JPH03124183A - Cell abort compensation picture decoding system - Google Patents

Abstract

PURPOSE:To reduce an error of a DC component of an aborted block at a reception side by using a data transmitted from a transmission side quantity relation information transmission means so as to apply compensation thereby reducing an error of the DC component of the block with respect to the block aborted by the reception side. CONSTITUTION:An abort compensation means 11 provided in the inside of a decoding section 15 uses quantity relation information transmitted from a quantity relation information transmission means 10 at the transmission side when cell abort exists to compensate a DC component of a block corresponding to the abort cell by using a DC component of the block corresponding to one preceding frame from the frame including the block as to a block having a small DC component difference among blocks corresponding to the aborted cell and by using the DC component of the block adjacent to the block in the frame including the block as to a block with a larger DC component.

Description

[Detailed Description of the Invention] [Summary] In an image data transmission system that encodes input image data using a plurality of pixels as a block and transmits it in units of cells, when cell discard occurs, a block corresponding to a discarded cell is provided. Regarding the cell discard compensation image decoding method that compensates for the DC component of the data of In an image data transmission system that encodes input image data in blocks of multiple pixels and transmits it in units of cells, the aim is to reduce errors in DC components on the receiving side even if cells are discarded. On the transmitting side of the data transmission system, the difference between the DC component of the block currently being transmitted and the DC component of the block corresponding to the block in a frame temporally one frame before the current frame including the block is determined in advance. A size-related information transmitting means for transmitting size-related information indicating whether the cell is larger or smaller than a threshold value to the receiving side of the image data transmission system is provided on the receiving side of the image data transmission system. According to the size relationship information sent from the size relationship information transmitting means when the cell is discarded, among the blocks corresponding to the discarded cell, a block with a small difference in the DC component is selected from the frame containing the block.
The DC component of the block corresponding to the block in the previous frame is calculated using the DC component of the block adjacent to the block in the frame including the block for which the difference in the DC component is large. It is configured to have a waste compensation means for compensating for the DC component.

[Industrial application field]

The present invention relates to an image data transmission system, and more specifically, the present invention relates to an image data transmission system that encodes input image data using a plurality of pixels as a block and transmits it cell by cell. The present invention relates to a cell discard compensation image decoding method that compensates for the DC component of data in a block.

[Prior Art] Recently, digitization of image equipment and transmission systems for image digital signals have been widely developed and researched in many application fields such as videophones. Generally, image data has a very large amount of information. For example, if current televisions were to be digitized as is, a transmission capacity of 100 Mb/s would be required, and image data has 1,500 times more information than audio. On a television, the screen changes 30 times per second, and frames change every 1/30 of a second. Pixels are generally arranged two-dimensionally on one screen corresponding to one frame,
Even if the level of one pixel is represented by 8 bits, for example, the amount of image data for one frame is extremely large.

However, even when frames are switched at 1/30 seconds on a television, the content of the pictures in each frame often does not change significantly. In other words, there are many parts of an image that remain static throughout, such as the sky and background, and there are few parts that change from frame to frame.Therefore, one method for compressing and encoding image data is the interframe coding method. .

In the interframe coding method, a difference in level data is calculated for each corresponding pixel between two temporally consecutive frames, and the difference is subjected to, for example, Huffman coding and sent onto a transmission path. If the level for most pixels does not change, the difference for those pixels will be 0, and the difference can be represented by one bit. Pixels with large level differences require codes of 8 bits or more, but by coding only these differences, the amount of information to be transmitted can be greatly compressed.

On the other hand, a method of encoding data within one frame as is is called an intraframe encoding method.

Regardless of whether interframe coding or intraframe coding is used, two operations, sampling and quantization, are basically required for digital representation of an image signal. There are two ways to sample an image; the first is to represent the image in terms of level values for discrete sequences of dots corresponding to pixels within a frame.

The second method is to perform orthonormal expansion of a level value function (image function) defined on the XY plane corresponding to the frame surface, and use the expansion coefficients as sample values.

FIG. 19 is an overall block diagram of an image transmission system using orthonormal transformation, which is the second method.

In Fig. 19, the input image is divided into blocks of 1 and 2
After being subjected to orthogonal transformation at step 3, it is quantized at step 3, and subjected to, for example, Huffman encoding at step 4, and then sent from the sending side to transmission line 5.

Then, on the receiving side, the signal from transmission line 5 is decoded by 6,
7 performs inverse orthogonal transformation, 8 blocks the inverse block, and outputs it as image data. The 122 and 1 blocks are two-dimensional KXK.
Instead of orthogonally transforming the data of each pixel (for one screen), we convert the data of about 4 x 4 to 16 x 16 pixels into one
This is done in order to shorten the conversion time by determining the conversion coefficients in block units.

Examples of the types of orthogonal transformations in 2 are Hadamard transform,
There are cosine transformations, Karhunen-Loewe transformations, etc.
In recent years, discrete cosine transform (DCT), which is a discrete representation of cosine transform and will be described later, is often used.

