JP2007088687A - Image processing apparatus, image processing method and program thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of obtaining a high compression rate by a comparatively light processing load. <P>SOLUTION: An image processing apparatus 2 divides an input image into image blocks of a predetermined size, and determines a plurality of representative colors for each of the image blocks. The apparatus 2 determines a combination of a unique representative color and a reference representative color so that the number of reference representative colors increases (e. g. variations of the unique representative colors increase) from among the representative colors determined by each of image blocks, and corrects the determined combination of the unique representative color and the reference representative color depending on a predetermined existing coding system. Correction in this case is performed so that the number of reference representative colors is not reduced. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing apparatus that performs color reduction processing on image data.

For example, Patent Document 1 performs a color reduction process on an original image, outputs color information and an index color image, and generates a binary image for each color using color information other than the highest level and the index color image. An image encoding apparatus that performs compression processing by a method such as MMR is disclosed.
JP 2003-309727 A

  The present invention has been made from the above-described background, and an object thereof is to provide an image processing apparatus that achieves a high compression rate with a relatively light processing load.

[Image processing device]
In order to achieve the above object, an image processing apparatus according to the present invention stores image data including a representative color of an image block of a predetermined size and reference information instructing reference to a representative color of another image block. An image data acquisition unit to be acquired; a correction unit that corrects at least one of representative colors and reference information included in at least a part of the image data acquired by the image data acquisition unit; and the correction unit. Coding means for coding image data corrected at least in part.

  Preferably, the correction unit corrects some image data so that representative colors of a plurality of consecutive image blocks have the same value.

  The image processing apparatus according to the present invention includes a block setting unit that sets an image block of a predetermined size in input image data, and an image based on a color included in the image block set by the block setting unit. Representative color determining means for determining a predetermined number or less of representative colors for each block, and color reducing means for generating reduced color image block data using the representative colors determined for each image block by the representative color determining means And reference information generating means for comparing the representative colors determined for each of the plurality of image blocks with each other to generate reference information for designating a reference destination of the representative color, and the color reduction means for each image block. Replacement means for replacing a part of representative colors with reference information generated by the reference information generation means. Encoding means for encoding another representative color of an image block in which some representative colors are replaced by reference information by the replacement means, and the reference information generation means is generated by the encoding means The reference information is generated so that the amount of code to be generated is small.

  Preferably, the reference information generating unit compares the representative color of the image block to be processed with the representative color of the image block around the image block, and generates a reference information. Reference information correction means for correcting the reference information generated by the surrounding reference means in accordance with the encoding method made by the encoding means, and the replacement means is the reference corrected by the reference information correction means A replacement process is performed using the information.

  Preferably, the encoding unit compares the representative color of the image block to be processed with the representative color of the image block at a predetermined position with respect to the image block, and performs code data by predictive encoding processing. And the reference information correcting unit corrects the reference information so that the prediction accuracy in the predictive encoding process is improved.

  Preferably, the reference information correction unit corrects the reference information on the condition that the representative color of the image block to be processed is not referred to by another image block.

[Image processing method]
The image processing method according to the present invention obtains image data including a representative color of an image block of a predetermined size and reference information instructing reference to a representative color of another image block, and a predetermined code At least one of the representative color and the reference information included in at least a part of the acquired image data is corrected, and at least a part of the acquired image data is corrected so that the code amount generated by the conversion method is reduced. Image data is encoded by the encoding method.

[program]
Further, the program according to the present invention includes a step of obtaining image data including a representative color of an image block of a predetermined size and reference information instructing reference to a representative color of another image block; The step of correcting at least a part of the acquired image data so that the amount of code generated by the encoding method is small, and the image data corrected at least partly by the encoding method And causing the computer to execute the encoding step.

  According to the image processing apparatus of the present invention, a high compression rate can be realized with a relatively light processing load.

