JP3862969B2 - Image compression processing apparatus, image compression processing method, and program for causing computer to execute the method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image compression processing apparatus, an image compression processing method, and a program for causing a computer to execute the method.
[0002]
[Prior art]
Currently, in order to reduce the amount of hardware and time required for processing digital data, a technique for compressing data has become widespread. A technique for compressing image data (image compression) is a particularly important technique for image data that is generally known to have a large amount of data.
[0003]
As an image compression technique for compressing a binary image, a one-dimensional compression run length image compression and an MH code used for a facsimile are known. In recent years, it has also been proposed to further increase the compression rate of image data by further repeating the processing after the run length processing of image data (processing for obtaining the run length of image data). However, in any of these techniques, depending on the image data to be processed, the data amount may increase after the image data compression (expansion after compression).
[0004]
When expansion after compression can occur in image data, an image compression processing apparatus that compresses image data must include a memory that can store the maximum amount of data that can be taken by the image data compressed in the image compression processing apparatus. For this reason, there has been a drawback that the cost of the image compression processing apparatus is increased, or the processing speed is lowered due to an increase in the number of memory accesses.
[0005]
In order to solve the above-described drawbacks, the inventions described in JP-A-5-183760 and JP-A-7-264417 encode image data line by line, and the amount of encoded image data is the original. When the amount exceeds the amount of data (image data before encoding), the original data is stored with a flag indicating that expansion after compression has occurred. According to the inventions described in JP-A-5-183760 and JP-A-7-264417, the data amount of the entire image data after encoding does not become larger than the data amount of the entire original data, and compression is performed. Later expansion does not occur.
[0006]
The inventions described in Japanese Patent Application Laid-Open Nos. 9-248943 and 10-117289 do not encode image data line by line, but instead have a plurality of blocks (line direction and direction orthogonal to the line). The original data is stored by setting a flag for the image data of the block that has been encoded in units of (arranged pixels) and expanded after compression.
[0007]
[Problems to be solved by the invention]
However, the above-described technology for encoding image data line by line or block by block is a line memory or block for storing original data for one line or one block and encoded image data. Requires memory. When a line memory or a block memory is added to the image compression processing apparatus, the cost of the image compression processing apparatus may increase.
[0008]
In the technique of encoding image data for each line or for each block, the image is encoded by dividing the image into lines or blocks. For this reason, even if the image is blank over several lines or several blocks, this region is encoded. For this reason, a sufficient effect of increasing the compression rate may not be obtained.
[0009]
The present invention has been made to solve the above-described problems, and minimizes the configuration for storing encoded data and original data, enabling lower cost and higher speed processing. It is a first object of the present invention to provide an image compression processing apparatus, an image compression processing method, and a program for causing a computer to execute the method.
[0010]
The present invention also efficiently compresses an image that is blank over several lines or blocks, and provides an image compression processing apparatus, an image compression processing method, and a program for causing a computer to execute the method with a higher compression rate. The second purpose is to provide it.
[0011]
[Means for Solving the Problems]
  In order to solve the above-described problems and achieve the object, an image compression processing device according to claim 1 compresses image data as word data cut out every predetermined number of bits. BecauseOf the plurality of continuous word data thus cut out, one word data is obtained by a run length method based on the first bit of the next word data following the word data.Encoding means for encoding;The code amount of one word data and the next word data following the word data among the plurality of extracted word data is measured.Whether the code amount of the word data is larger than the threshold value by comparing the code amount measured by the code amount measurement unit and the code amount measured by the code amount measurement unit with a preset code amount threshold value Code amount comparison means for determiningAnd the code amount comparing means determines that the code amount of the second word data encoded next to the first word data which is one of the word data is larger than the threshold value. In this case, the encoding means cuts a run length extending between the first word data and the second word data when encoding the first word data.It is characterized by that.
[0012]
  According to the invention described in claim 1,When it is determined that the code amount of the second word data is larger than the threshold value, the one length over the first word data and the second word data can be cut when the first word data is encoded. .
[0013]
  An image compression processing device according to the invention of claim 2 is provided.The code word data encoded by the encoding means and the original word data that is the word data before being encoded by the encoding means are held, and the code amount comparison means determines the code amount of the word data. If it is determined that the code amount is smaller than the threshold value, the code word data is output. On the other hand, if the code amount comparison unit determines that the code amount of the word data is larger than the threshold value, the original word data is output. Comprising word data selection output means for outputtingFeatures.
[0014]
  According to the invention described in claim 2,The code amount of the image data is compared with a threshold value for each word data, and if it is determined that the code amount of the word data is smaller than the threshold value, the encoded word data is output, while the code of the word data If it is determined that the amount is larger than the threshold value, the original word data before encoding can be output.
[0015]
  An image compression processing device according to the invention of claim 3 is provided.The encoding means obtains a repetition number that is the number of times the run length appears repeatedly in the image data and encodes the word data based on the repetition number.It is characterized by that.
[0016]
  According to the third aspect of the present invention, the number of repetitions, which is the number of times the run length appears repeatedly in the image data, is obtained, and the image data is obtained based on the number of repetitions.For each word dataIn the encoding configuration, the code amount of the image data is compared with a threshold value for each word data, and if it is determined that the code amount of the word data is smaller than the threshold value, the encoded word data is output. On the other hand, when it is determined that the code amount of the word data is larger than the threshold value, the original word data before encoding can be output.
[0019]
  Claim4The image compression method according to the invention described inAn image compression processing method for compressing image data as word data cut out for each predetermined number of bits, wherein one word data of the cut out continuous word data is continued to the word data An encoding step of encoding by the run length method based on the first bit of the next word data, and a holding step of holding the same word data as the word data encoded in the encoding step in a state before encoding A code amount measuring step of measuring a code amount of one word data and the next word data following the word data among the plurality of extracted word data, and the code measured in the code amount measuring step The amount is compared with a preset code amount threshold value to determine whether the code amount of the word data is larger than the threshold value. An amount comparison step, wherein the amount of code of the second word data encoded next to the first word data which is one of the word data by the code amount comparison step is greater than the threshold value. The encoding step cuts a run length over the first word data and the second word data when encoding the first word data. .
