JP3270861B2 - Decoding device for block transform code - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a block conversion code decoding apparatus for compressing a data amount by dividing a digital image signal into small blocks and processing the divided blocks.
In particular, the present invention relates to a decoding device that, when an important word that is an error is modified, makes a determination indicating the reliability of the modification.

[0002]

2. Description of the Related Art When a digital video signal is recorded on a recording medium such as a magnetic tape, the amount of information is large. Therefore, in order to achieve a transmission rate that can be recorded / reproduced, the digital video signal is encoded by a high efficiency encoding. Is usually compressed. ADRC and DCT (Discrete Cosine Trunking), which divide a digital video signal into a number of small blocks and perform encoding processing for each block, are performed as high efficiency coding.
ansform) are known. ADRC can be used, for example, as described in Japanese Patent Application Laid-Open No.
This is a high-efficiency coding that obtains a dynamic range defined by the maximum value and the minimum value of a plurality of pixels included in a dimensional block, and performs encoding adapted to the dynamic range.

The coded outputs obtained by block transform coding do not have equal importance. In the ADRC, if the dynamic range information is not known on the reproduction side, decoding of all the pixels in the block becomes impossible. Therefore, the dynamic range information detected for each block is more important than the code signal for each pixel. high. In the case of DCT, D
In the coefficient data generated by CT, the DC component has a higher importance than the AC component. Also, in the case of DCT, not only the DC component but also information of the quantization step for each block is important. These encoded outputs with high importance are referred to as important words.

[0004]

In a digital VTR using ADRC, even if an important word is an error, all encoded outputs are decoded by using the value of the important word, or a block in which the important word is erroneous is replaced with an error block. In order to correct the error block with the surrounding decoded data. Regardless of the process, the block whose key word is an error is
There is a problem that block-shaped distortion is caused and deterioration of a restored image is conspicuous. Therefore, the error of the important word is estimated by a statistical method based on the spatial correlation between the peripheral block and the block of interest. However, the accuracy of estimation may be low depending on the design. In the case of DCT as well, important words are modified using spatial correlation, but there is a similar problem.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a decoding method of a block transform code capable of suppressing the deterioration of a restored image by making a decision indicating whether or not the estimation has been successful when correcting an erroneous important word. It is to provide a device.

[0006]

According to the first aspect of the present invention, there is provided a computer system comprising:
The digital image signal consisting of a pixel, for each block comprising a plurality of pixels spatially close, produced by block coding to compress the amount of transmitted information, for decoding
In the decoding apparatus of lube b <br/> click transform coding to decode the transmission data importance including high critical words, if important word error, a circuit for retouch important word errors The transmission data is supplied, and a decoding circuit for decoding the encoded data using the correct important word or the modified important word for each block, and a decoding circuit are combined, and the important word is related to the error target block. Successful modification of the key word by referring to the difference between the decoded value at a position in the vicinity of the block of interest and the decoded value in the block around the block of interest and near the boundary with the block of interest. And a circuit for determining whether or not the block transform code is used.

According to a second aspect of the present invention, when referring to a difference, if the sum of absolute values of the difference is smaller than a predetermined value, it is determined that the modification has been successful. According to a third aspect of the present invention, when referring to the difference, when the maximum value of the absolute values of the differences is smaller than a predetermined value, it is determined that the modification has been successful.
According to a fourth aspect of the present invention, when referring to the difference, when the sum of the absolute values of the differences is smaller than a predetermined value or when the maximum value of the absolute values of the differences is smaller than the predetermined value, it is determined that the modification has been successful. Things. According to a fifth aspect of the present invention, when referring to the difference, when the sum of the absolute values of the differences is smaller than a predetermined value and the maximum value of the absolute values of the differences is smaller than the predetermined value, it is determined that the modification is successful. Things. According to a fifth aspect of the present invention, when an important word is an error, the error of the important word is corrected, the number of errors in data of peripheral blocks used for the correction is counted, and the number of errors is compared with a predetermined value. ,
A flag indicating whether or not the estimation of the important word is successful is generated, and this flag is reset with reference to the difference.

[0008]

When the block of interest has an error, an important word relating to the block of interest is estimated by a statistical method using the coded value of the block of interest and its surrounding data (decoded value). Decoding is performed using the correct key word or the modified key word. The difference between the decoded value of the block of interest in which the key word is erroneous and the decoded value in a block around the block of interest and located near the boundary with the block of interest is checked. If it is larger, a determination circuit is provided for determining that the estimation has failed. With this determination, the reliability of the modification can be increased.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the decoding device according to the present invention will be described below. In this embodiment, ADRC is used as a block conversion code. As a specific example of the ADRC, an effective area of one frame is divided into blocks each having a size of (4 × 4) pixels. The ADRC encoder provided on the recording side detects the dynamic range DR and the minimum value MIN of each block, and requantizes the video data from which the minimum value has been removed in a quantization step. In the case of 4-bit fixed length ADRC, the dynamic range DR is
By setting it to 1/16, a quantization step Δ is obtained. In the quantization step Δ, the video data from which the minimum value has been removed is divided, and a value obtained by rounding down the quotient to an integer is used as a code signal. Since the dynamic range DR, the minimum value MIN, and the code signal have been converted into the structure of the recording data, they are recorded on the magnetic tape.