First, the background will be described in order to help understanding of the present invention.
In order to reduce the code amount of the image data, a color reduction process may be performed. The color reduction processing is to reduce a plurality of colors included in the image data into one color. Here, the color means a combination of pixel values, for example, a combination of R value, G value and B value in an RGB image, or a pixel value in a monochrome image.
For example, in the color reduction process, colors included in an input image (input image) are classified into a plurality of groups, and a representative color (a color for replacing a color belonging to each group) is determined for each classified group. Then, the colors belonging to each group are replaced with the determined representative color. Since colors similar to each other are classified into each group, and the representative color is similar to the color group belonging to each group, the image replaced with the representative color is approximate to the input image.
Thus, the compression efficiency is improved by reducing the colors included in the input image to a predetermined number (the number of groups) of representative colors.

However, if such a color reduction process is performed in consideration of the entire image, the processing load increases. For example, in order to determine the representative color, the storage area for storing the entire image becomes large, or the calculation amount for determining the representative color becomes enormous.
Further, when the representative color is determined based on the entire image, one representative color is determined based on a plurality of distant colors, so that image quality deterioration is likely to be manifested.

Therefore, it is desirable to perform the color reduction processing locally from the viewpoint of processing load and image quality.
FIG. 1 is a diagram illustrating a specific example of local color reduction processing.
As illustrated in FIG. 1A, the image processing apparatus 2 divides the input image into image blocks of a predetermined size (in this example, square blocks of 4 pixels in the main scanning direction and 4 pixels in the sub scanning direction). Then, the colors included in each image block are classified into a plurality of groups (in this example, two of the first group and the second group), representative colors are determined based on the colors included in each group, The colors belonging to each group are replaced with the determined representative color.
In this example, the first representative color is an average value of colors belonging to the first group, and the second representative color is an average value of colors belonging to the second group. Since the color group belonging to the first group is replaced with the first representative color, the pixels having the color of the first group are filled with the first representative color. In addition, since the color group belonging to the second group is replaced with the second representative color, the pixels having the color of the second group are filled with the second representative color.
As described above, by performing the color reduction processing in units of blocks, the processing load is reduced, and image quality deterioration due to the color reduction processing is hardly realized.

As described above, the image data of the image block subjected to the local color reduction processing can be expressed by the pixel values of a plurality of representative colors and a mask for designating the representative colors as illustrated in FIG. 1B. Become. The image data illustrated in FIG. 1B corresponds to 16-pixel image data included in an image block.
In this example, the first representative color and the second representative color each have an R value (8 bits), a G value (8 bits), and a B value (8 bits). .
In the mask of this example, “0” designates the first representative color and “1” designates the second representative color. Since the mask in this example is “0001...”, It indicates that the first three pixels in the scanning direction are the first representative color and the third pixel in the scanning direction is the second representative color. Has been.

In consideration of image continuity, the correlation between representative colors of adjacent image blocks often increases. Using this characteristic, among a plurality of representative colors of an image block to be processed (hereinafter referred to as a target block), a predetermined number of representative colors are referred to adjacent image blocks (hereinafter referred to as surrounding blocks). Replace with information. That is, when the difference between a part of representative colors included in the target block and the representative colors of the surrounding blocks is equal to or less than the reference value, the image processing apparatus 2 refers to the representative color of the target block with respect to the surrounding blocks. Replace with the reference information that indicates
For example, when the first representative color of the block of interest approximates the representative color of the surrounding block adjacent on the left side, the first representative color (24 bits) is set as illustrated in FIG. It can be replaced with reference information (in this example, any of the eight surrounding blocks can be represented by 3 bits because it is a reference destination).
In the following description, the representative color replaced with the reference information (in this example, the first representative color) is referred to as the reference representative color, and the representative color not replaced with the reference information (in this example, the second representative color). ) Is called a unique representative color.

  By limiting the reference destination of the representative color to the unique representative color of the adjacent image block, the range of the color reduction process can be limited to a narrow range. In addition, the data amount can be reduced by replacing a part of the representative color with the reference information.

As described above, the amount of data can be reduced by expressing the image data with the configuration illustrated in FIG. 1C (that is, the configuration including the reference information, the unique representative color, and the mask). In order to reduce the image quality and suppress deterioration in image quality, it is desirable that the unique representative colors of consecutive image blocks have different values as much as possible.
That is, when the variation of the unique representative colors of consecutive image blocks increases, the presence probability of surrounding blocks whose difference from the representative color of the target block is not more than the replacement criterion increases. Further, when the variation of the unique representative colors in the surrounding blocks increases, it becomes possible to refer to the representative colors that are closer to each other, and the deterioration of the image quality is difficult to be realized.