[0020]
  This claim4According to the invention described inWhen it is determined that the code amount of the second word data is larger than the threshold value, the one length over the first word data and the second word data can be cut when the first word data is encoded. .
[0021]
  Claim5The program according to the invention described inA program for causing a computer to execute an image compression processing method for compressing image data as word data cut out for each predetermined number of bits, wherein the computer includes a plurality of pieces of the extracted continuous word data, An encoding procedure for encoding one word data by a run length method based on the first bit of the next word data following the word data, one word data among the plurality of extracted word data, The code amount measurement procedure for measuring the code amount of the next word data following the word data, and the code amount measured by the code amount measurement procedure is compared with a preset code amount threshold value. A code amount comparison procedure for determining whether or not the code amount of word data is larger than the threshold value, the code amount comparison procedure Accordingly, when it is determined that the code amount of the second word data encoded next to the first word data which is one of the word data is larger than the threshold value, the encoding procedure is In addition, when the first word data is encoded, a run length extending between the first word data and the second word data is cut.
[0022]
  This claim5According to the invention described inWhen it is determined that the code amount of the second word data is larger than the threshold value, the one length over the first word data and the second word data can be cut when the first word data is encoded. .
[0023]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments 1 and 2, which are preferred embodiments of an image compression processing apparatus according to the present invention, will be described below in detail with reference to the accompanying drawings. FIG. 1 is a block diagram for explaining the overall configuration of the image compression processing apparatus according to the first embodiment.
[0024]
(Embodiment 1)
The image compression processing apparatus according to the first embodiment, for example, receives word data created by cutting out data such as image data for each predetermined bit, and performs compression processing for each word data. The word data will be described with reference to FIG. FIG. 2A is a diagram showing data before being cut out into word data. In (a), the upper part shows the address of the data before being cut out, and the lower part shows the data stored in each address. FIG. 2B is a diagram showing 8-bit word data D obtained by cutting out the data shown in FIG. 2A at the positions indicated by P1 and P2.
[0025]
The image compression processing apparatus according to the first embodiment sequentially inputs word data and processes them in parallel. For this reason, for example, when the illustrated image compression processing apparatus inputs the word data D as “current word data”, the word data arranged immediately after the data D is input as “next word data”.
[0026]
The image compression processing apparatus in FIG. 1 includes a run-length encoding processing apparatus 101, code amount measurement apparatuses 102a and 102b, and a PASS code header creation apparatus 103. The run-length encoding processing apparatus 101 compresses word data and next word data in parallel. The code amount measuring device 102a measures the code amount when each word data is encoded by the run length encoder, and the code amount measuring device 102b measures the code amount of the next word data. Further, the PASS code header creation device 103 creates a PASS code that is added to the word data as a header.
[0027]
Further, the image compression processing apparatus in FIG. 1 stores the word data encoded by the run-length encoding processing apparatus 101 as a code, a register 105 that stores the PASS code generated by the PASS code header generation apparatus 103 as a word before compression. A register 106 that stores data (original data), a comparator 104a that compares a preset threshold th with a code amount of word data measured by the code amount measuring apparatus 102a, and a threshold th and code amount measurement Comparator 104b that compares the code amount of the next word data measured by apparatus 102b, word data encoded based on the result of comparison between the code amount and threshold th, and PASS code as code data And a MUX (Multiplexer) 107 for outputting.
[0028]
FIG. 3 is a diagram for explaining codes handled in the above-described image compression processing device. FIG. 3A shows a format when word data is encoded by a run-length encoder, and FIG. , Shows the format of the PASS code.
[0029]
Next, an outline of processing performed in the above-described configuration will be described with reference to the flowchart of FIG. First, the image compression processing apparatus of the first embodiment inputs the next word data to the code amount measuring device 102b and measures the code amount of this word data (step S401). The measured code amount is compared with the threshold value th by the comparator 104b (step S402). The comparison result is output to the run-length encoding processing apparatus 101.
[0030]
Further, word data (hereinafter simply referred to as word data) arranged immediately before the next word data is input to the code amount measuring device 102a, the run-length encoding processing device 101, and the PASS code header creating device 103. The run-length encoding processing apparatus 101 receives the MSB (Most Significant Bit (or Byte)) of the word data and the next word data, and encodes the input word data (step S403). At this time, the code amount measured by the code amount measuring apparatus 102b is compared with the threshold value th in the comparator 104b, and a signal indicating whether or not the code amount exceeds the threshold value th is displayed as a run-length encoding processing apparatus. 101.
[0031]
When the input signal indicates that the code amount of the next word data exceeds the threshold th, the next data is PASS-encoded, so the run length of the word data extends over the next word data In addition, the word data is cut out and encoded by the word boundary. If the code amount of the next word data does not exceed the threshold value, only the closed run length (the portion not including the run length over the next boundary) is encoded. The word data encoded by the run-length encoding processing apparatus 101 is stored in the register 105 (step S404).
[0032]
Next, the PASS code header creation device 103 determines whether the 0th bit of the data to be PASS encoded is 1 or 0 in order to clarify 0 and 1 of the start of the run length after the PASS code. Investigate. Then, the setting of the PASS code is selected by 1 or 0 of the 0th bit (step S405).
[0033]
  FIGS. 5A and 5B are diagrams for more specifically describing step S405. (A), (b) shows the example of the PASS encoding of word data, respectively, (II) in (a), (b) is a figure which shows the encoding of this Embodiment. Moreover, (I) is a figure which shows the encoding generally performed for a comparison. The top row of (a), (b), (I), and (II) shows the pixel data, and the next row shows the run length when each pixel data is cut into 8-bit word data. Further, the lower part of the run length indicates the code amount of each word data, and the lowermost part indicates the code assigned to each code amount. In the first embodiment, the threshold value th of PASS encoding is set to a code amount 3, and the code amount isGreater than 3Word data is PASS encoded.