The orthogonal transform encoding method used in FIG. 19 focuses on the correlation between small areas within the screen, and orthogonal transform is performed using the pixels of the small area as a numerical string. The conversion coefficients obtained as a result correspond to frequency components, and in a single image signal that represents each component from low frequency to high frequency, the low frequency component is generally large in power and the high frequency component is small in power, so the low frequency component The amount of information to be transmitted can be reduced by performing quantization that allocates a large number of bits to high-frequency components and a small number of bits to high-frequency components.

FIG. 20 is a diagram showing a transform coefficient area after orthogonal transform is applied to image data in one block composed of 4×4 pixels. In this conversion coefficient area, generally the upper leftmost area 9 represents the DC component of the signal, and all the coefficients in the other areas represent AC components. The AC components correspond to each frequency from low frequency to high frequency, and the frequency becomes higher toward the lower right.

[Problem to be solved by the invention]

As described above, when multiple pixels are encoded and transmitted as one block, regardless of whether the encoding method is between frames or within a frame, whether the transformation coefficients after orthogonal transformation are encoded or the pixel data itself is encoded. Regardless of whether it is encoded, image data is transmitted on a transmission path in the form of fixed-length packets of several bytes to which a header indicating destination information called a cell is added, for example.

However, if a header error occurs or the number of cells increases rapidly to exceed the processing capacity of the network, cells may be discarded within the network, that is, cell discard may occur. In such a case, there is a problem that the image deteriorates in the image data receiving section. In particular, there has been a problem in that when there is a large difference in DC component between a block in the frame currently being transmitted and a corresponding block in the temporally previous frame, the image quality deteriorates significantly.

The present invention uses information indicating whether the difference in DC components between the block in the frame currently being transmitted and the block in the previous frame is large or small, so that even if cell discard occurs, the DC component on the receiving side is The purpose is to minimize component errors.

[Means to solve the problem]

FIG. 1 is a block diagram showing the principle structure of an image data transmission system using the cell discard compensation image decoding system of the present invention.

The configuration of this block diagram is the same for the first invention to the fifth invention. In FIG. 1, input image data is blocked by a blocking unit 12, encoded by an encoding unit 13, transmitted over a transmission path 14, decoded by a receiving side decoding unit 15, and deblocked. The image data is deblocked in a section 16 and output as image data. Note that when the image data is orthogonally transformed and the orthogonal transform coefficients are encoded and transmitted, the orthogonal transform is performed on the output of the blocking section 12, and the inverse orthogonal transform is performed on the output of the decoding section 15. It is done against.

Size-related information transmitting means 10 provided on the transmitting side in FIG.
In the first invention, the difference between the DC component of the block currently being transmitted and the DC component of the corresponding block in the frame one frame before the current frame including that block is greater than a predetermined threshold. For example, the magnitude relationship information indicating whether the
4 to the receiving side.

Further, on the receiving side of the image data transmission system, the discard compensating means 11 provided inside the decoding unit 15, for example, detects the presence or absence of cell discard on the transmission path 14, and when there is cell discard, the discard compensation means 11 is installed on the receiving side of the image data transmission system. Using the size relationship information sent from the size relationship information transmitting means 10, among the blocks corresponding to the discarded cell, for a block with a small difference in DC components, a corresponding block one frame before the frame including the block is selected. The DC component of the block corresponding to the discarded cell is compensated for using the DC component of the block corresponding to the discarded cell, and for blocks with a large difference in DC component, the DC component of the block adjacent to that block within the frame including that block.

Note that the DC component for a block composed of a plurality of pixels is the 20th DC component when encoding orthogonal transform coefficients.
As explained in the figure, this is the upper leftmost coefficient in the transform coefficient area, and when data in the pixel area is encoded, it is the average of data for all pixels within one block.

In the second invention, the magnitude relationship information transmitting means 10 transmits magnitude relationship information indicating the magnitude relationship between the DC component difference and the threshold value, and the current frame information for the block for which the DC component difference is greater than the threshold value. Compensation information obtained by subjecting the DC component of the block to, for example, coarser quantization than that by the encoding unit 13 in FIG. 1, is transmitted to the receiving side via the transmission path 14.

Along with this, the discard compensating means 11 on the receiving side uses the DC component of the block in the current frame transmitted from the size relation information transmitting means 10 to deal with the discarded cell for the block with a large difference in the DC component described above. Their functions are the same as in the first invention, except that they compensate for the DC component of the block.

In the third invention, the magnitude relation information transmitting means 10 roughly quantizes the value of the DC component difference for the block whose DC component difference is larger than the threshold value, and transmits the same to the receiving side. Then, the discard compensating means 11 uses the DC component difference transmitted from the size relation information transmitting means 10 for blocks with a large difference in DC components and the DC component of the corresponding block of the 1" previous frame to determine which cells are to be discarded. except that it compensates for the DC component of the corresponding block.
The effect is the same as in the first invention.

