position that put into a one-dimensional address managed by the address management means Encoding means for
A compression processing apparatus comprising: Data storage means for storing the compression target data in the recording area following the blank data consisting of blank cells,
2. The address management means manages cell addresses so that cells of blank data are continuous and a cell at the end of blank data follows a head cell of the compression target data. The compression processing apparatus according to 1.   The data storage means stores the blank cells as many as the number of neighboring cells that can be used for encoding the encoding target cell to be encoded first in the compression target data. The compression processing apparatus according to claim 2. The compression target data is image data,
The compression processing apparatus according to claim 1, wherein the address management unit manages an address of a pixel of the image data as the cell. In a control method for a compression processing apparatus that compresses data to be compressed in which cells are arranged in a two-dimensional manner by sequentially selecting encoding target cells,
The cell addresses are set to 1 so that each row of cells in the compression target data is continuous in the same direction, and in each adjacent two rows, the last cell of one row is continuous with the first cell of the other row. Address management step to manage by dimension ,
Encoding target cells, in order of address, for the address of the coding target cell, encoding based on the plurality of neighboring cells in a predetermined relative address position that put into a one-dimensional address managed by the address management step An encoding step,
A control method for a compression processing apparatus, comprising:   A control program for operating the compression processing apparatus according to claim 1, wherein the computer functions as each unit in the compression processing apparatus. A computer-readable recording medium on which the compression processing control program according to claim 6 is recorded.

Images