[0034]
In the example shown in (a), the data of Dp1 has a code amount of 6, and has a code amount of 3 or more. For this reason, Dp1 is PASS encoded as shown in (II) of FIG. Whether or not to perform PASS encoding is determined in the word data after cutting away the run length between the word data before and after the word data to be PASS encoded.
[0035]
Further, the PASS code uses a code of 8 or 9 to distinguish whether the extracted word data starts with 0 or 1. In the example of (a), since Dp1 starts from 1, 9 is used for the PASS code (9D2).
[0036]
  In the example shown in (b), the data of Dp2 has a code amount of 5, and has a code amount of 3 or more. For this reason, Dp2 is PASS-encoded as in (II). In the example of (b), since Dp2 starts from 0, 8 is used for the PASS code (82D).
[0037]
The PASS code created by the PASS code header creation device 103 is attached to the original data as a header, and is stored in the register 106 together with the original data (step S406). The code amount measuring apparatus 102a measures the code amount of the word data (step S407). The measured code amount is input to the comparator 104a, and is compared with the threshold value th by the comparator 104a. The comparison result (whether or not the code amount is larger than the threshold th) is input to the MUX 107 (step S408).
[0038]
As a result of the comparison by the comparator 104a, the word data and the PASS code encoded by the run-length encoding processing apparatus 101 are input to the MUX 107. Based on the comparison result of the comparator 104a, the MUX 107 outputs the original data to which the PASS code is added as code data when the code amount is larger than the threshold value th. When the code amount is smaller than the threshold value th, the encoded word data is output as code data (step S409).
[0039]
Next, the configuration shown in FIG. 1 will be described in more detail.
(1) Run-length encoding processing device
FIG. 6 is a block diagram for explaining the run-length encoding processing apparatus 101. The run-length encoding processing apparatus 101 includes a run-length processing apparatus 601 and an encoding processing apparatus 602. The run-length processing device 601 is configured to obtain the run length of word data in parallel. Since the word data in the first embodiment is 8 bits, the run length processing device 601 obtains a maximum of 8 run lengths in parallel. Also, the encoding processing device 602 encodes the run length obtained by the run length processing device 601 based on the format shown in FIG.
[0040]
FIG. 7 is a diagram for explaining the configuration of the run length processing apparatus 601. As illustrated, the run length processing device 601 includes run counters 701 a and 701 b, a Left run value generation device 702, a Right run value generation device 703, a run value synthesis device 705, a cumulative addition processing device 708, and a register 709. . The cumulative addition processing device 708 includes an adder 706 and an adder 707. Hereinafter, each configuration of the run-length encoding processing device will be described.
[0041]
(Run counter)
The run counters 701a and 701b are 8-bit word data (IDOT) input from an image reading device (not shown) such as a scanner, and the upper 4 bits of IDOT from the bit (NDOT) at the head of the next word data. (IDOT [7: 4]), run count for processing IDOT [3] as NDOT, lower 4 bits of IDOT (IDOT [3: 0]), and run counter for processing NDOT A run length, a signal (mask value) indicating whether or not the run length is valid, and a run length (RUNX) to be carried over next are obtained.
[0042]
That is, the run counters 701a and 701b receive, for example, the upper 4 bits of 8-bit word data generated by cutting out image data and the first bit data of the next word data, and output a maximum of 4 run lengths. To do. The illustrated RUN3 to LUN0 (LRUN is a run value generated by the Left run value generation device, and RRUN is a run value generated by the Right run value generation device) indicate run lengths. DMASK3 to DMASK0 are flags indicating whether or not the run lengths of RUN3 to LUN0 are valid. The run count processing performed in the run counters 701a and 701b is shown as flowcharts in FIGS.
[0043]
(Left run value generator)
The Left run value generation device 702 receives a maximum of four run lengths and mask values (LDMASK3-0) from the run counter 701a, and a mask value (RDMASK3-0) from the run counter 701b, and has a maximum 8 bits of 8 bits. Find the run length. The processing by the Left run value generation device 702 is shown in FIG.
[0044]
(Right run value generator)
The right run value generation device 703 receives a maximum of four run lengths and mask values (RDMASK3-0) from the run counter 701b, and a mask value (LDMASK3-0) from the run counter 701a, and has a maximum of 8 bits and a maximum of 8 bits. Find the run length. The processing performed by the Right run value generation device 703 is shown in FIG.
[0045]
(Mask value synthesizer)
The mask value synthesizing device 704 receives mask values from the run counter 701a and the run counter 701b, and synthesizes the mask values to obtain eight mask values of 8 bits. Processes performed by the mask value synthesis apparatus 704 are shown in FIGS.
[0046]
(Run value synthesizer)
The run value synthesizer 705 receives a maximum of 8 run lengths from the Left run value generator 702 and the Right run value generator 703, respectively, and combines them to generate a maximum of 8 run lengths (step S1401). . The processing performed by the run value synthesis device 705 is shown in FIG. RUNX in the figure is the run length to be carried over next, AFL is 1 when all 4 bits are 0 or 1, and NDOT (the first bit of the next data) is the same value. Flag.
[0047]
(Cumulative addition processing device)
The cumulative addition processing device 708 includes adders 706 and 707. When the AFL of the word data is 1, the adder 706 adds the carry-over run length obtained by the run value synthesizer 705 and the carry-over run length stored in the register 709 (RBRUN7), and carries over again. Set as the total run length (RBRUN7). The adder 707 adds the RUN 7 input first and the total carry-over run length, and obtains a run length considering the run length carried over so far.
[0048]
(Encoding processing device)
FIG. 18 is a diagram for explaining the encoding processing device 602. The encoding processing device 602 receives RRUN 7 to 0 data, DMASK 7 to 0, a clock (CLK) signal, and the like from the register 709, and outputs encoded word data (CODE). 19 and 20 are flowcharts showing the encoding process performed by the encoding processing device 602. Note that the processing shown in the flowcharts of FIGS. 19 and 20 encodes word data according to the format shown in FIG.