However, the expressions illustrated in FIG. 1C are not necessarily one way. That is, among the plurality of representative colors of the image block, which can be a reference representative color and which is a unique representative color can have a plurality of combinations. For example, among the representative colors illustrated in FIG. 1B, the second representative color may be replaced with the reference information, and the first representative color may be the unique representative color.
Further, when considering the compression efficiency of the unique representative color, the method of determining the reference information and the unique representative color so as to increase the variation of the unique representative color is not always optimal.

Therefore, the image processing apparatus 2 according to the present embodiment determines a combination of the reference representative color and the unique representative color for each image block so that the compression rate is improved by a predetermined encoding method, and uses the determined reference representative color. Substituting with the reference information, the determined unique representative color is encoded by a predetermined encoding method.
More specifically, the image processing apparatus 2 divides input image data into image blocks of a predetermined size, determines a plurality of representative colors for each image block, and among the representative colors determined for each image block. From the above, the combination of the unique representative color and the reference representative color is determined so that the number of reference representative colors increases (that is, the variation of the unique representative color increases), and the determined combination of the unique representative color and the reference representative color is determined. Are corrected according to a predetermined encoding method. The correction in this case is made so as not to reduce the number of reference representative colors.
As a result, the compression efficiency of the unique representative color is improved, so that the data size of the image data can be further reduced.

[Hardware configuration]
Next, a hardware configuration of the image processing apparatus 2 in the present embodiment will be described.
FIG. 2 is a diagram illustrating the hardware configuration of the image processing apparatus 2 to which the image processing method according to the present invention is applied, centering on the control apparatus 21.
As illustrated in FIG. 2, the image processing apparatus 2 includes a control device 21 including a CPU 212 and a memory 214, a communication device 22, a recording device 24 such as an HDD / CD device, an LCD display device or a CRT display device, and a keyboard. A user interface device (UI device) 25 including a touch panel and the like is included.
The image processing apparatus 2 is provided, for example, inside the housing of the printer apparatus 3, and an image processing program 5 (described later) according to the present invention is installed to process image data used for image forming processing.

[Image processing program]
FIG. 3 is a diagram illustrating a functional configuration of the encoding program 5 which is executed by the control device 21 (FIG. 2) and implements the image processing method according to the present invention.
As illustrated in FIG. 3, the encoding program 5 includes a data acquisition unit 500, a color reduction processing unit 510, a reference replacement unit 520, a reference information correction unit 530, and an encoding unit 540.
The combination of the reference substitution unit 520 and the reference information correction unit 530 is an example of the reference information generation unit and the substitution unit according to the present invention.

In the encoding program 5, the data acquisition unit 500 acquires image data to be processed, and outputs the acquired image data to the color reduction processing unit 510 or the reference information correction unit 530.
When the data acquisition unit 500 of this example acquires the image data illustrated in FIG. 1C (that is, image data including reference information, a unique representative color, and a mask), the acquired image data is used as reference information. When output to the correction unit 530 and image data expressed in another format is acquired, the image data is output to the color reduction processing unit 510.

The color reduction processing unit 510 (color reduction means) performs local color reduction processing on the image data input from the data acquisition unit 500.
The color reduction processing unit 510 of this example divides the input image data into image blocks of a predetermined size, determines a predetermined number of representative colors for each of the divided image blocks, and each image block with the determined representative color Fill the pixels.

The reference replacement unit 520 (surrounding reference unit) generates reference information that indicates a reference to a representative color of another image block, based on the image data subjected to the color reduction processing by the color reduction processing unit 510, and the generated reference With information, a part of the representative color (that is, the reference representative color) is replaced.
In this example, the reference replacement unit 520 compares each representative color of the target block with the representative colors (specific representative colors) of the surrounding blocks around the target block, and the representative color whose difference is equal to or less than the reference value is determined. If it exists, a part of the representative color of the target block is replaced with reference information to the representative color of the surrounding block whose difference from the representative color is equal to or less than the reference value. When a part of the representative color is replaced with reference information by the reference replacement unit 520, image data illustrated in FIG. 1C is generated.
In addition, when there are a plurality of replaceable reference information, the reference replacement unit 520 of this example determines a reference representative color so as to leave a unique representative color that does not match the specific representative colors of surrounding blocks as much as possible, Replace with information.