In the fourth invention, the magnitude relation information transmitting means 10 transmits information indicating the magnitude relation between the DC component of the block currently being transmitted, the difference between the DC component of the block and the adjacent block within the same frame, and the threshold value; For a block with a large difference in DC components, the result of rough quantization of the DC component values of that block is transmitted to the receiving side.

Along with this, the discard compensating means 11 determines, based on the magnitude relation information sent from the magnitude relation information transmitting means 10 when a cell is discarded, a block having a small difference in the DC component among the blocks corresponding to the discarded cell. For blocks, the block corresponding to the discarded cell is determined using the DC component of the block adjacent to the block in the frame containing the block, and for blocks with a large difference in DC components, the DC component sent from the size relation information transmitting means 10 is used to determine the block corresponding to the discarded cell. Compensates for the DC component of

In the fifth invention, the size-related information transmitting means 10 transmits the DC component of the block and the block within the same frame for the block in which the difference between the DC component and the threshold value between the block currently being transmitted and the adjacent block is large. The point is that a value obtained by roughly quantizing the difference between the DC component of an adjacent block is transmitted to the receiving side, and accordingly, the discard compensation means 11 detects a block with a large difference in DC component, if the block is adjacent to that block within the same frame. The operation of the fourth invention is similar to that of the fourth invention, except that the DC component of the block corresponding to the discarded cell is compensated for using the difference between the DC component of the block corresponding to the discarded cell and the DC component transmitted from the size relation information transmitting means 10. are the same.

[For production]

In the present invention, in the image data transmission system,
For example, even when cell discard occurs on the transmission path, the data sent from the size-related information transmitting means 10 on the transmitting side is used to calculate the DC component of the block discarded on the receiving side by the discard compensating means 11. Compensation is performed to reduce the error. In the first invention, only the magnitude relationship with respect to the threshold value of the difference between the DC components of the block being transmitted and the corresponding block in the previous frame is transmitted to the receiving side, and the receiving side discards the block. For a block with a small difference in DC component, the DC component of the corresponding block in the previous frame is used as is. Also, for blocks with large differences in DC components,
The blocks adjacent to that block in the current frame,
For example, the DC component of the block directly above is used.

In the second invention, in addition to the information indicating the magnitude relationship, for blocks with a large difference in DC components, the value of the DC component of the block in the current frame is transmitted to the receiving side, and the receiving side transmits the value of the DC component of the block in the current frame. For blocks with small differences, the first
As in the invention, the DC component of the corresponding block in the previous frame is used, and for blocks with a large difference in DC component, the transmitted DC component value is used.

In the third invention, instead of the value of the DC component of the block in the current frame in the second invention, the difference in DC components is transmitted to the block with a large difference, and the receiving side receives the transmitted DC component. The sum of the difference and the DC component of the corresponding block of the previous frame is output.

In the fourth and fifth inventions, the difference in DC component is taken between the block currently being transmitted and a block adjacent to that block within the same frame, for example, a block directly above it,
Information indicating the magnitude relationship between the difference and the threshold, and in the fourth invention, the DC component value itself of the block for a block with a large difference, and in the fifth invention, the block with a large difference and the block adjacent thereto. The difference in DC component between the two is transmitted to the receiving side. On the receiving side, for a block with a small difference in DC components, the DC component of the block adjacent to the block, for example, directly above it, in the frame containing the block is used, and for blocks with a large difference, the fourth invention In this case, the transmitted DC component is used. Further, in the fifth invention, the sum of the difference between the transmitted DC components and the DC component of a block that is instantaneously connected to the block in the same frame, for example, directly above it, is output.

As described above, according to the present invention, even when cell discard occurs, the error on the receiving side of the DC component of the discarded block is reduced.

〔Example〕

FIG. 2 is a conceptual diagram of an embodiment of the first invention in the orthogonal transform coefficient domain. In the figure, only the upper leftmost coefficient of the orthogonal transform coefficient region, that is, the DC component is shown. FIG. 5A shows the coefficient area after decoding in the normal case where cell discard does not occur, and FIG. 2C shows the coefficient area after decoding when cell discard occurs. Here, it is assumed that the fourth and fifth blocks are discarded due to cell discard, and magnitude relationship information is received from the transmitting side indicating that the difference in DC component is large for the fourth block and small for the fifth block. Assuming that the value of the peripheral block in the current frame is used for the fourth block, the value of the DC component of the block directly above is used for the fourth block, and the value of the DC component of the block directly above is used for the fourth block, and the value of the DC component of the block directly above is used for the fourth block. The DC component of the corresponding plot in the frame is used. In the figure, the upper row shows the first to third blocks, and the lower row shows the fourth to sixth blocks.

FIG. 3 is a conceptual diagram of an embodiment of the first invention in a pixel area. In the figure, only the average value of pixel data in each block, that is, the DC component, is shown for the first to sixth blocks. Figure 2 (a) shows the DC component of each block after decoding in the normal case where no cell discard occurs, that is, the average value, and Figure 2 (b) shows the DC component, that is, the average value, when cell discard occurs. The average value in the pixel area after decoding processing is shown in the figure. Similarly to FIG. 2, it is assumed that the fourth and fifth blocks are discarded, and the difference in magnitude of the DC component is large for the fourth block (and small for the fifth block). For the fifth block, the average value of the block immediately above in the current frame is used, and for the fifth block, the average value of the pixel data of the corresponding block in the previous frame is used.