[0049]
(2) Code amount measuring device
FIG. 21 is a block diagram for explaining the code amount measuring apparatus 102a (code amount measuring apparatus 102b). The code amount measuring device 102a includes a run length processing device 1801 and a code amount measuring processing device 1802. The code amount measuring apparatus 102a and the code amount measuring apparatus 102b have the same configuration and operate in the same manner. Therefore, in this specification, only the code amount measuring apparatus 102a will be described, and the description of the code amount measuring apparatus 102b will be omitted.
[0050]
(Run length processing equipment)
FIG. 22 is a block diagram for explaining the run-length processing device 1801. The run-length processing device 1801 has the same configuration as the run-length processing device 601 shown in FIG. 7, and includes run counters 701a and 701b, a left run value generation device 702, a right run value generation device 703, a mask value synthesis device 704, A run value synthesis device 705 and a register 709 are provided. However, the run-length processing device 1801 does not include the cumulative addition processing device 708, and inputs the inverted LSB of the word data immediately before the run counter 701b so that the run length of the word data is always closed. This is different from the run-length processing apparatus 601.
[0051]
FIG. 23 is a block diagram for explaining the code amount measurement processing device 1802 shown in FIG. Similarly to the encoding processing device 602 shown in FIG. 18, the code amount measurement processing device 1802 receives RRUN 7 to 0 data, DMASK 7 to 0, a clock (CLK) signal, and the like from the register 709 and encodes the encoded word. Data (CODE) is output. The code amount measurement processing performed by the code amount measurement processing device 1802 is shown in FIGS.
[0052]
(3) PASS code header creation device
FIG. 25 is a flowchart for explaining processing of the PASS code header creation device 103. As shown in the figure, the PASS code header creation device 103 determines whether or not the first bit of the word data is 0 (step S2201), and when the first bit is 0 (step S2201: Yes). The header as the PASS code is set to 9 (step S2202). If the result of determination in step S2201 is that the first bit is not 0 (that is, 1) (step S2201: No), the header is set to 8 (step S2203).
[0053]
(Embodiment 2)
Next, a second embodiment of the present invention will be described.
FIG. 26 is a block diagram for explaining an image compression processing apparatus according to the second embodiment of the present invention. Note that the image compression processing apparatus according to the second embodiment has the same configuration as the image compression processing apparatus according to the first embodiment. It shall be omitted.
[0054]
The image compression processing device in FIG. 26 includes a run length / repetitive coding processing device 2301, code amount measurement devices 2302a and 2302b, and a PASS code header creation device 103. The run-length / repetitive encoding processing device 2301 compresses word data and next word data in parallel. However, unlike the word data of the first embodiment, the word data of the second embodiment is cut out with a run length that crosses the boundary between the word data. For this reason, the run-length / repetitive encoding processing device 2301 is configured to be able to cope with repetitive processing across word data.
[0055]
The code amount measuring device 2302a measures the code amount when the word data of the second embodiment is encoded by the run length encoder, and the code amount measuring device 2302b is the code of the next word data of the second embodiment. Measure the amount. Further, the PASS code header creation device 103 creates a PASS code that is added to the word data as a header.
[0056]
Further, the image compression processing apparatus of FIG. 26 includes comparators 104a and 104b, registers 105 and 106, and a MUX 107 similar to those in the first embodiment. The comparator 104a compares the threshold th with the code amount of the word data measured by the code amount measuring device 2302a. The comparator 104b compares the threshold th with the code amount of the next word data measured by the code amount measuring device 2302b. The register 105 stores the encoded word data as a code, and the register 106 stores the pre-encoded word data (original data) attached with the PASS code. The MUX outputs either encoded word data or original data as code data based on the result of comparison between the code amount and the threshold value th.
[0057]
FIG. 27 is a diagram for explaining codes handled in the above-described image compression processing apparatus. FIG. 27A shows a format when word data is encoded by a run-length encoder, and FIG. The format of the repetitive code is shown, and (c) shows the format of the PASS code.
[0058]
Next, an outline of processing performed in the above-described configuration will be described with reference to the flowchart of FIG. First, the image compression processing apparatus of the second embodiment inputs the next word data to the code amount measuring device 2302b and measures the code amount of this word data (step S2501). The measured code amount is compared with the threshold value th by the comparator 104b (step S2502). The comparison result is output to the run-length / repetitive encoding processing device 2301.
[0059]
The word data is input to the code amount measuring device 2302a, the run-length / repetitive encoding processing device 2301, and the PASS code header creating device 103. The run-length / repetitive encoding processing device 2301 receives the MSB of the input word data and the next word data, and encodes the input data (step S2503). At this time, the code amount measured by the code amount measuring device 2302b is compared with the threshold value th in the comparator 104b, and a signal indicating whether the code amount exceeds the threshold value th is run-length / repetitively encoded. Input to the processor 2301.
[0060]
When the input signal indicates that the code amount of the next word data exceeds the threshold th, the next data is PASS-encoded, so the run length of the word data extends over the next word data In addition, the word data is cut out and encoded by the word boundary. If the code amount of the next word data does not exceed the threshold value, only the closed run length (the portion not including the run length over the next boundary) is encoded. The word data encoded by the run-length encoding processing apparatus 101 is stored in the register 105 (step S2504).
[0061]
Next, the PASS code header creation device 103 determines whether the 0th bit of the data to be PASS encoded is 1 or 0 in order to clarify 0 and 1 of the start of the run length after the PASS code. Investigate. Then, the setting of the PASS code is selected by 1 or 0 of the 0th bit (step S2505).
[0062]
FIG. 29 is a diagram for explaining step S2505 more specifically, and shows an example of PASS encoding of word data. Moreover, (II) is a figure which shows the encoding of Embodiment 2, (I) is a figure which shows the encoding generally performed for a comparison. The top row of (I) and (II) shows pixel data, and the next row shows the run length when each pixel data is cut into 8-bit word data. Further, the lower part of the run length indicates the code amount of each word data, and the lowermost part indicates the code assigned to each code amount. Also in the second embodiment, the threshold value th for PASS encoding is set to code amount 3, and word data having a code amount of 3 or more is PASS encoded.