The reference information correction unit 530 (correction unit) encodes image data in which a part of the representative color is replaced with the reference information (that is, image data input from the data acquisition unit 500 or the reference replacement unit 520). The reference information is corrected so that the coding efficiency of the unit 540 is improved. The correction of the reference information means at least one of a change of a reference destination and a change of a representative color (that is, a reference representative color) serving as a reference source.
In this example, since the predictive coding method is applied, the reference information correction unit 530 causes the reference representative color of the target block so that the unique representative color of the adjacent image block matches the unique representative color of the target block. And replace the unique representative color. In addition, the reference information correction unit 530 of the present example does not reduce the number of reference information included in the image data, and the reference representative color when the unique representative color of the target block is referenced from another image block. And replacement with unique representative colors are prohibited.
In this example, a case where the predictive coding scheme is applied will be described as a specific example. However, the present invention is not limited to this. For example, when the JPEG scheme is applied, the reference information correction unit 530 The reference information is corrected so that the unique representative color changes gently.

The encoding unit 540 encodes the image data in which the reference information is corrected by the reference information correction unit 530.
For example, the encoding unit 540 separately encodes the reference information, the unique representative color, and the mask included in the image data using a predetermined encoding method.
The encoding unit 540 of this example encodes the unique representative color using a predictive encoding method.

4A illustrates the configuration of the color reduction processing unit 510 (FIG. 3), and FIG. 4B illustrates the configuration of the reference information correction unit 530 (FIG. 3).
As illustrated in FIG. 4A, the color reduction processing unit 510 includes a block extraction unit 512, a representative value determination unit 514, and a color reduction unit 516.
In the color reduction processing unit 510, the block extraction unit 512 (block setting unit) extracts an image block having a predetermined size from the input image data, and outputs the extracted image block to the representative value determination unit 514.
The color reduction processing unit 510 of this example divides the input image data into 4 × 4 size image blocks illustrated in FIG.

The representative value determining unit 514 determines a predetermined number of representative colors for each of the image blocks extracted by the block extracting unit 512.
As illustrated in FIG. 1A, the representative value determination unit 514 of this example classifies colors included in an image block into two groups according to pixel values, and average pixel values of colors belonging to each group. Is the representative color.

  As illustrated in FIG. 1A, the color reduction unit 516 replaces the color of each pixel included in the image block with the representative color of the group to which this color belongs.

As illustrated in FIG. 4B, the reference information correction unit 530 includes a candidate detection unit 532, a compressibility determination unit 534, a replacement prohibition unit 536, and a replacement unit 538.
In the reference information correction unit 530, the candidate detection unit 532 detects the unique representative color that can be corrected by comparing the unique representative color of the block of interest with the unique representative color of the surrounding blocks, and detects the detected unique representative color. Is output to the compressibility determination unit 534 and the replacement prohibition unit 536.
The candidate detection unit 532 of this example calculates the difference between the unique representative color of the block of interest and the unique representative colors of surrounding blocks around the block of interest, and if the difference is equal to or less than the reference value, the block of interest Are detected as replacement candidates.

The compressibility determination unit 534 determines whether or not the compression rate of the encoding process by the encoding unit 540 is improved when the unique representative color detected by the candidate detection unit 532 is replaced with the reference representative color. When the compression rate is improved, the replacement of the unique representative color that is the replacement candidate and the reference representative color is permitted. When the compression rate is not improved, the replacement of the replacement candidate is prohibited.
In this example, since the predictive coding method is applied, the compressibility determination unit 534 determines whether or not the predictive predictive ratio of the predictive coding process is improved when the unique representative color and the reference representative color are switched. The replacement is permitted only when the predictive predictive value improves.
When attention is paid only to a specific reference position (the immediately preceding image block) as in the run-length encoding method, the compressibility determination unit 534 determines that the reference representative color corresponding to the unique representative color of the replacement candidate is the immediately preceding reference color. It is determined whether or not it matches the unique representative color of the image block (in this example, the image block on the left on the left), and if it matches, this replacement candidate is allowed to be replaced. The replacement of is prohibited.

