JP2006005740A - Image data compression/expansion method, and image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image data compression/expansion method and an image processing apparatus which can attain a higher compression rate, and moreover, do not produce a blur at the edge part of a character. <P>SOLUTION: A reference image data is irreversibly compressed, while a mask data indicative of a transparent part of the reference image data is reversibly compressed. Next, each mask data indicative of a transparent part of one or a plurality of original image data following the reference image data is reversibly compressed, while difference data between the one or the plurality of original image data and the reference image data is irreversibly compressed respectively. Next, an original image data next to the one or the plurality of original image data is irreversibly compressed as a reference image data, while a mask data indicative of a transparent part of the reference image data is reversibly compressed. Then, each mask data of one or a plurality of original image data following the reference image data is reversibly compressed, while difference data between the one or the plurality of original image data and the references image data is irreversibly compressed respectively. The above-mentioned processing is repeated. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a method for compressing / decompressing image data for sprite display and an image processing apparatus using the same.

The following methods are known as conventional color image data compression methods.
(1) Cell reference method, line reference method These methods are effective for solid pictures with the same color code before and after, but there is a drawback that the compression rate does not increase if there is no same color code before and after. . In addition, there is a drawback that irreversible / high efficiency compression of RGB format image data is not possible.

(2) Dictionary method Large dictionary data is required to increase the reproducibility. Since the lossless compression is performed in the same manner as the cell reference method and the line reference method, there is a disadvantage that irreversible / high-efficiency compression of image data in RGB format cannot be performed.

(3) DCT transformation + 2-dimensional Huffman coding (JPEG; Joint Photographic image coding Experts Group)
Since a block of 8 × 8 size is formed and DCT (Discrete Cosine Transform) is performed, distortion in block units becomes conspicuous as the compression rate increases. In addition, since the approximation is essentially a sine wave, a portion having a strong edge such as an animation image cannot be approximated well. Further, a complicated Huffman table is required for two-dimensional Huffman encoding, and processing such as table reference is disadvantageous for hardware implementation. Furthermore, there is a drawback that the index format image data cannot be reversibly compressed on the premise of color palette conversion.

(4) Wavelet transform + arithmetic coding (JPEG2000)
In order to perform the two-dimensional wavelet conversion for image compression, a memory equivalent to the pixel size is required as a buffer, and therefore, it is not suitable for hardware and requires a long conversion time. Arithmetic coding is complicated in operation. In particular, when encoding is performed in units of bit planes as in JPEG2000, decoding processing takes time. Wavelet transform uses image correlation in the same way as DCT, and is not suitable for lossless compression of index-format image data.
In view of such a conventional compression method, the applicant of this application first,
Japanese Patent Application No. 2002-234984
Japanese Patent Application No. 2003-431199
Has been filed.

  On the other hand, in moving image display using sprites, sprite pattern data is usually composed of character image color data and transparent data. FIG. 5 is a diagram showing a moving image display in which a car sequentially moves in a background image on which mountains and trees are displayed. Sprite pattern data D1 to D4 obtained by sequentially moving a car are used as sprite pattern data BG of the background image. A moving image is displayed by displaying the images in a superimposed manner. In this case, the sprite pattern data D1 to D4 are composed of the color data of the automobile J and the transparent data of the transparent portion T.

By the way, when the sprite pattern data composed of the color data and the transparent data of the character image is compressed, the transparent data is used at the contact portion between the transparent portion and the character by using the above-described lossy compression in order to increase the compression rate. Changes to different data, which causes a problem that the edge of the character is blurred. On the other hand, when reversible compression is used, such a problem does not occur, but there is a disadvantage that the compression rate is low.
The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an image data compression / decompression method and an image processing apparatus that can obtain a high compression rate and that do not cause blurring of edge portions. It is in.

  SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems. The invention according to claim 1 is a compression method for compressing original image data to be displayed sequentially one after another. The image data is irreversibly compressed, and mask data indicating a transparent portion of the reference image data is reversibly compressed, and difference data between one or a plurality of original image data following the reference image data and the reference image data is respectively determined to be non-reversible. In addition to lossless compression, each mask data indicating a transparent portion of the one or more original image data is losslessly compressed, and the original image data next to the one or more original image data is irreversibly compressed as reference image data. , Reversibly compressing mask data indicating a transparent portion of the reference image data, and one or more original image data following the reference image data and the reference image data With each lossy compressed difference data, each mask data indicating the one or more transparent portions of the original image data and lossless compression, an image data compression method, characterized in that the above-described processing is repeated.

  According to a second aspect of the present invention, there is provided a compression method for compressing original image data to be displayed sequentially and continuously, irreversibly compressing one original image data as reference image data, and a transparent portion of the reference image data. And reversibly compress the mask data indicating the difference data between one or a plurality of original image data following the reference image data and the original image data before one display timing of each original image data, Each mask data indicating a transparent portion of the one or more original image data is reversibly compressed, the original image data next to the one or more original image data is irreversibly compressed as reference image data, and the reference image data The mask data indicating the transparent portion of the image is reversibly compressed, and one or a plurality of original image data following the reference image data and one display of each original image data Image data characterized by irreversibly compressing difference data from original image data before imming, reversibly compressing each mask data indicating a transparent portion of the one or more original image data, and repeating the above processing Compression method.

According to a third aspect of the present invention, in the image data compression method according to the first aspect, the reference image data is set every time a certain number of the difference data are compressed.
According to a fourth aspect of the present invention, in the image data compression method according to the first aspect, the reference image data is set when a data amount of the difference data becomes a certain amount or more.

  According to a fifth aspect of the present invention, there is provided a decompression method for decompressing image data compressed by the image data compression method according to the first aspect, wherein the compressed reference image data and mask data are decompressed and the mask is decompressed. Transparent data is allocated to the transparent portion of the reference image data expanded based on the data, the compressed differential data and the mask data are expanded, and the transparent portion of the differential data expanded based on the expanded mask data Transparent data is assigned to the transparent data, transparent data is output for the transparent portion of the difference data, and the difference data and the reference image data are added and output for the portions other than the transparent portion of the differential data. This is a data decompression method.

  According to a sixth aspect of the present invention, there is provided a decompression method for decompressing image data compressed by the image data compression method according to the second aspect, wherein the compressed reference image data and mask data are decompressed and the mask is decompressed. Transparent data is allocated to the transparent portion of the reference image data expanded based on the data, the compressed differential data and the mask data are expanded, and the transparent portion of the differential data expanded based on the expanded mask data The transparent data is assigned to the first difference data to which the transparent data is assigned and the reference image data to which the transparent data is assigned to obtain the first decompressed image data, and the transparent data is written in the first data 2 to obtain the second decompressed image data by adding the difference data of 2 and the first decompressed image data in which the transparent data is written. An image data decompression method and repeating the process.

  The invention according to claim 7 includes compressed image data obtained by irreversibly compressing image data, compressed difference data obtained by irreversibly compressing difference data, and mask data indicating each transparent portion of the image data and the difference data. An image processing apparatus for displaying an image on a display device based on display data comprising reversibly compressed compressed mask data, decompression means for decompressing the compressed image data, the compressed differential data, and the compressed mask data, respectively Transparent processing means for assigning transparent data to the decompressed image data and decompressed differential data decompressed by the decompressing means based on the decompressed mask data, the image data after the transparent processing, and the differential data after the transparent processing And adding means for adding one of the image data after the transparent processing or the output data of the adding means Selecting means, comprising a frame buffer for output of said selecting means is written, an image processing device and performs image display based on the data in the frame buffer.

  According to the present invention, image data and difference data are compressed by irreversible compression, and mask data corresponding to a transparent portion is separately prepared, and the mask data is compressed by lossless compression. Thereby, a high compression ratio by irreversible compression can be obtained, and at the same time, blurring of the edge portion can be prevented by the mask data.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is an explanatory diagram showing an image data compression / decompression method according to one embodiment of the present invention. In this figure, F1 and F2 are original image data by YUV (luminance / color difference / color difference) color data before compression, and are displayed in the order of F1 and F2 at the time of display. The image data F2 is image data when the elliptical character of the image data F1 is slightly moved in the right direction. The image data F1 is composed of character data C1 indicating the color of the character and transparent data E1 set around the character. Similarly, the image data F2 is composed of character data C2 indicating the color of the character and transparent data E2 set around the character.