[0063]
As shown in the figure, in the second embodiment, the run length over the word data arranged immediately after the PASS-encoded word is cut off. In the example shown in FIG. 29, the iterative process is added to the word data immediately before the word to be PASS encoded. Note that the run length between the word data before and after the PASS-encoded word data is also cut off in the second embodiment.
[0064]
The PASS code created by the PASS code header creation device 103 is attached to unencoded data as a header and stored in the register 106 together with the data (step S2506). The code amount measuring apparatus 2302a measures the code amount of word data (step S2507). The measured code amount is input to the comparator 104a, and is compared with the threshold value th by the comparator 104a. The comparison result (whether or not the code amount is larger than the threshold th) is input to the MUX 107 (step S2508).
[0065]
As a result of the comparison by the comparator 104a, the word data and the PASS code encoded by the run-length repetition / encoding processing device 2301 are input to the MUX 107. Based on the comparison result of the comparator 104a, the MUX 107 outputs the original data with the PASS code added as a header as code data when the code amount is larger than the threshold th. When the code amount is smaller than the threshold value th, the encoded word data is output as code data (step S2509).
[0066]
Next, the configuration shown in FIG. 26 will be described in more detail.
(1) Run-length repetition / encoding processing device
FIG. 30 is a block diagram for explaining the run-length / repetitive encoding processing device 2301. The run-length / repetitive encoding processing device 2301 includes a run-length processing device 2701, an iterative processing device 2702, and an encoding processing device 2703. The run length processing device 2701 is configured to obtain the run length of word data in parallel. Since the word data in the first embodiment is 8 bits, the run length processing device 2701 obtains a maximum of 8 run lengths in parallel. In addition, the iterative processing device 2702 obtains the number of repetitions of the run length in the run length processing device 2701. The encoding processing device 2703 encodes the run length obtained by the run length processing device 2701 and the number of repetitions obtained by the repetition processing device 2702 based on the format shown in FIG.
[0067]
FIG. 31 is a diagram for explaining the configuration of the iterative processing device 2702. As illustrated, the iterative processing device 2702 includes an iterative flag generator 2801, an iterative number generator 2802, a register 2803, a NEXT run length generator 2804, and a register 2805. The iterative processing device 2702 determines whether or not the plurality of run lengths obtained by the run length processing device 2701 are equal to the run length two previous to each run length, and the repetitive flag (RFL7˜ 0) is generated. Then, from the pattern of the repetition flag (RFL7 to 0) and the number of repetitions (REP) input so far, the number of repetitions to date (WREP7 to 0) and the number of repetitions to the next pattern (WREP) are obtained. . Hereinafter, each configuration of the iterative processing device 2702 will be described.
[0068]
(Repetition flag generator)
The repetitive flag generation device 2801 inputs a plurality of run lengths, a plurality of run length mask values, a run length of the previous pattern, and a run length of the previous pattern from the run length processing device 2701, It is determined whether each run length input from the processing device 2701 matches the run length input two times before. Based on the result of the determination, repeat flags RFL7-0 are output. The above processing performed by the repetition flag generation device 2801 is shown as a flowchart in FIG.
[0069]
As shown in FIG. 32, the repetition flag generator 2801 checks the mask value (step S2901). If this value is 0 (step S2901: NO), 2 is set in the repeat flag RFL7 because there is no corresponding run length (step S2903). If it is determined in step S2901 that the mask value is 1 (step S2901: YES), RUN7 is compared with the second run length RF2 after the previous pattern (step S2902).
[0070]
As a result, when RUN7 and RF2 do not match (step S2902: No), 0 is set to the repetition flag RFL7 (step S2904). If it is determined in step S902 that RUN7 and RF2 match (step S2902: Yes), 1 is set in the repeat flag RFL7 (step S2905). The repetition flag generation device 2801 executes the above processing for all bit values DMASK7-0.
[0071]
(Repetition number generator)
The repetition number generation device 2802 receives the repetition flags RF7 to 0 input from the repetition flag generation device 2801 and the current repetition number (REP), generates a plurality of repetition numbers (WREP7 to 0), and outputs them to the register 2803 To do. Processing for obtaining the number of repetitions from the number of repetition flags of the repetition number generation device 2802 is shown as a flowchart in FIGS.
[0072]
As illustrated in FIGS. 33 to 35, the repetition number generation device 2802 obtains a plurality of repetition numbers WREP7 to 0. The number of repetitions is a numerical value that occurs when the repetition stops. The repetition number generation device 2802 first obtains the seventh-bit repetition number WREP7.
[0073]
That is, the iteration number generation device 2802 determines whether or not the iteration number RF7 is 0 (step S3001). If it is 0 (step S3001: Yes), the result of the comparison by the iteration flag generation device 2801 does not match. The number of repetitions up to the present is settled, and the number of repetitions REP up to the present is substituted into the number of repetitions WREP7 (step S3003). If the repetition number RF7 is not 0 (step S3001: No), the repetition number WREP7 is set to 0 (step S3002) assuming that the repetition continues.
[0074]
The value WREP6 of the sixth bit of the repetition number is obtained in steps S3004 to S3006. That is, if the repetition flag RFL7 is 1 and RFL6 is 0 (step S3004: Yes), the repetition number WREP6 is set to 1 (RFL = 1) that is the current repetition number in the repetition number REP thus far. Add and substitute into WREP6 (step S3006). If the repetition flag RFL7 is not 1 and RFL6 is not 0 (step S3004: No), 0 is substituted for the repetition number WREP6 (step S3005). Hereinafter, as in the processing described above, the number of repetitions of the fifth bit to the zeroth bit is obtained.
