JPH08317391A - Device for compressing and extending dynamic image data - Google Patents

Abstract

PURPOSE: To reduce the sense of discontinuity to be generated in an extended image when compressing processing is performed to a dynamic image for each block. CONSTITUTION: A frame memory circuit 10 successively stores dynamic image data VIN for the unit of an image (frame). In a certain frame cycle, a read control circuit 12 generates and outputs a block equally divided in the range of an original image and in the next frame cycle, the frame memory circuit 10 is controlled so as to generate and output a block dividing a picture so as not to be overlapped with the boundary of the block in the preceding frame cycle. A compression encoding circuit 14 performs DCT processing successively to the blocks inputted from the frame memory circuit 10 and further executes compression encoding by performing quantizing processing and variable length encoding processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving image data compression apparatus for dividing an image included in moving image data into a plurality of blocks and performing compression processing, and decompressing moving image data thus compressed. And a moving image data decompressing device.

[0002]

2. Description of the Related Art FIG. 7 is a diagram showing an example of image data included in moving image data. FIG. 8 shows an orthogonal transform (DCT)
FIG. 3 is a diagram showing a method of compressing image data according to FIG.
Shows the block after division, (B) shows the frequency component obtained by orthogonally transforming the block, and the values of (0,0) to (7,7) attached to each pixel or each frequency component are Indicates a matrix vector. As a method of compressing moving image data, each image included in the moving image data is equally divided into square blocks, and orthogonal transformation such as discrete cosine transform (DCT) is performed on pixel values of pixels included in each block. Conventionally, a method of compressing moving image data has been used.

A method of compressing the above-mentioned moving image data will be described below. When the moving image data includes an image as shown in FIG. 7, each of the images is, for example, as shown in FIG.
It is equally divided into blocks including 8 × 8 (64) pixels shown in FIG. By performing orthogonal transformation such as DCT on the block thus obtained, the frequency component corresponding to each element in FIG. 7B can be obtained. Figure 7
Each element shown in (B) corresponds to a higher frequency component in the horizontal direction of the image as it goes to the right side, and corresponds to a higher frequency component in the vertical direction of the image as it goes to the lower side.

That is, the element (0,0) in FIG. 7B is a direct current component (offset value) of the pixel value of the pixel in the block, and when there are many small changes in the image in the block, FIG. The value of the frequency component in the B direction shown in B) also increases,
If the image in the block is smooth and has little change, then FIG.
The value of the frequency component in the A direction shown in (B) becomes large. The above-described orthogonally transformed moving image data is further quantized, variable length coded, and transmitted or recorded as compressed moving image data.

[0005]

However, when the compressed moving image data generated by orthogonally transforming an image in block units as described above is inversely orthogonally transformed and decompressed, the boundary of the blocks of the decompressed image is reduced. May appear discontinuous. This is because the boundary between adjacent blocks in which the difference in the offset value (value of the element (0,0) in FIG. 7B) and the like for each block due to the quantization processing after orthogonal transformation becomes larger than the original image, It always appears in the same position on the screen and is emphasized.

The present invention has been made in view of the above-mentioned problems of the prior art, in which each image included in moving image data is equally divided into predetermined blocks, and DC is divided into blocks.
An object of the present invention is to provide a moving image data compression apparatus capable of compressing moving image data so that the boundaries of blocks do not appear discontinuous in an image after decompression even when T or the like is performed. It is another object of the present invention to provide a moving picture data decompression device capable of decompressing moving picture data compressed by the moving picture data compression device according to the present invention and reproducing an original moving picture.

[0007]

In order to achieve the above object, a moving image data compression apparatus according to the present invention divides each image of moving image data into a plurality of blocks each including a predetermined number of pixels. A moving image data compression apparatus that performs compression processing by dividing the images into a plurality of blocks by shifting the boundaries in a horizontal direction and a vertical direction at predetermined intervals and generating the blocks. And a moving image data compression unit that compresses the moving image data by performing a predetermined compression process on the pixel values of the plurality of pixels included in each of the blocks.