FIG. 4 is a block diagram of the overall configuration of a pixel data transmission system using the first invention in the orthogonal transform coefficient domain.

In the figure, the transmitting side uses the output of the subtraction unit 22, which will be described later, to determine whether the difference in DC component between the block currently being transmitted and the corresponding block in the previous frame is larger or smaller than a predetermined threshold. A size determination unit 19 that transmits size relation information indicating the size to a discard compensation unit 30 on the receiving side, which will be described later, a DCT unit 20 that performs discrete cosine transformation (DCT), which is a type of orthogonal transformation, on input image data, and a DC
The output ■S of the T section 20 and the corresponding block data ■S in the previous frame stored in the frame memory 21
A subtractor 22 that takes the difference between An adder 25 adds the output S of the converter 24 and the block data S in the previous frame, which is the output of the frame memory 21.The output S of the adder 25 is multiplied by the leakage coefficient α and output to the frame memory 21. It consists of a leak coefficient multiplier 26.

Also, on the receiving side, the signal ■S input via the transmission path
An inverse quantizer 27 that returns the quantized input data to its original format ■R, and an addition that takes the sum of its output ■R and the block data ■R in the previous frame stored in the frame memory 28 a discard compensator 30 corresponding to the discard compensator 11 in FIG. 1; an inverse DCT unit 31 that performs inverse discrete cosine transform on the output R of the discard compensator 30 and outputs image data in the pixel domain; and Disposal Compensation Department 3
0 output ■R multiplied by leak coefficient α frame memory 2
It consists of a leak coefficient multiplier 32 which outputs to 8.

Here, the discrete cosine transform (DCT) uses a transformation matrix created from a cosine function, and it required a multiplier for calculation, making it difficult to perform high-speed calculations, but with the development of high-speed signal processing processors, stationary It is now widely used for semi-video transmission such as image transmission and video conference signals.

FIG. 4 shows an embodiment of the present invention using an interframe encoding method. For example, on the transmitting side, block data S in the frame to be transmitted currently and block data S in the previously transmitted frame are The difference, that is, the difference between frames is ■
S, which is quantized and, for example, Huffman encoded, and output onto a transmission path. On the receiving side, the difference ■R is received from the transmission path, and the block data ■R in the previous frame is
By calculating the sum with , block data ■R in the current frame is obtained. And on the sending side, the difference ■
The adder 25 calculates the sum of S and the block data S in the previous frame, multiplies it by a leakage coefficient α, and then stores it in the frame memory 21 in order to obtain the inter-frame difference for the next frame. Also in the receiving section,
Similarly, the current block data ■R is multiplied by the leakage coefficient α and then stored in the frame memory 28.

For example, the leak coefficient α is used to attenuate the effect of cell discard in a short time even if cell discard occurs. Block data FM
The sum (DH+FMo) is multiplied by the leakage coefficient α, and the block data for the current frame expressed by the following formula is the content F of the frame memory 28.
Assume that it has become M1.

FM+ = (D+ +FMo) α= D t α+
FM, α Then, when the difference D2, which is the block data between the next frame, is input from the transmission path, the frame memory 28
The content of is as follows.

FMz=(Dz+FM+)α=D2α+D1α”+FMoα2 Furthermore, when block data D3 between the next frames is input, the contents of the frame memory 28 are given by the following equation.

FM3 = (Ih +FM2) α = [): + a: +D2 rx2+D+ at” +
FMOa3 For example, the influence of prediction error on the entire image when cell discard occurs is attenuated to α7 after inputting n frames. Therefore, by setting α smaller than 1, the influence can be reduced to almost 0 within a short time.

FIG. 5 is an example of DC component data on the transmitting side and the receiving side. FIG. 4(a) shows an example of data on the transmitting side, and as explained in FIG.
Data for the previous frame stored in ■
The difference from S is taken by the subtractor 22, resulting in ■S. Here, the threshold value for determining the magnitude of the DC coefficient is ±8, and if it is larger than the threshold value, it will be represented by the code 1, and if it is smaller than the threshold value, it will be represented by the symbol O, then the magnitude relationship information transmitted to the receiving side will be the first to sixth values. The block order becomes "000100". Here, among each block data, the upper stage is assumed to be the first to third blocks, and the lower stage is assumed to be the fourth to sixth blocks.

FIG. 5(c) shows an example of data on the receiving side.

As explained in Fig. 4, on the receiving side, the received data ■R and the data ■R for the previous frame are summed, and this is taken as the data ■R for the current frame. Assuming that the DC components for the fourth and fifth blocks are lost due to cell discard, the effect of the discard compensator 3° will cause a large difference in the transmitted magnitude relationship information for the fourth block. DC component 3 of the block directly above in the current frame to show
0, and since the difference is small for the fifth block, the DC component 14 of the corresponding block in the previous frame is used.