[0075]
In addition, FIGS. 36 to 40 are flowcharts showing processing for obtaining the number of repetitions (REP) to be continued in the next pattern of the repetition number generation device 2802. As shown in FIGS. 36 to 40, when the repetition flag RFL7 is 2 (step S3101: Yes), the repetition number generation device 2802 cannot determine the run length value with the highest priority, and therefore sets the current repetition number REP. Save (step S3105). If the repetition flag RFL7 is not 2 (step S3101: No), it is determined whether the repetition flag RFL7 is 0 and RFL6 is 2 (step S3102). As a result of the determination, if the repetition flag RFL7 is 0 and RFL6 is 2 (step S3102: Yes), the run length with the highest priority is not a repetition, and since there is no run length thereafter, the repetition number WREP is 0. (Step S3106).
[0076]
If it is determined in step S3102 that the repeat flag RFL7 is 0 and RFL6 is not 2 (step S3102: No), it is determined whether the repeat flag RFL7 is 1 and RFL6 is 2 (step S3103). As a result, if the repeat flag RFL7 is 1 and RFL6 is 2 (step S3103: Yes), the run length with the highest priority is repeat, and there is no run length thereafter, so the repeat number REP at the present time is set. The number of repetitions 1 obtained this time is added and substituted into WREP (step S3107).
[0077]
Next, the repetition number generation device 2802 determines whether or not the repetition flag RFL6 is 0 and RFL5 is 2 (step S3104). As a result of the determination, if the repetition flag RFL6 is 0 and RFL5 is 2 (step S3104: Yes), if the effective termination (here, RFL5 is 2 because RFL5 is 2), the number of repetitions WREP Becomes 0 regardless of the value of RFL7 (step S3108). Thereafter, the repetition number generation device 2802 performs the above processing for all combinations of repetition flags.
[0078]
(NEXT run length generator)
The NEXT run length generation device 2804 is configured to obtain the second run length and the last run length counted from the end of the previous pattern in order to generate the next pattern repetition flag. The processing executed by the NEXT run length generating device 2804 is shown in FIG.
[0079]
As illustrated in FIG. 41, the NEXT run length generation device 2804 determines whether or not the mask (DMASK7) having the highest priority is 1 (step S3401). If DMASK7 is not 1 (step S3401: NO), DMASK7 is 0, and the last value and the last value of the previous pattern are stored assuming that there is no closed run length in this pattern (step S3405). ).
[0080]
Further, when DMASK7 is not 0 (step S3401: YES), the NEXT run length generation device 2804 determines whether DMASK6 is 1 (step S3402). If DMASK6 is not 1, that is, 0 (step S3402: No), there is only one run length. Therefore, the value of RUN7 is substituted into the first run length WREF1 counted from the end of the previous pattern. Further, the value of WREF1 is substituted into the second run length WREF2 counted from the end of the previous pattern (step S3406).
[0081]
Further, the NEXT run length generation device 2804 determines whether DMASK5 is 1 (step S3403). If DMASK5 is not 1 (step S3403: No), it is assumed that DMASK5 is 0 and there are only two closed run lengths in this pattern. Then, the value of RUN6 is substituted into the first run length WREF1 counted from the end of the previous pattern. Further, the value of RUN7 is substituted for the second run length WREF2 counted from the end of the previous pattern (step S3407).
[0082]
Further, the NEXT run length generation device 2804 determines whether DMASK4 is 1 (step S3404). If DMASK4 is not 1 (step S3404: NO), it is assumed that DMASK4 is 0 and there are only three closed run lengths in this pattern. Then, the value of RUN5 is substituted into the first run length WREF1 counted from the end of the previous pattern. Also, the value of RUN6 is substituted into the second run length WREF2 counted from the end of the previous pattern (step S3408).
[0083]
(Encoding processor)
FIG. 42 is a block diagram showing the configuration of the encoding processing device 2703 shown in FIG. The encoding processing device 2703 receives a maximum of 8 run lengths (ORUN7 to 0), the number of repetitions (OREP7 to 0), and a mask value (not shown). Then, the data encoded every 4 bits shown in FIG. 27 is output. However, the encoding processing device 2703 is not limited to the one that encodes data every 4 bits, and may use a format that encodes every bit.
[0084]
43 to 45 are flowcharts for explaining processing performed by the encoding processing device 2703. The flowchart shown in FIG. 43 shows the process performed by the encoding processing apparatus, and FIGS. 44 and 45 are flowcharts for explaining the encoding process in FIG. The encoding processing device 2703 obtains a run length based on the mask values DMASK7 to 0 and encodes it.
[0085]
That is, the encoding processing device 2703 first determines whether or not the value of DMASK7 is 1 (step S3601). If the value of DMASK7 is 1 (step S3601: YES), the run length (run length) is set to RUN7, the repetition number REP is set to REP7, and the run length is encoded (step S3602).
[0086]
When performing encoding, the encoding processing device 2703 first determines whether or not the input run length is 0, as shown in FIGS. 44 and 45 (step S3701). As a result of the determination, if the run length is not 0 (step S3701: NO), it is determined whether the run length is 10 or less (step S3702). If the run length is 10 or less (step S3702: YES), the run length is directly used as a code (step S3703).
[0087]
If the run length value is 10 or more (step S3702: NO), it is determined whether the run length is 43 or less (step S3704). As a result of the determination, if the run length value is 43 or less (step S3704: Yes), it is further determined whether or not the run length value is 27 or less (step S3705).
[0088]
As a result of the determination in step S3705, when the run length value is 27 or less (step S3705: Yes), a 4-bit 1010 is added as a header, and a value obtained by subtracting 11 from the run length is used as a sign (step) S3708). On the other hand, when the run length value is 27 or more (step S3705: No), 4 bits 1011 are added as a header, and a value obtained by subtracting 11 from the run length is used as a code (step S3707).
[0089]
Also, when the run length value is 43 or more (step S3704: NO), the encoding processing device 2703 adds 4 bits of 1100 as a header, and ends the fourth bit of the header through steps S3709 to S3716. Thus, the code length is expressed by a small number of data.
