KR0178221B1 - Improved pattern vector encoding system using average value of pixels - Google Patents

Abstract

The present invention relates to an improved pattern vector coding system that can be easily implemented in hardware and that can realize high-speed pattern vector coding by using an average value of each pixel in a divided block for pattern coding. Is a coding means for pattern coding an image signal, the DCT block for converting time-domain image signals of the differential signal into DCT transform coefficients of a frequency domain of 8x8 unit using a cosine function, and A mean value extraction block for dividing the DCT transform coefficient blocks into a plurality of subblocks of a predetermined size, and then calculating an average pixel value for each subblock; and an average value of pixel values of each divided subblock for each subblock. Code table with reference patterns corresponding to input patterns obtained by subtracting the corresponding average pixel value from the extraction block By performing pattern matching by examining from, the pattern matching block generating index values for the reference patterns, the index values corresponding to the input patterns output from the pattern matching block, and the respective average values provided from the average value extraction block. It has a multiplexer that multiplexes the level values and provides them to the output side for transmission.

In addition, the present invention, as a means for removing the redundancy on the time axis between frames, the average pixel values from the average value extraction block, the index values of the reference patterns corresponding to the respective input patterns for these average pixel values and the reference A code table having a codeword of a pattern is used to reconstruct each of the index values into a DCT block having DCT change coefficients before pattern coding, and then to generate a differential signal from the current frame signal. And a block reconstruction block provided to the generated motion prediction and compensation block.

Accordingly, the present invention can realize high-speed pattern vector coding that is easy to implement in hardware and effective.

Description

Improved Pattern Vector Coding System Using Average Values of Pixels

1 is a schematic block diagram of a pattern vector coding system according to a preferred embodiment of the present invention.

2 is a detailed block diagram of a pattern coding block and a pattern decoding block employed in the pattern vector coding system of the present invention.

3 is a diagram showing an example of dividing one DCT block into four subblocks or 16 subblocks according to the present invention.

4 is a schematic block diagram of a typical conventional video encoding system using MC-DCT.

* Explanation of symbols for main parts of the drawings

100: first frame memory 110: subtractor

120: pattern coding block 122: DCT block

124: block division block 126: average value extraction block

128: pattern matching block 130: code table

140: pattern decoding block 142: block recovery block

144: IDCT block 150: the adder

160: second frame memory 170: motion prediction and compensation block

180: multiplexer

The present invention relates to an image coding system using pattern vector quantization. More particularly, the present invention relates to a pattern vector coding system enabling high-compression image coding using pattern vector quantization and motion estimation compensation.

As is well known in the art, the transmission of discrete video signals can maintain better image quality than analog signals. When a video signal composed of a series of image frames is represented in digital form, a considerable amount of data must be transmitted, especially for high-definition television (HDTV). However, since the usable frequency range of the conventional transmission channel is limited, in order to transmit a large amount of digital data, it is necessary to compress the data to be transmitted and reduce its transmission amount. In addition, the compressed video signal and the audio signal are encoded through different encoding techniques due to the characteristics of the signals. In this encoding, a compression technique of a video signal that generates a greater amount of digital data than an audio signal is particularly important. It can be said to take part.

Therefore, when transmitting a video signal, the transmitting side compresses and encodes the video signal using the spatial and temporal correlations of the video signal and then transmits the compressed and encoded video signal to the receiving side.

On the other hand, as the various compression methods mainly used for encoding the video signal, the hybrid coding method combining the stochastic coding method and the temporal and spatial compression method is known to be the most efficient.

Most hybrid coding schemes, which are one of the efficient coding schemes described above, use motion compensated DPCM (differential pulse code modulation), two-dimensional discrete cosine transform (DCT), quantization of DCT coefficients, variable-length coding (DPCM), and the like. Here, the motion compensation DPCM determines a motion of the object between the current frame and the previous frame, and predicts the current frame according to the motion of the object to generate a differential signal representing the difference between the current frame and the prediction value. These methods are, for example, Staffan Ericsson's Fixed and Adaptive Predictors for Hybrid Predictive / Transform Coding, IEEE Transactions on Communication, COM-33, NO.12 (1985, December), or A motion Compensated Interframe from Ninomiy and Ohtsuka. Coding Scheme for Television Pictures, IEEE Transactions on Communication, COM-30, NO.1 (January, 1982).