Comparing these with the data of 1S in FIG. 5(a), in this case, the difference in the fourth block from the DC coefficient lost due to cell discard is still quite large. This is because the compensation for the large difference between the two blocks is simply made using the DC coefficients of adjacent blocks within the current frame.

FIG. 6 is a block diagram of the overall configuration of an embodiment of an image data transmission system using the first invention in the pixel area.

Comparing this figure with the embodiment in the orthogonal transform coefficient domain of FIG. The operation is exactly the same as in Fig. 4, except that it is performed using the data itself, and the DC component for each block is given as the average value of the plurality of pixel data composing each block. .

FIG. 7 is a flowchart of an example of the processing performed by the size determining unit 19 in FIGS. 4 and 6. FIG. In the figure, when the process is started, first, in step 33, the DC component is separated. This separation of DC components is performed in the pixel domain by calculating the average value of pixel data for each pixel in the block. The difference from the DC component of the corresponding block is taken, and in step 35 it is determined whether the difference is greater than a predetermined IJ value. If the difference in DC components is larger than the threshold, it is sent in step 36 that the difference is large, and if it is small, it is sent in step 37 that the difference is small, and the process ends. The above processing is performed for each block.

FIG. 8 is a flowchart of a processing example of the discard compensation unit 30. When the process starts in the figure, first, in step 3, it is determined whether or not cells are to be discarded. If no cell discard occurs, the discard compensator 30 does not perform any processing and outputs the input from the adder 29 as it is to the inverse DCT unit 31 in FIG. 4, for example.

If it is determined in step 3 that a cell has been discarded, then in step 39 the DC component of the block corresponding to the discarded cell is separated. This separation is performed for blocks with small differences in DC components for corresponding blocks in the previous frame, and for blocks with large differences in adjacent blocks in the current frame. Then, in step 40, the DC component is compensated, that is, for blocks with a small difference, the DC component of the corresponding block in the previous frame is used, and for blocks with a large difference, the DC component of the adjacent block in the current frame is used. The process ends.

FIG. 9 is a conceptual diagram of an embodiment of the second invention in a pixel area. Figure (a) shows the average value of pixel data of each block after normal decoding when no cell discard occurs, and (b) shows the average value after decoding when cell discard occurs. It shows. Here, similarly to FIGS. 2 and 3, it is assumed that the fourth block and the fifth block are discarded, and it is assumed that the difference between the fourth block and the fifth block is small in the first invention. In addition to the magnitude relationship information shown in FIG. 1, it is assumed that a value (ave4) obtained by roughly quantizing the DC component of the fourth block in the current frame is transmitted from the transmitting side to the receiving side as compensation information.

Therefore, the value of the compensation information sent from the transmitting side is used for the fourth block in FIG. 5(b) that has been discarded, and the corresponding block in the previous frame is used for the fifth block. The average value of is used.

The overall configuration of the embodiment of the image data transmission system in the second invention is the same as that in the first embodiment of FIGS. This is the same as the embodiment in the invention.

FIG. 10 is an example of DC component data for each block on the transmitting side and receiving side in the second invention. This figure corresponds to FIG. 5 in the first invention, and its meaning is exactly the same as that of FIG. 5. In the same figure, (a) is the data on the sending side, (b)
) is an example of data on the receiving side. On the transmitting side, the difference between data ■S and ■S is calculated, and as a result, ■S is transmitted to the receiving side as an interframe code, but at the same time, the difference between the magnitude relation information and the DC component is The DC component within the current frame is roughly quantized and transmitted to the receiving side.

As in Fig. 5, if the threshold value for size judgment is ±8, and if it is larger than the threshold value, it will be represented by the code 1, and if it is smaller than the threshold value, it will be represented by the code O, then the information sent to the receiving side will be 0001 (sign representing 8), 00 After the code 1 indicating that the difference in the DC component of the fourth block is large, a code representing the value 8 of the DC component in the current frame of the fourth block is inserted.

On the receiving side of the same figure (b), the value 8 received from the transmitting side is applied to the discarded fourth block, and the value 8 received from the transmitting side is applied to the discarded fourth block, and the DC component of the corresponding block in the previous frame is applied to the fifth block. A value of 14 is used.

FIG. 11 is a flowchart of a processing example of the size determining section in the second invention. In the figure, when the process starts, first, in step 41, the DC component is separated. In the pixel domain, separation of this DC component is performed by taking the average value of each pixel data in each block. Next, in step 42, the difference between the DC component of the block in the current frame and the DC component of the corresponding block in the previous frame is taken, and in step 43 it is determined whether the difference is greater than a predetermined threshold. It will be judged. If the difference is smaller than the threshold in step 43, step 44
The fact that the difference is small is transmitted to the receiving side, and the process ends.

If the difference is greater than the threshold in step 43, the large difference is transmitted to the receiving end in step 45, and in step 46, the value of the DC component of the block in the current frame is coarsely quantized and sent to the receiving end. It is sent and processing ends.