In the compression / decompression method according to this embodiment, first, mask data M1 (not shown) and M2 for designating each dot in the transparent area are generated based on the image data F1 and F2. In this case, in the mask data M1 and M2, data corresponding to each dot in the transparent area is “1”, and data corresponding to each dot in the character area is “0”. Next, the difference between the corresponding dot data of the image data F1 and F2
P2 = F2-F1
Is calculated to obtain the difference data P2. Then, the image data F1 and the difference data P2 are compressed at a high compression rate by the irreversible compression method, and the mask data M1 and M2 are compressed by the lossless compression and stored in the storage medium.

  As described above, in the compression processing according to the embodiment of the present invention, “original image data and mask data” and “difference data and mask data” are sequentially compressed and recorded on the storage medium every other image data.

  Next, the expansion method will be described with reference to FIG. First, the compressed image data F1 and the compressed mask data M1 are decompressed, and then transparent data is written in the area of the decompressed image data F1 corresponding to the data “1” of the mask data M1. Here, the transparent data is “00002 (hexadecimal)”. As a result, reproduced image data F1a (FIG. 1) in which the image data F1 is substantially reproduced is obtained, and the character peripheral portion of the reproduced image data F1a is not blurred. Next, the compressed differential data P2 and the compressed mask data M2 are expanded, and then transparent data is written in the expanded differential data P2 area corresponding to the data “1” of the mask data M2. Thereby, the difference data P2a (FIG. 1) consisting of the difference data P2 in the area corresponding to the character of the image data F2 and the surrounding transparent data is obtained.

Next, the reproduction image data F2a obtained by reproducing the image data F2 is obtained by adding the reproduction image data F1a and the difference data P2a for each corresponding dot data. However, in this case, when the difference data P2a is transparent data “00002”, addition is not performed and transparent data “00002” is written. The obtained reproduced image data F1a and F2a are converted into RGB (red / green / blue) color data and displayed.
As described above, the compression / decompression method according to the above embodiment increases the compression rate by taking the difference and compresses, and prevents blurring of the peripheral portion of the character by using the mask data.

  In the above embodiment, every other original image data is set as difference data. However, as shown in FIG. 2, a plurality of difference data (four in FIG. 2) may be used for one reference image data. Good. That is, in FIG. 2, F1 to F6 are original image data that are sequentially displayed. First, the original image data F1 serving as the reference image data is compressed as it is into the compressed image data K1, and then the original image data F2 The difference data from the original image data F1 is compressed into compressed difference data d2, and then the difference data between the original image data F3 and the original image data F1 is compressed into compressed difference data d3. Similarly, the compressed difference data d4 and d5 are obtained. Next, the original image data F6 is compressed as reference image data as it is to obtain compressed image data K2, and then difference data is sequentially obtained. Then, the obtained data K1, d2, d3, d4, d5, K2,... Are stored in a storage medium.

  In FIG. 2, the description of the mask data is omitted. In the figure, the squares increasing in the order of d2 → d5 generally indicate that the difference increases as the distance from the original image data F1 increases.

  Next, when the compressed data is reproduced, first, the compressed image data K1 is expanded to obtain reproduced image data I1 (= F1a). Next, the compressed difference data d2 is expanded, and the decompressed data and the reproduced image data I1 are added to obtain reproduced image data F2a. Thereafter, reproduced image data F3a, F4a, and F5a are obtained in the same manner. Next, the compressed image data K2 is expanded to obtain reproduced image data I2 (= F6a). Thereafter, the same processing is repeated.