More specifically, the motion compensation DPCM predicts the current frame from the previous frame according to the motion of the object estimated between the current frame and the previous frame. Here, the estimated motion may be represented by a two-dimensional motion vector representing the displacement between the previous frame and the current frame.

In general, there are various approaches to estimating the pixel displacement of an object, and they are generally classified into two types, one of which is a motion estimation method in block units and the other is a motion estimation method in pixel units. In motion estimation, the block of the current frame is compared with the blocks of the previous frame to determine an optimal matching block, and then, from this, the interframe displacement vector (how much the block has moved between frames) with respect to the entire frame is transmitted. It is estimated.

Therefore, when transmitting a video signal, the transmitting side compresses and encodes the image signal in the buffer of the output side in order by taking into account the spatial and temporal correlation of the image signal in block units or pixel units through the above-described encoding technique. The video data is transmitted to the decoding system on the receiving side through the transmission channel at a desired bit rate in response to the channel request.

More specifically, the encoding system on the transmitting side removes spatial redundancy of the video signal by using transform coding such as discrete cosine transform (DCT), and further uses temporal encoding of the video signal by using differential coding through motion estimation and prediction. By eliminating redundant redundancy, the video signal can be efficiently compressed.

As described above, an example of a conventional coding system using a hybrid coding scheme that compresses and encodes a video signal by using temporal and spatial correlation for compression transmission to a receiver is shown in FIG. There is. As shown in the figure, a typical coding system includes a first frame memory 8, a subtractor 10, a DCT block 20, a quantization block 30, a local decoding block 40, a motion prediction block 50 And a variable length coding block (hereinafter referred to as VLC) 60 and a buffer 70.

In addition, as indicated by the dotted lines in FIG. 4, the local decoding block 40 is composed of an inverse quantization block 42, an IDCT block 44, and an adder 46, and the motion prediction block 50 includes a second frame. Memory 52, motion estimation block 54, and motion compensation block 56.

First, the subtractor 10 subtracts the predicted previous frame signal from the motion compensation block 56 in the motion prediction block 50 for motion compensation differential coding from the current frame signal provided from the first frame memory 8. As a result, the data, that is, the difference signal representing the difference pixel value, is converted into a series of quantized DCT transform coefficients through the DCT block 20 and the quantization block 30. Then, the quantized DCT transform coefficients are simultaneously provided to the inverse quantization block 42 constituting the VLC block 60 and the local decoding block 40.

More specifically, the DCT block 20 uses a cosine function to time-domain video signals (pixel data) for differential signals based on input motion estimation and prediction. Convert to In addition, the quantization block 30 is for quantizing a DCT transform coefficient from the DCT block 20 to a finite number of values through nonlinear operations. The quantization block 30 performs a difference signal between a frame to be encoded and a predicted frame. Quantize. In such quantization, the quantization block 30 is adjusted by the quantization parameter QP based on the fullness of the data from the output side buffer 70.

Accordingly, the VIC block 60 encodes the quantized differentially coded image data (quantized DCT transform coefficients) through the quantization block 30 into run and coefficient through zigzag scanning or the like as described above. In more detail, the VLC block 70 is variable according to the frequency of occurrence of each major code by using a code table, that is, a code having a large number of occurrences of a code is a short length code, and a code having a long length of a code has a small frequency of occurrence. After encoding, the signal is provided to the output side buffer 70 for transmission to the receiving side. Here, the reason for allocating different lengths to all codes through the VLC block 60 is to substantially increase the coding efficiency by reducing the average of the code lengths.

