JPH05300485A - Decoding device of block conversion code - Google Patents

Abstract

PURPOSE:To interpolate picture element data to be an error with high accuracy when the encoded data generated by a block encoding is decoded. CONSTITUTION:Significant words are corrected and the decoding of an ADRC is performed using right significant words or the corrected significant words. The picture element data to be an error in this decoded data is interpolated by a spatial interpolation circuit 2. Next, an interpolation replacing the picture element data of the error is performed by the picture element data of the previous frame stored in a frame memory 5 by a time direction interpolation circuit 3. A motion flag M is formed from the difference of an inter-frame picture element data and is stored in a motion flag memory 9. In a memory controller 7, the decision of a motion/still area is performed referring to the motion flag M of a surrounded block, regarding the block where the significant words are errors. In the case of the still area, the interpolation of a time direction is performed with first priority even when the correction of significant words or the spatial interpolation is succeeded.

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 device for compressing a data amount by dividing a digital image signal into small blocks and processing each block.
In particular, it relates to one that interpolates an erroneous pixel.

[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. Is usually compressed. For high-efficiency encoding, a digital video signal is divided into a large number of small blocks, and ADRC, DCT (Discrete Cosine Tr
ansform) etc. are known. ADRC is disclosed in, for example, Japanese Patent Application Laid-Open No. 61-144989.
This is a high-efficiency coding in which a dynamic range defined by the maximum value and the minimum value of a plurality of pixels included in a dimensional block is obtained, and the coding adapted to this dynamic range is performed.

The coded output obtained by block transform coding is not of equal importance. In ADRC, if the dynamic range information is not known on the reproducing side, all the pixels in the block cannot be decoded. 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 is more important than the AC component. Also in the case of DCT, not only the DC component but also the quantization step information for each block is important. These coded outputs having high importance are called important words.

In a digital VTR using ADRC,
Even if the important word is an error, all the encoded outputs are decoded using the value, or the block having the important word in error is corrected as an error block by surrounding decoded data. In any processing, a block in which an important word is an error has a block-like distortion, and there is a problem that the restored image is conspicuously deteriorated. Therefore, the error of this important word is estimated by a statistical method based on the spatial correlation between the peripheral block and the target block.

Regarding an error in pixel (sample) units, a method of spatially interpolating the error and a method of interpolating the error in the time direction have been considered. The spatial interpolation is to interpolate the data of the pixel of interest, which is an error, with correct surrounding data. Interpolation in the time direction replaces the error data with the correct data from the previous frame at the same spatial location. Usually, it is set as a priority order in which important words are modified, spatial interpolation is then performed, and further temporal interpolation is performed.

[0006]

In the error correction on a pixel-by-pixel basis, the spatial error correction is prioritized over the one in the time direction because the spatial correlation can improve the accuracy of interpolation. It is based on what you can do. However, with respect to the blocks in the still area, the interpolation accuracy can be made higher by prioritizing the interpolation in the time direction.

Therefore, an object of the present invention is to provide a block conversion code decoding apparatus capable of further improving the accuracy of interpolation by giving priority to interpolation in the time direction in accordance with the state of motion. ..

[0008]

According to the present invention, a block encoding for compressing the amount of transmission information is performed for each block consisting of a plurality of pixels spatially adjacent to each other, and an important word for decoding is important. A block conversion code decoding device for generating encoded data including block data, adding parity of an error correction code to the encoded data to form transmission data, and decoding each pixel data from the received transmission data At, in the transmission data error correction,
A circuit for generating an error flag indicating the presence or absence of an error, a circuit for correcting the error of the important word when the important word is an error, and the transmission data are supplied to correct the important word or correct for each block. And a decoding circuit for decoding encoded data using the important word, a spatial interpolation circuit coupled to the decoding circuit for interpolating error pixel data by peripheral pixel data, and a spatial interpolation circuit. A time direction interpolation circuit that replaces the error pixel data with the correct pixel data of the same position in the previous frame, and interpolates with the spatial interpolation circuit when the error pixel data is in the still region. The block conversion code decoding device is characterized in that the interpolation by the time direction interpolation circuit is prioritized.