Preferably, the block generation means equally divides one of the two continuous images of the moving image data into the plurality of blocks by the first division method in which each of the plurality of blocks becomes a rectangle. Then, in the other of the two continuous images of the moving image data, each of the plurality of blocks is the rectangle, and a boundary between the blocks is a boundary between the blocks in the first division method. The image is divided into the plurality of blocks by a different second division method, and out of the plurality of blocks of the image divided by the second division method, a part of the block protrudes from the image, and the block protrudes from the image of each of the blocks. To compensate for the pixels in the area.

[0009] Preferably, the block generation means is an area of the plurality of blocks of the image divided by the second division method, which is partly protruding from the image and which is partly protruding from the image. Pixels of the same image, which are not included in any of the plurality of blocks as a result of being divided by the second dividing method, or by adding pixels of the same image. This is compensated by adding pixels in adjacent areas.

Further, the moving picture data decompression device according to the present invention divides each picture of the moving picture data into a plurality of blocks so as to shift the boundary between at least consecutive pictures, and divides the plurality of blocks into each of the plurality of blocks. Is a moving image data decompression device that decompresses compressed moving image data obtained by performing a predetermined compression process on pixel values of pixels of It has moving image data expansion means for performing a corresponding predetermined expansion processing to reproduce each of the blocks, and block moving means for moving the position of each of the reproduced blocks to restore the position in each of the images.

[0011]

In the moving picture data compression apparatus according to the present invention, each of the moving picture data images has a predetermined number, for example, eight.
A moving image data compression apparatus for performing compression processing by dividing into a plurality of square blocks including × 8 (64) pixels, wherein the block generation means blocks each image included in the moving image at the same position. In order to prevent the boundaries of No. 1 from always overlapping, for example, the positions of the boundaries of the blocks are shifted every other screen by 4 pixels in each of the vertical direction and the horizontal direction, and are equally divided to generate a block. The moving image data compressing unit performs a predetermined compressing process on pixel values of a plurality of pixels included in each block generated by the block generating unit,
That is, moving image data is compressed by performing orthogonal transform processing such as DCT, quantization processing, and variable length coding.

A moving image data decompression device according to the present invention is a device for decompressing the compressed image data generated by the above-described moving image data compression device according to the present invention, wherein the moving image data decompression means compresses the compressed image data. For the compressed moving image data corresponding to each of the previous original blocks, a predetermined decompression process corresponding to the compression process, that is, a process for decoding the compressed moving image data that has been variable length encoded (variable length decoding process), Inverse quantization processing and inverse discrete cosine transform (IDCT) are performed to reproduce each block before compression.

Since the moving picture data decompression means and the moving picture data compression means of the moving picture data decompression apparatus according to the present invention perform the same compression processing and decompression processing regardless of whether the boundaries of the blocks are deviated or not, The position of the block reproduced by the moving image data decompression means will deviate from the original position. Therefore, the block moving means moves the position of each reproduced block to restore the original position in each image.

[0014]

Embodiments of the present invention will be described below. Figure 1
It is a figure which shows the structure of the moving image data compression apparatus 1 which concerns on this invention. FIG. 2 is a diagram illustrating a method of dividing a block output from the frame memory circuit 10 shown in FIG.
As shown in FIG. 1, the moving image data compression apparatus 1 is composed of a frame memory circuit (FM) 10, a read control circuit 12, a compression encoding circuit 14 and a transmission circuit (TX) 16, and stores moving image data. This is a device that divides each of the constituent images into a plurality of square blocks including, for example, 8 × 8 (64) pixels, performs compression processing, and transmits the compressed data.