On the other hand, the inverse quantization block 42 and IDCT block 44 employed in a typical encoder as a means for performing the motion prediction differential coding as described above to form a local decoding block 40 and the DCT block 20 described above. The quantization block 30 restores the compression-encoded video signal (quantized DCT transform coefficient) to the original signal before being encoded for motion estimation and compensation, and provides the added signal to the adder 46, and thereafter, the adder 46 ) Adds the restored current frame signal (differential signal) provided from the IDCT block 44 and the predicted previous frame signal provided from the motion compensation block 56 to the second frame memory 52, thereby providing a frame memory. In 52, it is stored as a frame immediately before the restored current frame signal, that is, the current frame signal input for current encoding.

Thus, through this process, the previous frame signal stored in the second frame memory 52 is continuously updated with the image data immediately before the currently encoded input image data.

Next, the motion estimation block 54 is, for example, 16 × 16 units within a predetermined search range in the previous frame stored in the second frame memory 52 for the input frame from the first frame memory 8. The motion is estimated, i.e., the block of the previous frame most similar to the current frame is determined and provided to the motion compensator 56. The motion compensating block 56 outputs information from the motion estimation block 54 (search information). Based on this, the corresponding block of the previous frame is read from the frame memory 52 and provided to the subtractor 10 and the adder 46, respectively. In addition, although not shown in FIG. 1, the motion vector determined by the motion estimation block 54 is encoded through a predetermined encoding process and then sent to a transmitter (not shown) for transmission to the receiving-side decoding system.

As a result, the subtractor 10 subtracts the difference signal (differential pixel value) by subtracting the current frame signal from the first frame memory 8 with the predicted previous frame signal provided from the motion compensation block 56. The difference signal thus obtained is provided to the next stage DCT block 20 so that DCT and quantization for the difference signal as described above are performed.

Therefore, the DPCM / DCT hybrid coding scheme described above has a target bitrate of Mbps, and its application fields are CD-ROM, computers, home appliances (digital VCRs, etc.), broadcasting (HDTV), and the like. It is mainly concerned with MPEG-1,2 and H.261 coding algorithms for high bit rate encoding, which mainly consider only the statistical characteristics of block-by-block motion in the image, which has already been completed by the standard.

On the other hand, in recent years, due to the improvement and spread of PC performance, the development of digital transmission technology, the realization of high-definition display device, the development of memory device, various devices including home appliances can process and provide image information with huge data. In order to meet this demand, the transmission of audio-video data through the existing low-speed transmission path (PSTN, LAN, mobile network, etc.) and the limited capacity storage device to meet these demands. New storage techniques with high compression rates are needed for storage.

However, as described above, the conventional video coding system that removes temporal and spatial redundancy of video signals using MC-DCT (motion compensation DCT) can obtain some characteristics in terms of coding efficiency. Although it is possible, in practical implementations, there is a problem in that it is difficult to apply to a system requiring a high compression ratio as described above because of the limitation of the compression ratio.

In addition, although a pattern coding system has been proposed to apply a pattern coding method for dividing a DCT block into predetermined blocks and performing vector quantization in a typical hybrid coding system using a typical MC-DCT as described above, In this case, by searching for a pattern corresponding to each subblock without considering the average pixel value of each subblock, the effective pattern search is difficult, which is an obstacle to speedup for real-time processing under a desired compression ratio.

Accordingly, the present invention has been made in view of the above, and is an improved pattern vector that can be easily implemented in hardware and can realize high-speed pattern vector coding by using an average value of each pixel in a divided block for pattern coding. The purpose is to provide an encoding system.

In order to achieve the above object, the present invention provides a discrete cosine transform, a pattern of a differential signal between a current input frame signal to be encoded and a prediction frame signal through a motion prediction and compensation block based on the current frame signal and the immediately preceding frame signal. A pattern vector coding system for encoding the input frame signal at a desired compression rate by performing vector coding, wherein the time-domain image signals for the differential signal are transformed into DCT transform coefficients in a frequency domain of 8x8 units using a cosine function. A DCT block 122 for conversion; Average value extraction means (124,126) for dividing the transformed DCT transform coefficient blocks into a plurality of subblocks of a predetermined size and then calculating an average pixel value for each subblock; Pattern matching is performed by examining from the code table provided with reference patterns corresponding to the input patterns obtained by subtracting the divided pixel values of each subblock by the corresponding average pixel value from the average value extracting means for each subblock. Performing a pattern matching block 128 to generate index values for the reference patterns; The respective index values using a code table having average pixel values from the average value extracting means, index values of the reference patterns corresponding to respective input patterns for the average pixel values, and a codeword of the reference pattern A block reconstruction block (142) for reconstructing the data into the DCT block having the DCT change coefficients before the pattern coding and then providing the motion prediction and compensation block to the prediction frame signal; And multiplexing means (180) for multiplexing the index values corresponding to the input patterns output from the pattern matching block and the level values for the average values provided from the average value extracting means and providing them to the output side for transmission. An improved pattern vector coding system using an average value of pixels is provided.