  Also, as shown in FIG. 3, difference data from the previous original image data may always be obtained, compressed, and recorded on a storage medium. That is, in FIG. 3, F1 to F6 are sequentially displayed original image data. First, the original image data F1 serving as the reference image data is compressed as it is into the compressed image data K1, and then the original image data F2 and The difference data from the original image data F1 is compressed to be compressed difference data d2, and then the difference data between the original image data F3 and the original image data F2 is compressed to be compressed difference data d3. d4 and d5 are obtained. Next, the original image data F6 is compressed as reference image data as it is to obtain compressed image data K2, and then difference data is obtained sequentially in the same manner. Then, the obtained data K1, d2, d3, d4, d5, K2,... Are stored in a storage medium. In FIG. 3, the mask data is not shown.

Next, when the compressed data is reproduced, first, the compressed image data K1 is expanded to obtain reproduced image data I1 (= F1a). Next, the compressed difference data d2 is expanded, and the decompressed data and the reproduced image data I1 are added to obtain reproduced image data F2a. Next, the compressed difference data d3 is expanded, and the decompressed data and the reproduced image data F2a are added to obtain reproduced image data F3a. Thereafter, reproduced image data F4a and F5a are obtained in the same manner. Next, the compressed image data K2 is expanded to obtain reproduced image data I2 (= F6a). Thereafter, the same processing is repeated.
In the above example, the number of difference data is a fixed number, but instead, the reference image data may be prepared when the data amount of the difference data exceeds a certain level.

  Next, an image processing apparatus using the above-described image data expansion method will be described with reference to FIG. FIG. 4 is a block diagram showing the configuration of the image processing apparatus. The image processing apparatus shown in FIG. 4 stores the “compressed image data and compressed mask data” and “compressed differential data and compressed mask data” described in FIG. This is an apparatus for reproducing image data of a storage medium stored alternately.

  In FIG. 4, reference numeral 1 denotes a FIFO (first-in / first-out) memory in which compressed data (YUV data) read from a storage medium is temporarily stored. Reference numeral 2 denotes a decompression circuit that decompresses compressed image (or difference) data and compression mask data sequentially output from the FIFO memory 1, and the decompressed Y data, U data, and V data are respectively written in the buffers 3a to 3c. The mask data is written into the buffer 3d. Reference numeral 4 denotes a YUV → RGB conversion circuit for converting the YUV data in the buffers 3a to 3c into RGB data. Reference numeral 5 denotes a selector. When the mask data output from the buffer 3d is “0”, the RGB data output from the YUV → RGB conversion circuit 4 is selected and output. When the mask data is “1”, the transparent data “00002” is selected. Select "" to output. By the processing of the selector 5, the transparent data of the reproduced image is correctly set, and thereby blurring of the boundary portion of the character can be prevented.

6 is an adder circuit, 7 is a selector, and 8 is a frame buffer. The adder circuit 6 adds the image data read from the selector 5 to the image data before one display cycle read from the frame buffer 8 and outputs the result to the selector 7. That is, the adding circuit 6 is the same as the F2a = F1a + P2a described in FIG.
And outputs the result to the selector 7. However, when transparent data is output from the selector 5, the transparent data is output unconditionally. The selector 7 alternately selects and outputs the data output from the selector 5 and the data output from the adder circuit 6 every display cycle. That is, at the timing when the image data is output from the selector 5, the output data of the selector 5 is selected and output to the frame buffer 8, and at the timing when the difference data is output from the selector 5, the output data of the adder circuit 6 is output. Is output to the frame buffer 8.

  The frame buffer 8 has a two-buffer configuration. When the image data output from the selector 7 is written in one buffer, the image data in the other buffer is read out and output to the display device, and the next display cycle. 2 buffers are used in reverse, and this operation is repeated alternately.

  In the above embodiment, YUV data is compressed and stored in a storage medium, and the YUV data is decompressed and then returned to RGB data for display. However, the RGB data is compressed and stored, and decompressed. It may be displayed. However, in the case of YUV data, the compression rate can be further increased by using a thinning process for U and V data.

  The present invention is used for image display of game machines such as home game machines and pachinko machines.