The read control circuit 12 controls the moving image data VIN.
A clock signal CK synchronized with the moving picture data V
The moving image data V stored in the frame memory circuit 10 is generated by outputting the read address RA based on the frame synchronization signal SYN synchronized with the image cycle (frame cycle) included in IN and is output to the frame memory circuit 10.
The order in which IN is output as the block S10 is controlled.
The frame memory circuit 10 receives the moving image data VIN in digital format, which is sequentially input from, for example, a video tape recorder, as a clock signal CK and moving image data V.
The data is stored in image (frame) units according to IN, divided into blocks in the order according to the read address RA input from the read control circuit 12, and sequentially output as block data S10 to the compression encoding circuit 14. That is,
The read control circuit 12 and the frame memory circuit 10 cooperate to perform a process of dividing moving image data into blocks.

The contents of the block data S10 output from the frame memory circuit 10 are as illustrated in FIG. As in this embodiment, the frame memory circuit 10
When the read control circuit 12 divides the image into blocks of 8 × 8 pixels, the read control circuit 12 outputs the image to be output in the frame period within the original image range as shown by the dotted line in FIG. The read address RA is generated so as to be divided and output in the following manner, and in the next frame period, as shown by the solid line in FIG. Then, the read address RA is generated.

Since the pixels included in the image are originally only within the range indicated by the dotted line in FIG. 2, the pixels outside the range indicated by the dotted line (empty region) are not included in the range indicated by the solid line in FIG. It will be. The read control circuit 12 uses a fixed value such as a numerical value 0 in these empty areas, pixels that are not included in the area indicated by the dotted line because the boundary is shifted, or pixels in the row area within the solid line that are empty. The read address RA is generated so that the pixels of the area adjacent to the area are added. The method of adding pixels to the empty area is
Further, it will be described later with reference to FIG.

The compression coding circuit 14 performs DCT processing on the block data S10 input from the frame memory circuit 10 in block units, and further performs quantization processing and variable length coding processing to obtain compressed moving image data S14. Output to the transmission circuit 16. The transmission circuit 16 sends the compressed moving image data S14 as transmission data TRD to a predetermined communication line.

FIG. 3 is a diagram showing the configuration of the moving picture data decompression device 2 according to the present invention. As shown in FIG. 3, the moving image data decompressing device 2 includes a receiving circuit (RX) 20, an expanding / decoding circuit 22, a frame memory circuit 24, and a read control circuit 26, and the moving image data compressing device 1 (see FIG. 1) The transmission data TRD transmitted from 1) is received, decompressed and decoded, and the original moving image data is reproduced to obtain the moving image data V.
Output as OUT.

The receiving circuit 20 is, for example, a clock signal CK and a frame synchronization signal SY supplied from a communication line.
The transmission data TRD transmitted from the moving image data compression device 1 is received using N, and is output to the decompression decoding circuit 22 as compressed moving image data S20. The decompression decoding circuit 22 is the compression encoding circuit 14 of the moving image data compression apparatus 1.
The opposite operation is performed. That is, the compressed image data S20 is subjected to a process (variable length decoding) that is the reverse of the variable length encoding performed in the compression encoding circuit 14 and a dequantization process to reproduce the blocks, and each of these blocks is subjected to IDCT. Then, the reproduced block data S22 corresponding to each block data S10 of the moving image data compression apparatus 1 is output to the frame memory circuit 24.

The read control circuit 26 includes a clock signal CK,
The read address RA is generated based on the frame synchronization signal SYN and the information indicating the blocking pattern in the transmission data, and the read address RA is output to the frame memory circuit 24, and the reproduction block data S22 stored in the frame memory circuit 24. The output order of the reproduced moving image data VOUT is controlled. The frame memory circuit 24 sequentially stores the reproduced block data S22 in image units according to the clock signal CK, the frame synchronization signal SYN, and the information indicating the pattern of block formation in the transmission data, and is input from the read control circuit 26. In accordance with the read address RA, the reproduced moving image data VOUT corresponding to the moving image data VIN in the moving image data compression device 1 is sequentially output.

The operations of the frame memory circuit 24 and the read control circuit 26 will be further described. In the moving image data compression apparatus 1, every other image is divided into blocks at the boundary shown by the solid line in FIG. 2, and the predetermined pixels are added to the region outside the range of the solid line and outside the range of the dotted line as described above. ing.