The replacement prohibition unit 536 determines whether or not the unique representative color detected by the candidate detection unit 532 is referenced from the reference representative color of another image block. If the unique representative color is referenced, the replacement prohibition unit 536 replaces the replacement candidate. If it is prohibited and not referenced, this replacement candidate is allowed to be replaced.
In other words, the replacement prohibition unit 536 sets the unique representative color and the reference representative color that are replacement candidates on the condition that the unique representative color detected by the candidate detection unit 532 is not referenced from the reference representative color of another image block. Allow replacement of.

When the replacement determination unit 534 and the replacement prohibition unit 536 permit replacement of replacement candidates, the replacement unit 538 replaces the unique representative color that is a replacement candidate and the reference representative color, and the compression determination unit 534 and replacement replacement unit 536. If replacement is prohibited by any of the above, replacement is not performed.
That is, the replacement unit 538 can refer to the unique representative color of the target block as the unique representative color of the target block. Only when the replacement is permitted, the reference representative color of the target block is set as the unique representative color, and the unique representative color is replaced with the reference information.
Further, when the unique representative color and the reference representative color are exchanged, the replacement unit 538 inverts the mask for selecting the unique representative color and the reference representative color.

FIG. 5 is a diagram for explaining the correction processing by the reference information correction unit 530 (FIG. 4B). FIG. 5A illustrates the reference relationship before correction, and FIG. The following reference relationship is illustrated.
FIG. 6 is a diagram illustrating a case where correction is prohibited by the replacement prohibition unit 536 (FIG. 4B).
Each of “D1” to “D8” in FIGS. 5 and 6 indicates the value of the representative color of each image block.
As illustrated in FIG. 5A, the reference replacement unit 520 (FIG. 3) displays the representative color “D1” of the target block and the representative color “D1” of the surrounding block 1 immediately before (on the left) of this target block. Therefore, the representative color “D1” of the block of interest is replaced with the reference information to the surrounding block 1. As a result, the representative color “D2” of the block of interest becomes the unique representative color.
In the reference relationship (before correction) illustrated in FIG. 5A, the unique representative color “D1” of the surrounding block 1 and the unique representative color “D2” of the target block do not match. When applied, the continuous length is interrupted and the compression efficiency is lowered.

Here, when the surrounding block 2 above the block of interest is viewed, the unique representative color “D2” of the block of interest and the unique representative color “D2” of the surrounding block 2 match. The color “D2” is also a representative color (that is, a replacement candidate) that can be referred to. Accordingly, the candidate detection unit 532 detects the unique representative color “D2” of this block of interest as a replacement candidate.
In addition, since the reference representative color “D1” of the target block matches the unique representative color “D1” of the surrounding block 1, the same unique representative color “D1” continues by becoming a unique representative color. It will be. That is, by making the reference representative color “D1” of the block of interest the unique representative color, the code amount when the run-length encoding method or the like is applied becomes small. Thus, as in this example, the compressibility determination unit 534 permits replacement when the reference representative color “D1” of the target block that is a replacement candidate is replaced with a unique representative color, and the compression rate is improved.
Further, the replacement prohibition unit 536 determines whether the unique representative color “D1” of the block of interest is not referenced from the surrounding blocks, and permits replacement when it is not referenced.
Then, as illustrated in FIG. 5A, the replacement unit 538 matches the unique representative color “D2” of the target block with the unique representative color of the surrounding block, and sets the reference representative color “D1” of the target block. When the unique representative color is used, the compression efficiency is improved and the unique representative color “D2” of the target block is not referred to from the surrounding blocks, as illustrated in FIG. Swap the representative color and the reference representative color.