The above and other objects and various advantages of the present invention will become more apparent from the preferred embodiments of the present invention described below with reference to the accompanying drawings by those skilled in the art.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, the pattern vector encoding to be used in the present invention has a code table and the data of each subblock divided into blocks of a predetermined size, for example, a 4x4 region, among the reference patterns (codewords) in the code table. Instead of outputting the coded bit values for each pixel of the corresponding subblock by selecting the most similar data, that is, the index value of the reference pattern most suitable for the input pattern of the divided subblock and the average pixel of the subblock. The method of encoding a value by using a pattern vector for each subblock divided into a predetermined size as an output, that is, has the biggest technical feature. Of course, according to the present invention, the encoded image data may be restored to the original signal through a reception signal decoding system having the same code table as that of the transmitting system.

1 is a schematic block diagram of an improved pattern vector coding system using average values of pixels according to a preferred embodiment of the present invention. As can be seen from the figure, the image coding system according to the present invention is a block for pattern-coding a motion-compensated differential signal using a portion blocked by a dotted line and denoted by reference numeral A, that is, a code table, for motion compensation. The remaining members 100, 110, 150, 160, and 170 except for the block portions 120, 130, and 140 for restoring the pattern coded image signal to the original signal are substantially corresponding to those shown in FIG. 4. The same element performing the same function of the constituent members is shown.

More specifically, the first frame memory 100 of FIG. 1 according to the present invention is the first frame memory 8 of FIG. 4 according to the prior art, the subtractor 110 to the subtractor 10, and the adder 150. Is the adder 46, the second frame memory 160 is the second frame memory 52, and the motion prediction and compensation block 170 corresponds to the motion estimation block 54 and the motion compensation block 56, respectively. . Therefore, since the functions of the constituent members substantially corresponding to each other have already been described in detail in the above-described prior art, the following will focus on the constituent parts A constituting the main technical features of the present invention.

Referring to FIG. 1, the pattern coding block 120 uses the respective codewords (or reference patterns) in the code table 130 to calculate the difference of the motion-compensated predetermined block unit input from the subtractor 110 through the line L21. Pattern-code the signal. That is, as shown in FIG. 2, the pattern coding block 120 includes a DCT block 122, a block division block 124, an average value extraction block 126, and a pattern matching block 128.

In FIG. 2, the DCT block 122 in the pattern coding block 120 is a component that performs substantially the same function as the DCT block 20 of FIG. A video signal (pixel data) in the time domain with respect to the difference signal between the current frame to be encoded and the prediction frame based on motion estimation and prediction provided by the subtractor 110 of FIG. Convert to DCT conversion factor in the frequency domain of.

Then, the DCT transform coefficients transformed in 8 × 8 block units are provided to the next block division block 124. Therefore, in the block division block 124, for the pattern vector encoding according to the present invention, the DCT transform coefficients of the 8x8 block are further divided into subblocks of a predetermined size, for example, as shown in FIG. As shown in FIG. 3, each 8 × 8 DCT transform coefficient block is divided into 4 subblocks of 4 × 4 size or each 8 × 8 DCT transform coefficient block is shown in FIG. Split into 16 subblocks of size 2x2. In this embodiment, it is assumed that each DCT block is divided into four subblocks of 4x4.