[0090]
Further, if it is determined in step S3701 that the run length is 0 (step S3701: YES), it is determined whether the repeat number REP is 1 (step S3717). As a result of the determination, the code is set to 13 (step S3718). If the repetition number REP is not 1 (step S3717: No), it is determined whether the repetition number REP is 2 (step S3719). If the number of repetitions is 2 (step S3719: Yes), the code is set to 14 (step S3720).
[0091]
If the repetition number REP is not 2 (step S3719: No), the encoding processing device 2703 determines whether the repetition number REP is 3 (step S3721). If the repetition number REP is 3 (step S3721: YES), the code is set to 13 and 14 (step S3722). When the repetition number REP is not 3 (step S3721: No), the encoding processing device 2703 determines whether the repetition number REP is 4 (step S3723). If the repetition number REP is 4 (step S3723: YES), the code is set to 14 and 14 (step S3724).
[0092]
When the number of repetitions is 4 or more (step S3723: No), 4-bit 1111 is added as a header, and the code length is expressed by a small number of data by using the fourth bit of the header as a termination flag in steps S3726 to S3733. .
[0093]
(2) Code amount measuring device
FIG. 46 is a block diagram for explaining the code amount measuring apparatus 2302a (2302b). The code amount measuring device 2302a includes a run length processing device 1801, an iterative processing device 2702, and a code amount measuring processing device 3801. The run-length processing device 1801 and the iterative processing device 2702 are the same as those described above. For this reason, the same code | symbol is attached | subjected and shown and description shall be abbreviate | omitted. The code amount measuring device 2302a and the code amount measuring device 2302b have the same configuration and operate in the same manner. Therefore, in this specification, only the code amount measuring device 2302a will be described, and the description of the code amount measuring device 2302b will be omitted.
[0094]
(Code amount measurement processing device)
The code amount measurement processing device 3801 measures the code length of the word data based on the run length obtained by the run length processing device 1801, the number of repetitions obtained by the repetition processing device 2702, and the format shown in FIG. is there. 47 and 48 show flowcharts of processing performed by the code amount measurement processing device 3801. FIG.
[0095]
【The invention's effect】
  As described above, the invention described in claim 1When the first word data is encoded, the run length over the first word data and the second word data can be cut. Therefore, it is possible to provide an image compression processing apparatus that can obtain a higher compression rate, can be manufactured at lower cost, and can be processed at high speed.
[0096]
  The invention described in claim 2The code amount of image data is compared with a threshold value for each word data, and word data or original word data encoded based on the comparison result is output. For this reason, it is possible to provide an image compression processing apparatus capable of minimizing the configuration for storing the encoded data and the original data and capable of processing at a lower cost and at a higher speed.
[0097]
  Claim3In the invention described in the above, the image data is obtained based on the number of repetitions of the image data.For each word dataIn the encoding configuration, the code amount of the image data is compared with a threshold value for each word data, and word data or original word data encoded based on the comparison result is output. For this reason, it is possible to provide an image compression processing apparatus capable of minimizing the configuration for storing the encoded data and the original data and capable of processing at a lower cost and at a higher speed. In addition, there is an effect that it is possible to efficiently compress an image and to provide an image compression processing device with a higher compression rate.
[0099]
  Claim4The invention described inWhen the first word data is encoded, the run length over the first word data and the second word data can be cut. Therefore, it is possible to provide an image compression processing apparatus that can obtain a higher compression rate, can be manufactured at lower cost, and can be processed at high speed.
[0100]
  Claim5The invention described inWhen the first word data is encoded, the run length over the first word data and the second word data can be cut. Therefore, it is possible to provide an image compression processing apparatus that can obtain a higher compression rate, can be manufactured at lower cost, and can be processed at high speed.
[Brief description of the drawings]
FIG. 1 is a block diagram for explaining an overall configuration of an image compression processing apparatus according to a first embodiment of the present invention.
FIG. 2 is a diagram for explaining word data according to the first embodiment of the present invention.
FIG. 3 is a diagram showing a format when encoding word data according to the first embodiment of the present invention.
FIG. 4 is a flowchart for explaining an outline of processing performed in Embodiment 1 of the present invention;
FIG. 5 is a diagram for explaining a PASS code according to the first embodiment of the present invention.
6 is a block diagram for explaining the run-length encoding processing apparatus shown in FIG. 1; FIG.
7 is a diagram for explaining the configuration of the run-length processing apparatus shown in FIG. 6;
8 is a flowchart showing a run count process performed in the run counter shown in FIG. 7;
FIG. 9 is another flowchart showing a run count process performed in the run counter shown in FIG. 7;
10 is a flowchart showing a run count process performed in the run counter shown in FIG. 7;
11 is a flowchart showing a run count process performed in the run counter shown in FIG. 7;
12 is a flowchart showing processing of the Left run value generation device shown in FIG.
FIG. 13 is a flowchart showing processing of the Right run value generation device shown in FIG. 7;
14 is a flowchart showing processing of the mask value synthesizing apparatus shown in FIG. 7;
15 is another flowchart showing processing of the mask value synthesizing apparatus shown in FIG.
16 is another flowchart showing processing of the mask value synthesizing apparatus shown in FIG. 7;
FIG. 17 is another flowchart showing the process of the run value synthesis apparatus shown in FIG. 7;
18 is a diagram for explaining the encoding processing apparatus illustrated in FIG. 6; FIG.
FIG. 19 is a flowchart showing processing of the encoding processing device.
FIG. 20 is another flowchart showing processing of the encoding processing device.
FIG. 21 is a block diagram for explaining the code amount measuring apparatus shown in FIG. 1;
22 is a block diagram for explaining the run length processing apparatus shown in FIG. 21; FIG.
FIG. 23 is a block diagram for explaining the code amount measurement processing apparatus shown in FIG. 21;
FIG. 24 is a flowchart illustrating a code amount measurement process performed by a code amount measurement processing apparatus.
25 is a flowchart for explaining processing of the PASS code header creation device shown in FIG. 1; FIG.