However, the compression encoding circuit 14 of the moving image data compression apparatus 1 does not discriminate whether it is a block divided as shown by the solid line in FIG. 2 or a block divided as shown by the dotted line. Further, since the decompression decoding circuit 22 of the moving picture data decompression device 2 similarly decompresses and decodes these blocks, the reproduced block data S22 should actually be at the position indicated by the solid line in FIG. The block will be included as the block at the position indicated by the dotted line. Therefore, it is necessary to return the reproduction block reproduced in such a shifted position to the position of the original image.

Therefore, the read control circuit 26 generates the read address RA for outputting the read block data S22 stored in the frame memory circuit 24 in such an order as to restore the original position on the screen, and the frame memory circuit 24
The output order of the reproduced moving image data VOUT is controlled.
In other words, in the moving image data decompressing device 2, it is necessary to perform a predetermined reverse operation to the frame memory circuit 10 and the read control circuit 12 of the moving image data compressing device 1. The frame memory circuit 24 and the read control circuit 26 cooperate to perform the processing for restoring the position of the reproduction block data S22 described here.

The operations of the moving picture data compression apparatus 1 and the moving picture data decompression apparatus 2 will be described below with reference to FIGS. FIG. 4 is a diagram showing a process of compressing / decompressing a block divided as shown by a dotted line in FIG.
(A) shows moving image data VIN or reproduced moving image data V
The image contained in OUT is shown, (B) is transmission data TR
The image contained in D is shown.

As shown in FIG. 4 (A), in the moving image data compression apparatus 1, the blocks divided by the frame memory circuit 10 and the read control circuit 12 as shown by the dotted lines in FIG. It is compression-encoded by the encoding circuit 14 and transmitted to the moving image data decompression device 2 via the transmission circuit 16. In the moving image data decompression device 2, the reception circuit 20 receives the transmission data TRD, and the decompression decoding circuit 22.
Decompresses and decodes the compressed moving image data S20 and outputs it to the frame memory circuit 24. Frame memory circuit 24
The read control circuit 26, as shown in FIG.
The reproduction block data S22 is output as reproduced moving image data VOUT without changing its position.

FIG. 5 is a diagram showing a process of compressing / decompressing a block divided as shown by a dotted line in FIG.
(A) is a frame memory circuit 10 and a read control circuit 1
2 shows the process of adding data to the empty area by
(B) shows an image included in the compressed moving image data S14 output from the compression encoding circuit 14, (C) shows processing by the frame memory circuit 24 and the read control circuit 26, and (D) shows a reproduced moving image. An image included in the data VOUT is shown.

In the moving image data compression apparatus 1, the frame memory circuit 10 and the read control circuit 12 divide the image as indicated by the solid line in FIG. 2, and the empty area is indicated by the arrow in FIG. a and b have regions c and
The block generated by adding the pixel of d is compression encoded by the compression encoding circuit 14 for each block shown in FIG. 5B, and the moving image data decompression device 2 is transmitted via the transmission circuit 16.
Be transmitted to.

In the moving picture data decompression device 2, the receiving circuit 20 receives the transmission data TRD and the decompression decoding circuit 22.
Decompresses and decodes the compressed moving image data S20 and outputs it to the frame memory circuit 24. Frame memory circuit 24
The read control circuit 26 returns the position of the reproduction block data S22 to the original position in the image as described above.

Further, the frame memory circuit 24 and the read control circuit 26 return the pixels added to the empty area to the original positions, as shown by the arrows in FIG. 5 (C). The reproduction block data S22 processed by the frame memory circuit 24 and the read control circuit 26 is output as reproduction moving image data VOUT corresponding to the original moving image data VIN, as shown in FIG. 5D.