[0009]

When the block of interest has an error, the 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 peripheral data (decoded value). Decoding is performed using the correct key word or the corrected key word. The error of the pixel data after decoding is interpolated by spatial interpolation, and when this is not possible, it is interpolated by temporal interpolation. When the erroneous pixel data is in the still region, even if the correction or spatial interpolation of the important word is successful, the interpolation in the time direction is prioritized, and as a result, the accuracy of the correction can be improved.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the decoding device according to the present invention will be described below. In this embodiment, for example, ADRC is used as the block conversion code. As a specific example of ADRC, the effective area of one frame is divided into blocks each having a size of (4 × 4) pixels. AD provided on the recording side
In the RC encoder, the dynamic range DR of each block and the minimum value MIN are detected, and the video data from which the minimum value has been removed is requantized in the quantization step. In the case of 4-bit fixed length ADRC, the quantization step Δ can be obtained by setting the dynamic range DR to 1/16. At this quantization step Δ, the video data from which the minimum value has been removed is divided, and the value obtained by converting the quotient into an integer is used as the code signal. Dynamic range DR,
Since 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 correction circuit. Although not shown, the data is reproduced from the magnetic tape, the channel code is decoded, and the TBC (time base correction),
The reproduced data (data 1) subjected to frame decomposition and error correction and an error flag EF1 indicating the presence or absence of an error are supplied to the important word correction circuit 1. The error flag is set to "1" for the sample for which the error cannot be corrected. The error flags are an error flag Ed related to the dynamic range DR, an error flag Em related to the minimum value MIN,
There are three types, that is, an error flag Ep relating to the code signal of the pixel.

The important word correction circuit 1 uses the important word (ie,
When the dynamic range DR or the minimum value MIN) is in error, the important word is modified by the least squares method using the correct decoded pixel data of the peripheral block, and then the ADRC decoding is performed. The important word correction circuit 1 determines whether or not the important word is estimated successfully. As a result of this determination, the flag S1 is generated. If S1 = "1", it means that the important word is estimated successfully, and S1 =
If it is “0”, it means that the estimation of the important word has failed.

Error flag EF from the important word correction circuit 1
2 (Ed, Em, Ep, S1) and the decoded data (data 2) are supplied to the spatial interpolation circuit 2. The spatial interpolation circuit 2 forms an error flag Es regarding spatial interpolation in pixel units. That is, the error flag Es is generated according to Table 1 below.

[0014]

[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) is the minimum value MIN.
Es = “1” because it means that the modification of No. has failed. On the other hand, (Ed, Em, Ep, S) = (0101)
The pattern indicates that the modification of the minimum value MIN has succeeded, and thus Es = “0”. (Ed, Em, Ep,
The pattern S) = (1110) means that the dynamic range DR and the minimum value MIN have been successfully modified, and thus Es = “0”.

The spatial interpolation circuit 2 refers to the above-mentioned error flag Es and interpolates the error pixel with the peripheral pixels when the pixel of interest to be interpolated has an error. Specifically, the error flag Es related to the pixel of interest is "1".
, See Es of 8 surrounding pixels (upper and lower, left and right 4 points and diagonal 4 points), and first interpolate horizontally, then vertically, then diagonally, last. Interpolation is performed in the priority order of interpolation that is simply replaced with the adjacent pixel. If the interpolation is successful, the flag S2 is set (S2 = "1"), and if the interpolation cannot be performed, S2 = "0" is set. In the case of a block in which the important word estimation fails, the interpolation of the pixels in the peripheral portion of the block can be performed using the correct pixels in the peripheral blocks, but the central pixel remains uninterpolated.

Error flag EF from the spatial interpolation circuit 2
3 (Ed, Em, Ep, S1, S2) and the decoded data (data 3) are supplied to the time direction interpolation circuit 3. An example of the time direction interpolation circuit 3 is shown in FIG. First, the error flag EF3 is supplied to the ROM 4, and the ROM 4 forms the error flag Et in the pixel unit in the time direction according to Table 2 below. This error flag Et is supplied to the memory controller 7 via the delay circuit 6. Table 2
For simplicity, Ed, Em, Ep, and S1 are represented by an error flag Es.

[0018]

[Table 2]

In FIG. 2, reference numeral 5 is a frame memory, and in the past, spatial interpolation was performed by referring to only the flag Et. That is, the time-direction interpolation circuit 3 replaces the error pixel with the pixel data of the previous frame at the same spatial position as the pixel of the error (Et = “1”). Specifically, the pixel data of Et = “0” is written in the frame memory 5, and Et =
The pixel data of "1" is not written, and the correct pixel data previously written is read instead of this pixel data.