It is explanatory drawing for demonstrating the image data compression / decompression method by one Embodiment of this invention. It is explanatory drawing for demonstrating the other example of the compression method. It is explanatory drawing for demonstrating the other example of the compression method. 1 is a block diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention. It is explanatory drawing for demonstrating the moving image display by a sprite.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... FIFO memory, 2 ... Decompression circuit, 3a-3d ... Buffer, 4 ... YUV-> RGB conversion circuit, 5 ... Selector, 6 ... Adder circuit, 7 ... Selector, 8 ... Frame buffer.

Claims (7)

In a compression method for compressing original image data to be sequentially displayed,
Irreversibly compress one original image data as reference image data, and reversibly compress mask data indicating a transparent portion of the reference image data,
The difference data between one or a plurality of original image data following the reference image data and the reference image data is irreversibly compressed, and each mask data indicating a transparent portion of the one or more original image data is reversibly compressed. ,
The original image data next to the one or more original image data is irreversibly compressed as reference image data, and the mask data indicating the transparent portion of the reference image data is reversibly compressed,
The difference data between the one or more original image data following the reference image data and the reference image data is irreversibly compressed, and each mask data indicating a transparent portion of the one or more original image data is reversibly compressed. ,
An image data compression method characterized by repeating the above processing. In a compression method for compressing original image data to be sequentially displayed,
Irreversibly compress one original image data as reference image data, and reversibly compress mask data indicating a transparent portion of the reference image data,
The difference data between one or a plurality of original image data following the reference image data and the original image data before one display timing of each original image data is irreversibly compressed, and the one or more original image data Reversibly compress each mask data indicating the transparent part,
The original image data next to the one or more original image data is irreversibly compressed as reference image data, and the mask data indicating the transparent portion of the reference image data is reversibly compressed,
The difference data between one or a plurality of original image data following the reference image data and the original image data before one display timing of each original image data is irreversibly compressed, and the one or more original image data Reversibly compress each mask data indicating the transparent part,
An image data compression method characterized by repeating the above processing.   The image data compression method according to claim 1, wherein the reference image data is set every time a certain number of the difference data are compressed.   The image data compression method according to claim 1, wherein the reference image data is set when a data amount of the difference data becomes equal to or greater than a certain value. A decompression method for decompressing image data compressed by the image data compression method according to claim 1.
Decompress the compressed reference image data and mask data,
Assigning transparent data to transparent portions of the reference image data expanded based on the expanded mask data;
Decompress the compressed difference data and mask data,
Assigning transparent data to the transparent portion of the differential data expanded based on the expanded mask data;
An image data decompression method, comprising: outputting transparent data for a transparent portion of the difference data, and adding and outputting the difference data and reference image data for a portion other than the transparent portion of the difference data. A decompression method for decompressing image data compressed by the image data compression method according to claim 2,
Decompress the compressed reference image data and mask data,
Assigning transparent data to transparent portions of the reference image data expanded based on the expanded mask data;
Decompress the compressed difference data and mask data,
Assigning transparent data to the transparent portion of the differential data expanded based on the expanded mask data;
Adding first difference data to which transparent data is assigned and reference image data to which transparent data is assigned to obtain first decompressed image data;
Adding the second difference data in which the transparent data is written and the first decompressed image data in which the transparent data is written to obtain the second decompressed image data;
Hereinafter, an image data expansion method characterized by repeating the above processing. Compressed image data obtained by irreversibly compressing image data, compressed differential data obtained by irreversibly compressing difference data, compressed mask data obtained by reversibly compressing the image data and mask data indicating each transparent portion of the difference data, In an image processing apparatus that displays an image on a display device based on display data consisting of:
Decompression means for decompressing each of the compressed image data, the compressed differential data, and the compressed mask data;
Transparent processing means for assigning transparent data to the decompressed image data and decompressed differential data decompressed by the decompressing means based on the decompressed mask data;
Adding means for adding the image data after the transparent process and the difference data after the transparent process;
A selecting means for selecting and outputting one of the image data after the transparent processing or the output data of the adding means;
A frame buffer into which the output of the selection means is written;
And an image processing apparatus for displaying an image based on data in the frame buffer.

Images