On the other hand, as illustrated in FIG. 6A, when the unique representative color “D2” of the target block matches the reference representative color “D2” of the surrounding block 3 and is referenced from the surrounding block 3. 6B, if the unique representative color “D2” of the target block is the reference representative color and the reference representative color “D1” of the target block is the unique representative color, the reference from the surrounding block 3 Becomes impossible.
Therefore, when the unique representative color “D2” of the target block is referenced from the surrounding block 3, the replacement prohibition unit 536 (FIG. 4B) replaces the specific representative color and the reference representative color in the target block. Is prohibited.

FIG. 7 is a diagram for explaining the configuration of the encoding unit 540 (FIG. 3) in more detail.
As illustrated in FIG. 7, the encoding unit 540 includes a prediction unit 542, a prediction error calculation unit 544, a run counting unit 546, a selection unit 548, and a code generation unit 550. In the following description, the case where the unique representative color is encoded will be described. However, the reference information and the mask may be encoded in the same manner.

When the unique representative color of the block of interest is encoded, the prediction unit 542 generates prediction data of the block of interest based on the unique representative color of another image block, and the generated prediction data and the unique block of the block of interest The representative color is compared, and the comparison result is output to the run counter 546.
The prediction unit 542 of this example uses the unique representative color of the surrounding block immediately before the target block (the peripheral block 1 illustrated in FIG. 5) as prediction data, and compares this prediction data with the specific representative color of the target block. When the unique representative color and the prediction data match (that is, when the prediction is correct), the prediction unit ID for identifying itself is output to the run counting unit 546, and in other cases, the prediction data matches. The fact that there was no output is output to the run counter 546.
Note that the prediction unit 542 may be two or more types, and for example, a prediction unit that refers to the unique representative color of the surrounding block immediately above the target block (the surrounding block 2 in FIG. 5) may be added.

The prediction error calculation unit 544 generates prediction data of the unique representative color of the block of interest by a predetermined prediction method, calculates a difference between the generated prediction data and the unique representative color of the block of interest. The difference is output to the selection unit 548 as a prediction error.
The prediction error calculation unit 544 of this example calculates a prediction error value using the unique representative color of the surrounding block immediately before the target block as prediction data.

The run counting unit 546 counts the number of consecutive prediction unit IDs, and outputs the prediction unit ID and the number of continuous predictions to the selection unit 548.
The run counting unit 548 of the present example outputs the number of continuations counted by the internal counter when the prediction unit 542 does not match the unique representative color of the block of interest.

  When the continuous number is input from the run counting unit 546, the selection unit 548 outputs the input continuous number to the code generation unit 550, and when the prediction error value is input from the prediction error calculation unit 544, the selection unit 548 calculates the continuous number. After the output, the input prediction error value is output to the code generation unit 550.

The code generation unit 550 encodes the continuous number and the prediction error value input from the selection unit 548, and outputs them to the communication device 22 (FIG. 2) or the recording device 24 (FIG. 2).
The code generation unit 550 ignores this continuous number when the continuous number 0 is input.

FIG. 8 is a flowchart of the encoding process (S10) by the image processing program 5 (FIG. 3). In this example, a case where image data that has not undergone color reduction processing is input will be described as a specific example.
As shown in FIG. 8, in step 100 (S100), the data acquisition unit 500 (FIG. 3) outputs the input image data to the color reduction processing unit 510.

In step 120 (S120), the color reduction processing unit 510 performs local color reduction processing on the image data input from the data acquisition unit 500, as illustrated in FIG. The processed image data is output to the reference replacement unit 520.
More specifically, the block extraction unit 512 (FIG. 4A) of the color reduction processing unit 510 divides the input image data into 4 × 4 size image blocks, and the representative value determination unit 514 Two representative colors are determined for each of the image blocks, and the color reduction unit 516 replaces the color of each pixel included in the image block with the determined representative colors.

  In step 140 (S140), the reference information replacement unit 520 (FIG. 3) generates reference information for instructing to refer to the representative colors of the surrounding blocks based on the image data input from the color reduction processing unit 510. As illustrated in 1 (C), some representative colors (that is, reference representative colors) of the block of interest are replaced with the generated reference information.

  In step 160 (S160), the reference information correction unit 530 corrects the reference information generated by the reference information replacement unit 520 so that the compression efficiency of the encoding unit 540 is improved, and corrects at least part of the reference information. The processed image data is output to the encoding unit 540.