In FIG. 1, reference numeral 1 denotes an important word modifying circuit. Although not shown, the data is reproduced from the magnetic tape, the channel code is decoded, and TBC (time axis correction),
The reproduced data subjected to frame decomposition and error correction and an error flag indicating the presence or absence of an error are supplied to the important word modifying circuit 1. The error flag is set to "1" for a sample for which error correction has failed. There are three types of error flags: an error flag Ed for the dynamic range DR, an error flag Em for the minimum value MIN, and an error flag Ep for the code signal of the pixel.

The important word modifying circuit 1 is a key word modifying circuit 1
When the dynamic range DR or the minimum value MIN) is an error, an important word is corrected by the least squares method using correct decoded pixel data of a peripheral block, and ADRC decoding is performed thereafter. For Keyword Modification Circuit 1,
As will be described later, here, it is determined whether or not the estimation of the important word is successful. As a result of this determination, a flag S relating to the important word is generated. If S = “1”, it means that the estimation of the important word has succeeded, and if S = “0”, it means that the estimation of the important word has failed.

The decoded data from the key word correction circuit 1 is supplied to a spatial interpolation circuit 2, and flag information Ed,
Em, Ep, and S are supplied to the ROM 3. The error flag Es for each pixel formed in the ROM 3 is supplied to the spatial interpolation circuit 2. The ROM 3 generates an error flag Es for each pixel from the flag information added as an address according to Table 1 below.

[0013]

[Table 1]

In Table 1, the error flag Es is
It is "1" in the case of an error, "0" in the case of no error, and * means that the value of the error flag is ignored. Some patterns in Table 1 will be described. (Ed, Em, Ep, S) = (00
1 *), Es = "1". (Ed, Em,
The pattern of (Ep, S) = (0100) has the minimum value MIN
Es = “1” because it means that the retouching has failed. On the other hand, (Ed, Em, Ep, S) = (0101)
Indicates that the modification of the minimum value MIN has been successful, so that Es = "0". (Ed, Em, Ep,
The pattern of (S) = (1110) means that the modification of the dynamic range DR and the minimum value MIN was successful, so Es = "0".

The spatial interpolation circuit 2 receives the above-mentioned error flag Es, and when the pixel of interest to be interpolated has an error with reference to the error flag Es, interpolates the error pixel with neighboring pixels. Specifically, when the error flag Es for the target pixel is “1”, the Es of the pixels at eight surrounding points (upper, lower, left and right, and four oblique points) are looked at, and the horizontal interpolation is performed first. In the vertical direction, then in the diagonal direction, and finally in the priority order of interpolation that simply replaces the next pixel. When the interpolation is performed, the error flag Es is reset.
= Remains "1". In the case of a block for which key word estimation has failed, the interpolation of the pixels at the periphery of the block can be performed using the correct pixels of the surrounding blocks, but the pixels at the center remain without being interpolated.

The temporal interpolation circuit 4 is connected to the spatial interpolation circuit 2. As described above, the pixel data that could not be interpolated by the spatial interpolation circuit 2
Is interpolated by The last-stage interpolation circuit 4 replaces the erroneous pixel with the pixel data of the previous frame at the same spatial position as the erroneous pixel. Specifically, the interpolation circuit 4 has a frame memory, and Es =
The pixel data of “0” is written to the frame memory, and the pixel data of Es = “1” is not written, and instead of this pixel data, the previously written correct pixel data is read.

As described above, the modification of the important word, the adaptive spatial interpolation, and the interpolation in the time direction are performed according to the error situation. When this embodiment is applied to a digital VTR, a sufficient error correction can be performed not only during a normal reproduction operation but also during a variable speed reproduction operation, so that deterioration in image quality due to error propagation can be prevented.

FIG. 2 shows an example of the important word modifying circuit 1.
In FIG. 2, the error flag indicates the dynamic range D
R, the minimum value MIN, and Ed, Em, and Ep for each of the pixel code signals are collectively referred to. Data and error flags are supplied to the memory 11, the ADRC decoder 12, and the memory 13. The memories 11 and 13 are provided for time adjustment.