FIG. 12 is a flowchart of a processing example of the discard compensation section in the second invention. In the figure, when the process starts, it is first determined in step 47 whether or not cells have been discarded, and if no cells have been discarded, the discard compensator outputs the input data as is without performing any processing.

If it is determined in step 47 that a cell has been discarded, then in step 48, the DC component is separated, that is, the DC component of the corresponding block in the previous frame with respect to the block with a small difference is determined. Next, in step 49, the DC component is compensated, that is, for blocks with a small difference, the DC component in the previous frame obtained in step 48 is used, and for blocks with a large difference, the value of the DC component sent from the transmitting side is used. used to end the process.

FIG. 13 is a conceptual diagram of the third embodiment of the invention in the pixel area. In the third invention, unlike the second invention, compensation is provided from the transmitting side to the receiving side for blocks in which there is a large difference between the DC component of the block in the current frame and the DC component of the corresponding block in the previous frame. The difference in DC components is roughly quantized and transmitted as information.

Comparing FIG. 13 with FIG. 9 for the second invention,
(b) The fourth cell after decoding process when cell discard occurs
Compensation information sent from the transmitting side as data for the block, i.e. the value of the difference between the DC component of the block in the current frame and the DC component of the corresponding block in the previous frame, and the corresponding block in the previous frame. The results are exactly the same, except that the sum with the DC component is used.

The embodiment of the image data transmission system in the third invention is completely similar to FIGS. 4 and 6 for the first invention. Further, the embodiment of the DC component data of each block on the transmitting side and the receiving side is exactly the same as that shown in FIG. 10, except that the code representing the difference in DC components is sent from the transmitting side as compensation information.

Furthermore, the flowchart of the processing example of the size determination section is also shown in the first section.
They are exactly the same except that in step 46 of Figure 1, instead of the value of the DC component of the block in the current frame, the difference from the corresponding DC component in the previous frame is transmitted, and the discard compensator's In the processing embodiment, in step 48 of FIG. 12, the DC components of the corresponding blocks in the previous frame are separated even for blocks with large differences in DC components;
They are exactly the same except that in step 49, the value of the auxiliary information sent from the transmitting side for the block with a large difference is added to the value of the DC component in the previous frame determined in step 48.

FIG. 14 shows the fourth invention in the coefficient domain, that is, information indicating the magnitude relationship between the DC component of the block currently being transmitted and the DC component of the block adjacent to that block within the same frame, and the difference between the DC components is a threshold value. FIG. 6 is a conceptual diagram of an embodiment in which, for a larger block, a roughly quantized value of the DC component of the block is sent to the receiving side.

Figure (a) shows the DC component data after normal decoding when no cell discard occurs, and (opening) shows the DC component of the block data after decoding when cell discard occurs. ing.

It is assumed that the 4th and 5th blocks have been discarded in the same way as in Figure 2, and the difference between the 4th block and the adjacent block, for example the block directly above it, is large, and the difference between the 5th block and the 5th block is small. Assuming that for the fourth block, a value obtained by roughly quantizing the DC component of that block is sent from the transmitting side as compensation information.
For the block, the value of the sent compensation information is used, and for the fifth block, the DC component of the block directly above in the same frame is used.゛When implementing the fourth invention in the pixel domain, the upper left coefficient in the coefficient domain, that is, the average value obtained by averaging the pixel data of all pixels in the block instead of the DC coefficient, is used as the DC component. The concept is exactly the same as that in FIG. 14, except that

The overall configuration of the embodiment of the image data transmission system using the fourth invention is substantially the same as that shown in FIGS. 4 and 6, so a description thereof will be omitted. FIG. 15 is an example of DC component data on the transmitting side and the receiving side in the fourth invention. In the fourth invention, since only blocks within the same frame matter, FIG. 15 shows only the block data ``S'' being transmitted and the data ``R'' on the receiving side.

On the transmitting side of (a), for the fourth to sixth blocks,
Using a threshold value of ±8, the code is set to 1 when it is larger than the threshold value, and the code is set to O when it is smaller than the threshold value. )00 to the receiver. Here, the first 1 indicates that the difference with respect to the fourth block, that is, the difference with the first block directly above, is greater than the threshold value;
A code representing the block's DC coefficient of 8 is inserted.

Assuming that cell discard occurs and the fourth and fifth blocks are lost on the receiving side of (Fig. 15), the auxiliary information 8 sent from the transmitting side for the fourth block is
Also, for the fifth block, the coefficient value of the block directly above is 2.
5 is output as the DC component of the block for the current frame.

FIG. 16 is a flowchart of a processing example of the size determining section in the fourth invention. In the figure, when the process starts, first, in step 50, the DC component is separated. For example, in a pixel area, the average value of data of a plurality of pixels constituting a block is taken as the DC component. Next, in step 51, the difference between the DC component of the current block and the DC component of an adjacent block, for example, the block directly above, is calculated, and in step 52, it is determined whether the difference is greater than a certain threshold value.