  In step 180 (S180), the encoding unit 540 separately encodes the image data (inherent representative color of each block, reference information, and mask) input from the reference information correction unit 530, and outputs the code data to the outside. To do.

  In this example, after replacing some representative colors with reference information by the reference information replacing unit 520, the reference information is corrected (that is, the unique representative color and the reference representative color are corrected so that the compression rate is improved). However, after generating the reference information and correcting the generated reference information so that the compression rate is improved, a part of the representative color is replaced with the corrected reference information. May be.

FIG. 9 is a flowchart of the reference information correction process (S160) shown in FIG.
As shown in FIG. 9, in step 162 (S162), the reference information correction unit 530 (FIGS. 3 and 4) sequentially selects the image data of the target block from the input image data (FIG. 1C). Set.

In step 164 (S164), the candidate detecting unit 532 (FIG. 4B) can refer to the unique representative color of the block of interest by comparing the unique representative colors of the surrounding blocks around the block of interest. It is determined whether or not there is. More specifically, the candidate detection unit 532 calculates a difference between the unique representative color of the block of interest and the unique representative colors of surrounding blocks around the block of interest, and the calculated difference is equal to or less than the reference value. In this case, it is determined that reference is possible, and in other cases, it is determined that reference is impossible.
When the candidate detection unit 532 determines that the unique representative color of the target block can refer to the unique representative colors of the surrounding blocks, the reference information correction unit 530 proceeds to the process of S166 and cannot refer to it. If it is determined, the process proceeds to S172.

In step 166 (S166), the compressibility determination unit 534 (FIG. 4B) determines whether or not the compression rate is improved when the reference representative color and the unique representative color of the block of interest are exchanged. In this example, when the unique representative color of the immediately preceding surrounding block matches the unique representative color of the block of interest, the prediction is correct and the compression rate is improved. Therefore, the compressibility determination unit 534 of this example It is determined whether or not the reference representative color of the block matches the unique representative color of the surrounding block immediately before this block of interest. If they match, it is determined that the compression ratio is improved. It is determined that the compression rate is not improved.
The reference information correction unit 530 proceeds to the processing of S168 when the compressibility determination unit 534 determines that the compression rate is improved by replacement, and proceeds to the processing of S172 when it is determined that the compression rate is not improved. To do.

In step 168 (S168), the replacement prohibition unit 536 (FIG. 4B) determines whether or not the unique representative color of the block of interest is referenced by the representative colors of the surrounding blocks.
The reference information correction unit 530 proceeds to the process of S170 when the replacement prohibition unit 536 determines that no reference is made from any of the surrounding blocks, and when it is referenced from any of the surrounding blocks, the reference information correction unit 530 proceeds to S172. Move on to processing.

In step 170 (S170), the replacement unit 538 (FIG. 4B) replaces the unique representative color of the block of interest with reference information to surrounding blocks, and sets the reference representative color based on the reference information of the block of interest. The identified reference representative color is specified as the unique representative color.
Note that the replacement unit 538 inverts the mask when the unique representative color and the reference representative color of the block of interest are replaced.

  In step 172 (S172), the reference information correction unit 530 determines whether or not the processing has been completed for all the image blocks, and ends the reference information correction processing when the processing has been completed for all the image blocks. If there is an unprocessed image block, the next block of interest is set, and the process returns to S164.

  In this figure, replacement candidates are detected, and it is confirmed whether or not the compression rate is improved and whether the detected replacement candidates are not referenced from other image blocks. Prepare a line buffer for 4 lines. First, exclude the image block referenced by the surrounding blocks from the image blocks belonging to the line of interest, and then select the unique representative color from the line of interest. It is also possible to detect an image block that matches the unique representative color of the surrounding block and improves the compression ratio by exchanging the unique representative color and the reference representative color, and replace the detected image block. Good.

  As described above, the image processing apparatus 2 according to the present embodiment performs local color reduction processing and corrects the result of color reduction processing so as to conform to the encoding method. Thereby, the image processing apparatus 2 can generate code data with a high compression rate with a relatively light processing load. In addition, image quality deterioration due to the color reduction process is hardly realized.