FIG. 26 is a block diagram for explaining an image compression processing apparatus according to the second embodiment of the present invention;
FIG. 27 is a diagram illustrating a format when encoding word data according to the second embodiment of the present invention.
FIG. 28 is a flowchart for explaining an overview of processing performed in Embodiment 2 of the present invention;
FIG. 29 is a diagram for explaining a PASS code according to the second embodiment of the present invention.
30 is a block diagram for explaining the run-length / repetitive coding processing apparatus shown in FIG. 26;
31 is a diagram for explaining the configuration of the iterative processing apparatus shown in FIG. 30;
32 is a flowchart showing processing performed by the repetition flag generation device shown in FIG. 31. FIG.
FIG. 33 is a flowchart showing processing performed by the repetition number generation device.
FIG. 34 is another flowchart showing processing performed by the repetition number generation device.
FIG. 35 is another flowchart showing processing performed by the repetition number generation device.
FIG. 36 is a flowchart showing a process for obtaining the number of repetitions to be continued in the next pattern performed by the repetition number generation device.
FIG. 37 is another flowchart showing a process for obtaining the number of repetitions continued in the next pattern made by the repetition number generation device.
FIG. 38 is another flowchart showing a process for obtaining the number of repetitions continued in the next pattern made by the repetition number generation device.
FIG. 39 is another flowchart showing a process for obtaining the number of repetitions continued in the next pattern made by the repetition number generation device.
FIG. 40 is another flowchart showing a process for obtaining the number of repetitions continued in the next pattern made by the repetition number generation device.
41 is a diagram showing processing executed by the NEXT run length generation device shown in FIG. 31; FIG.
42 is a block diagram showing a configuration of the encoding processing device shown in FIG. 30. FIG.
43 is a flowchart showing processing performed by the encoding processing device shown in FIG. 30. FIG.
FIG. 44 is another flowchart showing the processing performed by the encoding processing device.
FIG. 45 is another flowchart showing processing performed by the encoding processing device.
46 is a block diagram for explaining the code amount measuring apparatus shown in FIG. 26; FIG.
FIG. 47 is a flowchart showing processing performed by the code amount measuring apparatus.
FIG. 48 is another flowchart showing processing performed by the code amount measuring apparatus.
[Explanation of symbols]
101 Run-length encoding processing apparatus
102a, 102b code amount measuring apparatus
103 PASS Code Header Creation Device
104a, 104b comparator
105, 106, 709, 2803, 2805 registers
601, 1801, 2701 Run length processing apparatus
602, 2703 encoding processing apparatus
701a, 701b Run counter
702 Left run value generator
703 Right run value generator
704 Mask value synthesis apparatus
705 Run value synthesizer
706,707 adder
708 Cumulative addition processing device
1802 Encoding measurement processing device
2301 Run-length / repetitive encoding processing apparatus
2302a, 2302b Code amount measuring apparatus
2702 Repeat processing equipment
2801 Repeat flag generator
2802 Repeat number generator
2804 NEXT run-length generator
3801 Code amount measurement processing apparatus

Claims (5)

An image compression processing apparatus that compresses image data as word data cut out for each predetermined number of bits,
Encoding means for encoding one word data among a plurality of consecutive extracted word data by a run length method based on the first bit of the next word data following the word data ;
Code amount measuring means for measuring the code amount of one word data and the next word data following the word data among the plurality of extracted word data ;
A code amount comparison for determining whether or not the code amount of the word data is larger than the threshold value by comparing the code amount measured by the code amount measuring means with a threshold value of a predetermined code amount Means,
With
When it is determined by the code amount comparison means that the code amount of the second word data encoded next to the first word data that is one of the word data is larger than the threshold value, The image compression processing apparatus , wherein the encoding means cuts a run length over the first word data and the second word data when the first word data is encoded . The code word data encoded by the encoding means and the original word data that is the word data before being encoded by the encoding means are held, and the code amount comparison means determines the code amount of the word data. If it is determined that the code amount is smaller than the threshold value, the code word data is output. On the other hand, if the code amount comparison unit determines that the code amount of the word data is larger than the threshold value, the original word data is output. 2. The image compression processing apparatus according to claim 1, further comprising word data selection output means for outputting . 3. The encoding device according to claim 1, wherein the encoding unit obtains a repetition number that is the number of times the run length repeatedly appears in the image data, and encodes the word data based on the repetition number. 4. Image compression processing device. An image compression processing method for compressing image data as word data cut out for each predetermined number of bits,
An encoding step of encoding one word data out of the plurality of extracted word data by a run length method based on the first bit of the next word data following the word data;
A holding step of holding word data similar to the word data encoded in the encoding step in a state before encoding;
A code amount measuring step of measuring a code amount of one word data and the next word data following the word data among the plurality of extracted word data;
A code amount comparison for comparing the code amount measured in the code amount measuring step with a preset code amount threshold value to determine whether or not the code amount of the word data is larger than the threshold value Process,
Including
When the code amount comparison step determines that the code amount of the second word data encoded next to the first word data that is one of the word data is larger than the threshold value, In the image compression processing method, the encoding step cuts a run length extending between the first word data and the second word data when the first word data is encoded . A program for causing a computer to execute an image compression processing method for compressing image data as word data cut out for each predetermined number of bits,
  On the computer,
  An encoding procedure for encoding one word data out of the plurality of continuous word data extracted by the run length method based on the first bit of the next word data following the word data;
  A code amount measurement procedure for measuring the code amount of one word data and the next word data following the word data among the plurality of extracted word data;
  A code amount comparison for comparing the code amount measured by the code amount measurement procedure with a preset code amount threshold value to determine whether or not the code amount of the word data is larger than the threshold value. Procedure and
  Including
  When it is determined by the code amount comparison procedure that the code amount of the second word data encoded next to the first word data which is one of the word data is larger than the threshold value, The encoding procedure is characterized in that a run length over the first word data and the second word data is cut when the first word data is encoded.

Images