On the other hand, in the mean value extraction block 126, the absolute mean pixel value is calculated from the vector of the subblocks from the block division block 124 described above, i.e., as shown in FIG. When divided, all 16 pixel values a1 to a16 constituting one subblock are added together, and then divided by 16 to calculate an average pixel value (here, as shown in FIG. 3 (c), as 16 subblocks). In case of division, all four pixel values a11 to a14 constituting one subblock are added together and divided by 4 to calculate an average pixel value.The absolute average pixel values for each subblock calculated as described above are Simultaneously provided to the pattern matching block 128, the levels of the absolute average pixel values are provided to the output multiplexer (MUX) 180 via line L22.

Here, the reason for sending the level value of the absolute average pixel values for each subblock to the multiplexer 180 on the output side shown in FIG. 1 via the line L22 is according to the present invention in the decoding system on the receiving side at the remote location. This is because it is practically necessary when restoring the pattern vector coded video signal to the original signal in consideration of the average pixel value.

On the other hand, in the pattern matching block 128, the codeword most similar to the input pattern is the difference value obtained by dividing each pixel value of the corresponding subblock (input pattern) by the absolute average pixel value provided from the above average value extraction block 126. (Reference pattern), i.e., an optimal reference pattern having the lowest error value is determined for an input pattern corresponding to each subblock of the image, and then an index value for the determined reference pattern is determined through a line L23. The multiplexer 180 and the block reconstruction block 142 illustrated in FIG. Therefore, according to the present invention, dividing the pixel values in each sub-block by an absolute average pixel value leaves only pixel values above the absolute average value and the remaining pixel values become zero values, thereby enabling more reliable pattern symmetry.

That is, in the encoding system according to the present invention, as in the above-described conventional technique, 8 × for the entire image without compression and coding of the image signal by removing spatial and temporal redundancy of the image through MC-DCT, quantization, and VLC, etc. By dividing 8 DCT blocks into four 4 × 4 subblocks, and performing pattern vector encoding on each of the divided subblocks using an absolute average pixel value, an effective high compression rate image coding is possible. do.

Meanwhile, referring back to FIG. 1, the pattern decoding block 140 may use the original signal before the DCT and the pattern vector coded image signal are encoded through the pattern coding block 120 to predict motion between frames. As shown in FIG. 2, the pattern decoding block 140 includes a block reconstruction block 142 and an IDCT block 144.

In FIG. 2, in the block reconstruction block 142, as described above, each DCT block provided through the line L23 is divided into four 4x4 subblocks, respectively, and an absolute average for each of the divided subblocks. The index value for the reference vector pattern-coded using the pixel value and the absolute average pixel value provided from the average value extraction block 126 through the line L22 and the 8x before the pattern vector-coded using the code table 130. Restore to DCT transform coefficient blocks of 8. At this time, the code table 130 used in the pattern restoration block 142 is, as an example, as shown in FIG. 2, the code table used in the pattern matching block 128 in the pattern coding block 120 of FIG. May be used as a common code, or may be configured as a separate code table in consideration of the possibility of collision or waiting (priority setting), which may be caused when searching a reference pattern for an input pattern.

Accordingly, in the IDCT block 144, the original signal (pixel data) before the transform encoding is performed to estimate and compensate for the motion of each DCT transform coefficient of the reconstructed image frame provided from the block reconstruction block 142. Restore to and provide to the adder 150 shown in FIG.

The adder 150 then adds the reconstructed current frame signal (differential signal) provided from the IDCT block 144 and the predicted previous frame signal provided from the motion prediction and compensation block 170 via line L26 to add the first. By providing the second frame memory 160, the second frame memory 160 is stored as a frame immediately before the restored current frame signal, that is, the current frame signal input for current encoding.

Therefore, through this process, the previous frame signal stored in the second frame memory 160 is continuously updated with the image data immediately before the input image data currently encoded. The previous frame signal updated in this way is provided to the motion prediction and compensation block 170 through the line L25 for motion prediction and compensation, and the temporal redundancy between successive frames is removed through this series of processes.