It is a figure explaining the specific example of a local color reduction process. It is a figure which illustrates the hardware constitutions of the image processing apparatus 2 with which the image processing method concerning this invention is applied centering on the control apparatus 21. FIG. It is a figure which illustrates the functional structure of the encoding program 5 which is performed by the control apparatus 21 (FIG. 2) and implement | achieves the image processing method concerning this invention. (A) illustrates the configuration of the color reduction processing unit 510 (FIG. 3), and (B) illustrates the configuration of the reference information correction unit 530 (FIG. 3). FIGS. 5A and 5B are diagrams illustrating correction processing by the reference information correction unit 530 (FIG. 4B). FIG. 5A illustrates a reference relationship before correction, and FIG. 5B illustrates a reference relationship after correction. Is illustrated. It is a figure which illustrates the case where correction | amendment is prohibited by the replacement prohibition part 536 (FIG.4 (B)). It is a figure explaining the structure of the encoding part 540 (FIG. 3) in detail. It is a flowchart of the encoding process (S10) by the image processing program 5 (FIG. 3). It is a flowchart of the reference information correction process (S160) shown in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 2 ... Image processing apparatus 5 ... Encoding program 500 ... Data acquisition part 510 ... Color reduction process part 512 ... Block extraction part 514 ... Representative value determination part 516 ... Color reduction part 520 Reference replacement unit 530 Reference information correction unit 532 Candidate detection unit 534 Compressibility determination unit 536 Replacement prohibiting unit 538 Replacement unit 540 Encoding unit

Claims (8)

Image data acquisition means for acquiring image data including a representative color of an image block of a predetermined size and reference information instructing reference to a representative color of another image block;
Correction means for correcting at least one of the representative color and the reference information included in at least a part of the image data acquired by the image data acquisition means;
An image processing apparatus comprising: encoding means for encoding image data at least partially corrected by the correction means. The image processing apparatus according to claim 2, wherein the correction unit corrects some image data so that representative colors of a plurality of continuous image blocks have the same value. Block setting means for setting an image block of a predetermined size in the input image data;
Representative color determining means for determining a predetermined number of representative colors or less for each image block based on the color included in the image block set by the block setting means;
Color reduction means for generating data of reduced image blocks using the representative color determined for each image block by the representative color determination means;
Reference information generating means for comparing the representative colors determined for each of the plurality of image blocks with each other and generating reference information for specifying a reference destination of the representative color;
Substitution means for replacing some representative colors among the representative colors generated for each image block by the color reduction means with reference information generated by the reference information generation means,
Encoding means for encoding other representative colors of the image block in which some representative colors are replaced with reference information by the replacement means, and
The image processing apparatus, wherein the reference information generation unit generates the reference information so that a code amount generated by the encoding unit is small. The reference information generating means includes
A surrounding reference means for generating reference information by comparing a representative color of an image block to be processed with a representative color of an image block around the image block;
Reference information correction means for correcting the reference information generated by the surrounding reference means in accordance with the encoding method performed by the encoding means, and
The image processing apparatus according to claim 3, wherein the replacement unit performs a replacement process using the reference information corrected by the reference information correction unit. The encoding means compares the representative color of the image block to be processed with the representative color of the image block at a predetermined position with respect to the image block, and generates code data by predictive encoding processing,
The image processing apparatus according to claim 4, wherein the reference information correction unit corrects the reference information so that a prediction accuracy in the predictive encoding process is improved. The image processing apparatus according to claim 4, wherein the reference information correction unit corrects the reference information on the condition that a representative color of the image block to be processed is not referenced from another image block. Obtain image data including the representative color of the image block of the default size and the reference information instructing reference to the representative color of another image block,
Correcting at least one of the representative color and the reference information included in at least some of the acquired image data so that the amount of code generated by a predetermined encoding method is small;
An image processing method for encoding image data, at least a part of which is corrected, by the encoding method. Obtaining image data including a representative color of an image block of a predetermined size and reference information instructing reference to a representative color of another image block;
Correcting at least some of the acquired image data so that the amount of code generated by a predetermined encoding method is small; and
A program for causing a computer to execute a step of encoding image data, at least a part of which has been corrected, by the encoding method.

Images