If the difference is smaller than the threshold, the fact that the difference is small is transmitted to the receiving side in step 53, and the processing for that block ends. If the difference is greater than the threshold in step 52, it is determined in step 54 that the difference is large, and in step 55, the value of the DC component of the current block is coarsely quantized and transmitted, and the process ends.

FIG. 17 is a flowchart of a processing example of the discard compensation section in the fourth invention. In the figure, when the process starts, it is first determined in step 56 whether or not cells are discarded. If there is no cell discard, the discard compensator outputs the input data as is without performing any processing.

If cell discard occurs, the DC component is separated in step 57, and in step 58, it is determined whether the difference from the DC component of the adjacent block is large based on the magnitude relationship information from the transmitting side, and the difference is determined. If it is smaller, compensation is performed in step 59 using the DC component of a block adjacent to the discarded block, for example, a block directly above it, and the process ends. If the difference is large in step 58, the compensation information is used to compensate the DC component of the block in step 60, and the process ends.

FIG. 18 shows the fifth invention, that is, the case where instead of the DC component of a block with a large difference between adjacent blocks as auxiliary information in the fourth invention, the difference between the DC components of the adjacent blocks is sent to the receiving side. It is a conceptual diagram of an example of. Same figure (
A) is the DC component data after normal decoding when no cell discard occurs, and g) is the decoding when the 4th and 5th blocks are discarded, as in Figure 2 etc. This is DC component data after oxidation processing.

In the figure, the fourth block has information indicating that the difference in DC component with the adjacent block, for example, the block directly above, is large, and the fifth block has information indicating that the difference is small, and compensation information for the fourth block, that is, the fourth block. It is assumed that the difference between the DC component of the first block directly above and the DC component of the first block directly above is sent to the receiving side, and for the fourth block, the receiving side calculates the value of the DC component of the first block and the received compensation information,
In other words, the sum of the roughly quantized value of the difference in DC components between the fourth block and the first block is used, and for the fifth block, the DC component value of the block directly above, that is, the second block, is used as is. ing.

Regarding the fifth invention, the difference between the DC component of that block and the DC component of the adjacent block is received as compensation information from the transmitting side, targeting an adjacent block, for example, a block that has a large difference from the block directly above. The operation is the same as in the fourth invention except that the data is sent to the side, and the embodiment of the image data transmission system, the embodiment of DC component data, the processing embodiment of the magnitude determination section, and the discard compensation section. Description of processing examples will be omitted.

In the above explanation, using the embodiments of the image data transmission system shown in FIGS. 4 and 6, the embodiments were mainly explained using the interframe encoding method, but the encoding method is limited to this. It is also possible to use an intra-frame encoding method without having to do so. Regarding the quantization of the compensation information transmitted to the receiving section, for example, the quantizer 2 in FIG.
3. This is because the compensation information is used only when cell discard occurs, and the same level of quantization as normal quantization is used. Of course you can go. Furthermore, in the first, fourth, and fifth inventions, the block directly above is used as an example of the adjacent block, but it goes without saying that the adjacent block is not limited to this.

〔Effect of the invention〕

As described in detail above, according to the present invention, even when cell discard occurs, the error in the DC component of the block can be reduced, and the influence of cell discard on the image quality in the receiving section can be reduced. .

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the principle configuration of the image data transmission system according to the present invention. FIG. 2 is a diagram showing the concept of an embodiment of the first invention in the coefficient domain. FIG. 3 is a diagram showing the concept of the embodiment of the first invention in the pixel domain. 4 is a block diagram showing the overall configuration of an embodiment of an image data transmission system using the first invention in the coefficient domain. FIG. FIG. 6 is a block diagram showing the overall configuration of an embodiment of an image data transmission system using the first invention in the pixel domain; FIG. 7 is a diagram showing an embodiment of DC component data on the receiving side; FIG. FIG. 8 is a diagram showing a flowchart of a processing example of the size determination unit in the first invention; FIG. 9 is a flowchart of a processing example of the discard compensation unit in the first invention; FIG. 9 is an embodiment of the second invention in the pixel area 10 is a diagram showing an example of the DC component data on the transmitting side and the receiving side in the second invention. FIG. 11 is a flowchart of the processing example of the magnitude determination section in the second invention. 12 is a diagram showing a flowchart of a processing example of the discard compensation unit in the second invention, FIG. 13 is a diagram showing the concept of the embodiment of the third invention in the pixel area, and FIG. 14 is a diagram showing the coefficient FIG. 15 is a diagram showing an example of DC component data on the transmitting side and receiving side in the fourth invention. FIG. 16 is a diagram showing the concept of the embodiment of the fourth invention in the area. FIG. 17 is a diagram showing a flowchart of a processing example of the discard compensation section in the fourth invention; FIG. 18 is a concept of the embodiment of the fifth invention in the coefficient domain. FIG. 19 is a block diagram showing the overall configuration of an image data transmission system using orthonormal transform, and FIG. 20 is a diagram showing a transform coefficient area after orthogonal transform of image data. 12... Blocking section, 13... Encoding section, 14... Transmission path, 15... Decoding section, 16... Deblocking section, 19... Size determination section, 20. ...Discrete cosine transform (DCT) section, 21.28... Frame memory, 30... Discard compensation section, 31... Discard discrete cosine transform (DCT) section.