FIG. 6 is a diagram for explaining the principle by which the moving image data compression apparatus 1 and the moving image data decompression apparatus 2 according to the present invention make the sense of discontinuity at the boundaries of blocks in the reproduced moving image data VOUT inconspicuous. , (A) exemplifies the pixel value of the moving image data VIN, and (B) shows the pixel value when the compressed moving image data S14 generated from the blocks divided as shown by the solid line in FIG. 2 is decompressed and decoded. (C) illustrates a pixel value when decompressing and decoding the compressed moving image data S14 generated from the blocks divided as shown by the dotted line in FIG. 2, and (D) illustrates (B). , (C)
7 shows the effect of the pixel value when the compressed moving image data S14 shown in FIG.

The pixel value included in the moving image data VIN is
For example, when the compressed moving image data S14 generated from the blocks divided as shown by the solid line in FIG. 2 is decompressed and decoded when distributed horizontally as shown in FIG. 6A,
Due to the influence of the quantization processing and the dequantization processing, FIG.
As shown in (B), discontinuity may occur at block boundaries. Similarly, the compressed moving image data S14 generated from the blocks divided as shown by the solid line in FIG.
In the case of decompressing and decoding, the block boundary may be discontinuous as shown in FIG. 6 (C).

However, according to the moving picture data compression apparatus 1 and the moving picture data decompression apparatus 2 according to the present invention, the reproduced moving picture data VOU as shown in FIGS. 6B and 6C is obtained.
T is output alternately for each image, and when this reproduced moving image data VOUT is displayed on the monitor device, from the eyes of the viewer, the reproduced moving image data VOUT shown in FIG.
The overlapping portions in (B) and (C) appear to be averaged as shown by the thick solid line in FIG. 6 (D). Therefore, the sense of discontinuity and the sense of distortion at the boundary portion of the block are less noticeable than in the case where the reproduced moving image data VOUT of only one of FIGS. 6B and 6C is displayed.

In the moving picture data compression apparatus 1 and the moving picture data decompression apparatus 2, the compression encoding circuit 14 and the decompression decoding circuit 22 are provided with the frame memory circuit 10 having a simple structure.
And the read control circuit 12 or only by adding the frame memory circuit 24 and the read control circuit 26, the device scale is larger than that of the conventional moving image data compression apparatus and moving image data decompression apparatus. It does not increase significantly.

The method of dividing an image into blocks, the number of pixels included in one block, and the like shown in the embodiments are examples, and can be changed according to the compression method. In the moving image data compression device 1,
The transmission circuit 16 is used to store the compressed moving image data S on a predetermined recording medium.
In the moving image data decompressing device 2, the receiving circuit 20 is replaced with a recording device that records 14, and the reproducing circuit that reproduces the compressed moving image data S14 from the recording medium in which the data is recorded by the moving image data compressing device 1. Alternatively, the compressed image data S14 may be input from the moving image data compression device 1 to the moving image data decompression device 2 via a recording medium.

If necessary, the compression / encoding circuit 14 of the moving picture data compression apparatus 1 and the moving picture data decompression apparatus 2 may be used.
The decompression / decoding circuit 22 may be directly connected. Further, when a block is generated from an image as shown by the solid line in FIG.
May be added. In addition to the examples described above, the moving image data compression apparatus 1 and the moving image data decompression apparatus 2 according to the present invention can have various configurations, for example, as in the modified example shown here.

[0037]

As described above, according to the moving image data compression apparatus of the present invention, each image included in moving image data is equally divided into predetermined blocks, and DC is divided into blocks.
Even if T or the like is performed, moving image data can be compressed so that the boundaries of blocks do not appear discontinuous in an image after decompression. Further, according to the moving image data decompressing apparatus of the present invention, it is possible to decompress the moving image data compressed by the moving image data compressing apparatus of the present invention, and moreover, it is not possible to reproduce the image at the block boundary of the reproduced image. It is possible to reduce the feeling of continuity and the feeling of distortion.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a moving image data compression apparatus according to the present invention.

FIG. 2 is a diagram illustrating a method of dividing a block output from the frame memory circuit shown in FIG.

FIG. 3 is a diagram showing a configuration of a moving image data decompression device according to the present invention.