The ADRC decoder 12 generates a decoded value Li for each pixel when the number of bits of the code signal is 4 bits. This decoded value Li is represented by the following equation. Li = [(DR / 2 ⁴ ) × xi + MIN] = [Δ × xi + MIN]

Here, xi is a code signal value, Δ is a quantization step, and [] is a Gaussian symbol. The ADRC decoder 12 has a configuration in which the operation in [] of the above equation is realized by, for example, a ROM and the addition of the minimum value MIN is performed. ADR
The decoded data from the C decoder 12 is supplied to the modification circuit 14. The correction circuit 14 is supplied with the error flag, the dynamic range DR, and the minimum value MIN from the memory 11.

When the important word of the block of interest is an error, the modifying circuit 14 estimates the important word that is in error using the encoded value of the block of interest and the boundary decoded data of the peripheral block. The word is
Simply pass through. For example, the modification is performed by the least square method. This estimation is performed when both the dynamic range DR and the minimum value MIN are errors, when an error occurs only in the minimum value MIN, and when an error occurs only in the dynamic range DR.

Since this estimation is a statistical method, a relatively large error occurs depending on the picture. For example, a sharp level peak in the block of interest is present, the estimation error is large Kikunaru. In order to increase the accuracy of estimation, in addition to excluding error samples from estimation targets, a process of checking the number of errors in data to be used and forming a flag S according to the number of errors is performed.

That is, the number of errors in the data used for estimation is counted by the counter 15, and the counted value is
At 6 the threshold value TH0 is compared. If the number of errors is larger than the threshold value TH0, the flag S is set to "0" in the modification circuit 14, and information indicating that the estimation has failed is sent to the subsequent stage.

The estimation method in the modification circuit 14 is as follows.
Not only estimation by the method of least squares, but also estimation by considering the maximum value and the minimum value of the boundary data of the peripheral block as the maximum value and the minimum value of the block of interest can be performed.

The dynamic range D from the modifying circuit 14
R, the minimum value MIN, and the flag S are the ADRC decoder 1
7 is supplied. The ADRC decoder 17 is also supplied with data and an error flag from the memory 13.
From the ADRC decoder 17, an error flag, decoded data of each pixel, and a flag S are taken out and supplied to a memory 18 for time alignment.

The decoded data of each pixel and the error flag are supplied to a decision circuit 19. The threshold value TH1 is supplied to the determination circuit 19. The output of the decision circuit 19 is supplied to the switch circuit 20 as a control signal.
The flag S is supplied to the switch circuit 20, and the flag S is forcibly set to “0” (in other words, the estimation has failed) according to the state of the switch circuit 20.

The decision circuit 19 decides whether or not the key word has been successfully estimated. That is, the decoded value at the boundary between the block of interest in which the important word is erroneous and the surrounding block is compared. If the sum of the absolute values of the differences is larger than the threshold value TH1, the flag S is forcibly set to "0". I do.

FIG. 3 shows the peripheral decoded values y1 to y1 of the block of interest.
It shows y12 and boundary data (decoded value) of its peripheral blocks. The determination circuit 19 generates the following absolute value sum of the difference. () * Means an absolute value. These decoded values are excluded if they are in error.

Sum of absolute values of difference = (y1 ″ −y1) * +
(Y1'-y1) * + (y2'-y2) * + (y3'-
y3) * + (y4'-y4) * + (y4 "-y4) *
+ (Y5'-y5) * + (y6'-y6) * + (y7 '
-Y7) * + (y7 "-y7) * + (y8'-y8)
* + (Y9'-y9) * + (y10'-y10) * + (y10
"" -Y10) * + (y11'-y11) * + (y12'-y1
2) *

This sum of absolute values is compared with a threshold value TH1. If the sum of absolute values is greater than TH1, the flag S is forcibly set to "0". Since the existence of a very large difference from the surrounding blocks is contrary to the existence of the spatial correlation of the image, it is determined that the keyword estimation has failed. As a result, the accuracy of the modification of the important words is improved.

FIG. 4 shows another embodiment of the present invention. The difference from the important word modification circuit in FIG. 2 is that both the judgment circuit 19 and the judgment circuit 21 are provided. As described above, the determination circuit 19 compares the absolute value sum of the difference between the decoded value of the target block and the boundary data of the peripheral block with the threshold value TH1 to determine whether the estimation of the important word is good or not. The other determination circuit 21 compares the maximum value of the absolute values of the difference between the decoded values of the corresponding two pixels with the threshold value TH2.

[0032] During maximum value is greater than the threshold value TH2 of the absolute value of the difference, as the important word modification (estimated) fails, the switch circuit 20 the flag S of important words,
Forced to "0". If the difference between adjacent decoded values becomes very large due to erroneous estimation of the dynamic range, the pixel becomes conspicuous. The deterioration of the image quality is prevented by the determination circuit 21. Two judgment circuits 19
By means of and 21, the deterioration of the image quality can be prevented more reliably than in the above-described embodiment.