Meanwhile, in the multiplexer 180, the level value of the absolute average pixel value for each subblocks provided from the average value extraction block 126 through the line L22 and each subblock provided from the pattern matching block 128 through the line L23. By multiplexing the index values of the reference pattern that is optimal for the input pattern to the input pattern and providing them to the transmitter (not shown), transmission of the compression-coded image data to the receiving side is performed.

As can be seen from the above, the pattern vector coding to be used in the present invention has a code table and is divided into blocks of a predetermined size, for example, a 4x4 region, among the reference patterns (codewords) in the code table. Instead of selecting the most similar data and outputting the coded bit value for each pixel of the subblock, the index value of the most optimal reference pattern is matched to the input pattern of the subblock. By outputting an average pixel value of each subblock, that is, by using a pattern vector for each subblock divided into a predetermined size, fast pattern vector coding can be implemented. Of course, according to the present invention, the encoded image data may be restored to the original signal through a decoding system on the receiving side having the same code table as the encoding system on the transmitting side.

As described above, according to the pattern vector coding system according to the present invention, a code table having a level value of an average pixel value and a plurality of predetermined reference patterns for each subblock obtained by dividing DCT blocks into a predetermined size is used. By performing compression encoding on each image frame, that is, by transmitting only the level value of the index value for the reference pattern corresponding to each subblock and the average pixel value for the subblock to the receiving side, an effective fast pattern vector encoding is performed. It can be realized.

On the other hand, in the above-described preferred embodiment of the present invention, each 8 × 8 DCT transform coefficient block is divided into four subblocks into four subblocks to perform pattern vector encoding using the average pixel values of each subblock and a code table. Although described, it should not be understood that the present invention is necessarily limited thereto. Therefore, the present invention does not perform the pattern vector coding by dividing each 8 × 8 DCT transform coefficient block into four subblocks, for example, by dividing it into 16 subblocks of 2 × 2 for each subblock. Even if the pattern vector quantization is applied using the average pixel values and the code table, it is possible to realize substantially the same result, that is, fast pattern coding.

Claims (5)

The desired input frame signal is encoded by discrete cosine transform and pattern vector encoding of the difference signal between the current input frame signal to be encoded and the prediction frame signal through the motion prediction and compensation block based on the current frame signal and the immediately preceding frame signal. A pattern vector coding system for encoding at a compression rate, comprising: a DCT block (122) for converting video signals in a time domain with respect to the differential signal into DCT transform coefficients in a frequency domain of 8x8 units using a cosine function; Average value extracting means (124, 126) for dividing the transformed DCT transform coefficient blocks into a plurality of subblocks of a predetermined size and then calculating an average pixel value for each subblock; Pattern matching is performed by examining from the code table provided with reference patterns corresponding to the input patterns obtained by subtracting the divided pixel values of each subblock by the corresponding average pixel value from the average value extracting means for each subblock. Performing a pattern symmetry block (128) for generating index values for the reference patterns; The respective index values using a code table having average pixel values from the average value extracting means, index values of the reference patterns corresponding to respective input patterns for the average pixel values, and a codeword of the reference pattern A block reconstruction block (142) for reconstructing the signal into the DCT block having the DCT change coefficients before the pattern coding and then providing the motion prediction and compensation block to the prediction frame signal; And multiplexing means (180) for multiplexing the index values corresponding to the input patterns output from the pattern matching block and the level values for the average values provided from the average value extracting means and providing them to the output side for transmission. An improved pattern vector coding system using the mean value of pixels. The system of claim 1, wherein the plurality of subblocks are four subblocks obtained by dividing the 8 × 8 DCT blocks into 4 × 4 units. 2. The improved pattern vector encoding system of claim 1, wherein the plurality of subblocks are 16 subblocks obtained by dividing each of the 8x8 DCT blocks into 2x2 units. The pixel table of claim 1, 2 or 3, wherein the code table for pattern matching and the code table for block reconstruction are one code table having the plurality of reference patterns. Improved Pattern Vector Coding System. The average value of the pixel of claim 1, 2 or 3, wherein the code table for pattern matching and the code table for block restoration are different code tables each having the same plurality of reference patterns. An Improved Pattern Vector Coding System Using.