Claims (1)

[Claims] 1) In an image data transmission system in which input image data is encoded as a block of a plurality of pixels and transmitted in cell units, a direct current component of the block currently being transmitted is transmitted to the transmitting side of the image data transmission system. and the DC component of a block corresponding to the block in a frame temporally one frame earlier than the current frame including the block, the size relationship information indicating whether the difference is larger or smaller than a predetermined threshold value is stored in the image data. A size-related information transmitting means (10) for transmitting to the receiving side of the image data transmission system; ), among the blocks corresponding to the discarded cell, if the difference in DC components is small, the DC component of the block corresponding to the block one frame before the frame containing the block is determined. , a discard compensation means (11) for compensating for the DC component of the block corresponding to the discarded cell using the DC component of the block adjacent to the block within the frame including the block with a large difference in the DC component. A cell discard compensation image decoding method characterized by: 2) In an image data transmission system that encodes input image data using a plurality of pixels as a block and transmits it in cell units, the transmitting side of the image data transmission system has a DC component of the block currently being transmitted and a current state including the block. Size relationship information indicating whether the difference between the block and the DC component of the block corresponding to the block in the frame temporally preceding the frame is larger or smaller than a predetermined threshold; size-related information transmitting means (10) for transmitting the DC component value of the block currently being transmitted with respect to the large block to the receiving side of the image data transmission system; is detected, and based on the size relationship information sent from the size relationship information transmitting means (10) when a cell is discarded, among the blocks corresponding to the discarded cell, the block with a small difference in the DC component is detected. The DC component of the block corresponding to the block one frame before the frame including the DC component is sent to the discarded cell using the DC component sent from the size relation information transmitting means (10) for the block with a large difference in the DC component. A cell discard compensation image decoding system characterized by comprising a discard compensating means (11) for compensating a DC component of a corresponding block. 3) In an image data transmission system that encodes input image data using a plurality of pixels as a block and transmits it cell by cell, the transmitting side of the image data transmission system has a DC component of the block currently being transmitted and a current state including the block. Size relationship information indicating whether the difference between the block and the DC component of the block corresponding to the block in the frame temporally preceding the frame is larger or smaller than a predetermined threshold; size-related information transmitting means (10) for transmitting the DC component difference for a large block to the receiving side of the image data transmission system; According to the size relationship information sent from the size relationship information transmitting means (10) when the cell is discarded, among the blocks corresponding to the discarded cell, a block with a small difference in the DC component is selected from the frame containing the block. The DC component of the block corresponding to the block preceding the frame, and the DC component of the block corresponding to the block one frame before the frame containing the block for a block with a large difference in the DC component, and the magnitude relation information transmitting means. A cell discard compensation image decoding system, comprising a discard compensating means (11) for compensating for a DC component of a block corresponding to a discarded cell using the difference between the DC components transmitted from (10). 4) In an image data transmission system that encodes input image data using a plurality of pixels as a block and transmits it in units of cells, the DC component of the block currently being transmitted and the frame containing the block are sent to the transmitting side of the image data transmission system. size relationship information indicating whether the difference between the DC component of the block adjacent to the block is larger or smaller than a predetermined threshold, and the DC current of the block currently being transmitted for the block whose DC component difference is larger than the threshold. a size-related information transmitting means (10) for transmitting the component values to the receiving side of the image data transmission system; Among the blocks corresponding to the discarded cell according to the size relationship information sent from the size relationship information transmitting means (10), a block with a small difference in the DC component is adjacent to the block in the frame including the block. Discard compensation that compensates for the DC component of the block corresponding to the discarded cell using the value of the DC component sent from the size relationship information transmitting means (10) for blocks with a large difference in the DC component. A cell discard compensation image decoding system comprising means (11). 5) In an image data transmission system that encodes input image data using a plurality of pixels as a block and transmits it cell by cell, the DC component of the block currently being transmitted and the frame containing the block are sent to the transmitting side of the image data transmission system. size relationship information indicating whether the difference between the block and the DC component of the block adjacent to the block is larger or smaller than a predetermined threshold; size-related information transmitting means (10) for transmitting the image data to the receiving side of the image data transmission system;
On the receiving side of the image data transmission system, the presence or absence of cell discard is detected, and when a cell is discarded, the block corresponding to the discarded cell is detected based on the size relationship information sent from the size relationship information transmitting means (10). Then, for a block with a small difference in the DC component, the DC component of the block adjacent to the block in the frame including the block is calculated, and for a block with a large difference in the DC component, the DC component of the block adjacent to the block in the same frame is calculated. and a difference between the DC component transmitted from the magnitude relationship information transmitting means (10) to compensate for the DC component of the block corresponding to the discarded cell. Cell discard compensation image decoding method.