In the above embodiment, when the important word is estimated, the flag S indicating whether or not the estimation is successful is generated in the modifying circuit 14, but this is not always necessary. This flag S may be generated. Although ADRC is used as block coding, other block coding such as DCT may be used.

[0034]

According to the present invention, for a block in which an important word is erroneous, an important word is estimated using decoded values existing around the block, and a flag indicating success / failure of the estimation is formed. Since the interpolation of the error pixel is performed with reference to this flag, a good restored image can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of an example of an important word modifying circuit in one embodiment of the present invention.

FIG. 3 is a schematic diagram illustrating a decoded value of a block of interest and a decoded value of a boundary between peripheral blocks.

FIG. 4 is a block diagram showing a configuration of another example of the important word modifying circuit of the present invention.

[Explanation of Signs] 1 Key word correction circuit 2 Spatial interpolation circuit 4 Time direction interpolation circuit 12, 17 ADRC decoder 14 Correction circuit 19, 21 Judgment circuit

Claims (5)

(57) [Claims] 1. A digital image signal comprising a plurality of pixels.
And for each block comprising a plurality of pixels spatially close, raw by block coding to compress the amount of transmission information
A transmission that includes the important words that are important for decoding
In the decoding apparatus of lube lock conversion code to decode the data, if the upper Symbol key word errors, the important word and means for modification errors, the transmission data is supplied, right above for each of the blocks A decoding unit for decoding the encoded data using the important word or the modified important word; and a decoding unit coupled to the decoding unit, wherein the important word is related to an error block of interest in the vicinity of the block of interest. The decoded value of the position and the peripheral block of the noted block,
Means for determining whether the modification of the important word has been successful by referring to a difference from a decoded value located near the boundary with the block of interest. apparatus. 2. A digital image signal comprising a plurality of pixels.
And for each block comprising a plurality of pixels spatially close, raw by block coding to compress the amount of transmission information
A transmission that includes the important words that are important for decoding
In the decoding apparatus of lube lock conversion code to decode the data, if the upper Symbol key word errors, the important word and means for modification errors, the transmission data is supplied, right above for each of the blocks A decoding unit for decoding the encoded data using the important word or the modified important word; and a decoding unit coupled to the decoding unit, wherein the important word is related to an error block of interest in the vicinity of the block of interest. The decoded value of the position and the peripheral block of the noted block,
Means for determining that the modification of the key word has been successful when the sum of absolute values of the difference from the decoded value located near the boundary with the block of interest is smaller than a predetermined value. Decoding device for transform code. 3. A digital image signal comprising a plurality of pixels.
And for each block comprising a plurality of pixels spatially close, raw by block coding to compress the amount of transmission information
A transmission that includes the important words that are important for decoding
In the decoding apparatus of lube lock conversion code to decode the data, if the upper Symbol key word errors, the important word and means for modification errors, the transmission data is supplied, right above for each of the blocks A decoding unit for decoding the encoded data using the important word or the modified important word; and a decoding unit coupled to the decoding unit, wherein the important word is related to an error block of interest in the vicinity of the block of interest. The decoded value of the position and the peripheral block of the noted block,
Means for determining that the modification of the important word has been successful when the maximum value of the absolute value of the difference between the decoded word and the decoded value located near the boundary with the block of interest is smaller than a predetermined value. For decoding block transform codes. 4. A digital image signal comprising a plurality of pixels.
And for each block comprising a plurality of pixels spatially close, raw by block coding to compress the amount of transmission information
A transmission that includes the important words that are important for decoding
In the decoding apparatus of lube lock conversion code to decode the data, if the upper Symbol key word errors, the important word and means for modification errors, the transmission data is supplied, right above for each of the blocks A decoding unit for decoding the encoded data using the important word or the modified important word; and a decoding unit coupled to the decoding unit, wherein the important word is related to an error block of interest in the vicinity of the block of interest. The decoded value of the position and the peripheral block of the noted block,
Modification of the key word is successful when the sum of absolute values of the difference from the decoded value located near the boundary with the block of interest is smaller than a predetermined value or the maximum value of the absolute value of the difference is smaller than a predetermined value. A decoding device for a block transform code, the decoding device comprising: 5. A digital image signal comprising a plurality of pixels.
And for each block comprising a plurality of pixels spatially close, raw by block coding to compress the amount of transmission information
A transmission that includes the important words that are important for decoding
In the decoding apparatus of lube lock conversion code to decode the data and modification means for generating an error flag indicating the presence or absence of error of the transmission data, if the key word is an error, an error of the important words Means for supplying the transmission data, decoding means for decoding the encoded data using the correct important words or the modified important words for each block, and the decoding means A decoded value of a peripheral position in the block of interest with respect to the block of interest in which the important word is an error,
By referring to the difference from the decoded value located near the boundary with the block of interest, it is determined whether or not the modification of the important word has been successful, and as a result, means for resetting the error flag. An apparatus for decoding a block transform code.