FIG. 4 is a diagram showing a process of compressing / decompressing a block divided as shown by a dotted line in FIG. 2, in which (A) shows an image included in the moving image data VIN or the reproduced moving image data VOUT; (B) shows an image included in the transmission data TRD.

5 is a diagram showing a process of compressing / decompressing a block divided as shown by a dotted line in FIG. 2, in which (A) shows a vacant area by the frame memory circuit (10) and the read control circuit (12). The data addition processing is shown, and (B) shows the compressed moving image data S14 output from the compression encoding circuit.
Shows an image included in the video data, the (C) shows processing by the frame memory circuit (24) and the read control circuit (26), and the (D) shows an image included in the reproduced moving image data VOUT.

FIG. 6 is a diagram for explaining the principle by which the moving image data compression apparatus and the moving image data decompression apparatus according to the present invention make the sense of discontinuity at the boundaries of blocks in the reproduced moving image data VOUT inconspicuous. The pixel value of the moving image data VIN is illustrated, and (B) is the compressed moving image data S1 generated from the blocks divided as shown by the solid line in FIG.
4 exemplifies a pixel value in the case of decompressing and decoding No. 4 and FIG.
The pixel values when the compressed moving image data S14 generated from the blocks divided as shown by the dotted line are decompressed and decoded are illustrated, and (D) is the compressed moving image shown in (B) and (C). The effect that the pixel value when the data S14 is decompressed and decoded has on the visual sense of the viewer is shown.

FIG. 7 is a diagram exemplifying image data included in moving image data.

8A and 8B are diagrams showing a method of compressing image data by orthogonal transformation, in which FIG. 8A shows a block after division, and FIG. 8B shows frequency components obtained by orthogonally transforming the block.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Moving image data compression device, 10 ... Frame memory circuit, 12 ... Read control circuit, 14 ... Compression encoding circuit, 16
... transmission circuit, 2 ... moving image data decompression device, 20 ... reception circuit, 22 ... decompression decoding circuit, 24 ... frame memory circuit,
26 ... Read Control Circuit

Claims (4)

[Claims] 1. A moving image data compression apparatus for performing compression processing by dividing each image of moving image data into a plurality of blocks each including a predetermined number of pixels, wherein each of the images is moved in a predetermined cycle. A block generation unit that divides the boundaries in the horizontal direction and the vertical direction into the plurality of blocks and generates the blocks, and performs a predetermined compression process on the pixel values of the plurality of pixels included in each of the blocks. And a moving image data compression unit that compresses the moving image data. 2. The block generation means sets one of two consecutive images of the moving image data,
A first division method in which each of the plurality of blocks is a rectangle is equally divided into the plurality of blocks, and the other of the two continuous images of the moving image data is
Each of the plurality of blocks is the rectangle, and the boundaries between the blocks are divided into the plurality of blocks by a second division method different from the boundaries between the blocks in the first division method; The moving image data compression apparatus according to claim 1, wherein among the plurality of blocks of the image divided by the second division method, the pixels of the area of each of the blocks that partially protrudes from the image are supplemented. . 3. The block generating means is a pixel in an area protruding from the image of each of the blocks partially protruding from the image among the plurality of blocks of the image divided by the second dividing method. By adding pixels of the same image that are not included in any of the plurality of blocks as a result of being divided by the second dividing method, or are adjacent to the same image. The moving picture data compression apparatus according to claim 2, wherein the pixel of the area is added to compensate. 4. An image of moving image data is divided into a plurality of blocks so that boundaries between at least consecutive images are shifted, and a predetermined compression is performed on pixel values of a plurality of pixels included in each of the blocks. A moving image data decompression device for decompressing processed compressed moving image data, wherein predetermined decompression processing corresponding to the compression processing is performed on the compressed moving image data corresponding to each of the blocks, A moving image data decompression device having moving image data decompressing means for reproducing each of them and block moving means for moving the position of each of the reproduced blocks to restore the position in each of the images.