However, when there is an error in the data in the still area, it is preferable to use the data of the previous frame even if the important word can be corrected or the spatial interpolation is successful. In this embodiment, a comparison circuit 8 is provided, a motion flag M is formed by this comparison circuit 8, and this motion flag M is stored in the motion flag memory 9. The error flag EF3 and the data 3 are supplied to the comparison circuit 8.

The comparison circuit 8 reads the data of the previous frame from the frame memory 5 and compares the input data with the data of the previous frame. A difference between pixel data at the same position is obtained between two temporally consecutive frames, and the absolute value Δ of this difference is compared with an appropriate threshold value THt. Δ
If> THt, it is determined to be a pixel in the motion area, and as a result, the motion flag M = “1”, and if Δ ≦ THt,
It is determined that the pixel is in the still area, and the motion flag M is set to "0". If Ed or Em in the error flag EF3 is "1", M = "1" is forcibly set. this is,
This is to exclude the case where the entire block is in error when making a determination described later. The motion flag M is stored in the motion flag memory 9.

The memory controller 7 is supplied with the error flags EF3 and Et and the data 3 via the delay circuit 6, and the motion flag M read from the motion flag memory 9. FIG. 3 shows the memory controller 7
It is a flowchart showing the processing of 1., and whether Ed or Em is "1" is checked in the first determination step. If these flags are "0", the control shifts to the determination step 14 to determine whether Et = "1".

If Et = “1”, time-direction interpolation is necessary, so writing of data to the frame memory 5 is prohibited (step 15). If Et = “0”, this pixel data is written to the frame memory 5 (step 16). These processes are performed regardless of whether it is a still area or a moving area.

In step 11, with respect to a block (block in which Ed or Em is "1") which means that the important word is an error, a process of determining whether the block is a motion block or a still block is performed. That is, the motion flags M of eight blocks around this block are read from the flag memory 9 (that is, 1).
2). Then, it is determined in step 13 whether or not it is a moving area.
Made in. In this determination, the total of the number of motion flags M of eight surrounding blocks is obtained, and when the total is larger than a certain threshold value THm, it is determined as a motion area, and when the total is larger than the threshold value THm, the motion area is stationary. The area is determined.

When it is determined that the area is the still area, the flag Et
Regardless of the above, it is judged that it is better to output the data of the previous frame, and the writing of the data is stopped Step 15
Control is transferred to. If it is determined to be in the motion area, the control shifts to the determination step 14, and writing to the frame memory 5 or prohibition of writing is performed in response to the flag Et.

As described above, in the normal interpolation, the interpolation is performed according to the priority of the correction of the important word, the spatial interpolation, and the interpolation in the time direction. In the still region of the image, the interpolation in the time direction has the highest priority. Will be done.

The estimation method in the modifying circuit 1 is not limited to the estimation by the least squares method, and the estimation in which the maximum value and the minimum value of the peripheral data are the maximum value and the minimum value of the target block can be performed.

Although ADRC is used as the block coding in the above embodiments, other block coding such as DCT may be used.

[0029]

According to the present invention, when the pixel data in error is interpolated, it is determined whether the pixel data is included in the motion area or the still area. In the case of the still area, the spatial interpolation is successful. Whether or not it is done, the interpolation in the time direction has the highest priority. As a result, the accuracy of interpolation of error pixels can be increased.

[Brief description of drawings]

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

FIG. 2 is a block diagram showing a configuration of an example of a time direction interpolation circuit according to an embodiment of the present invention.

FIG. 3 is a flowchart for explaining an operation of an example of a time direction interpolation circuit.

[Explanation of symbols]

 1 key word correction circuit 2 spatial interpolation circuit 3 time direction interpolation circuit 5 frame memory 9 motion flag memory

Claims (1)

[Claims] 1. A block coding for compressing a transmission information amount is performed for each block consisting of a plurality of pixels spatially adjacent to each other, and the coded data including an important word with high importance for decoding is the block code. In the decoding device for the block conversion code for decoding each pixel data from the received transmission data, the transmission data is generated by the above-mentioned transmission data. Means for correcting the error of, and generating an error flag indicating the presence or absence of an error, a means for correcting the error of the important word when the important word is an error, and the transmission data is supplied, A decoding means for decoding the encoded data using the correct important word or the modified important word for each block, and the decoding means. And the spatial interpolation means for interpolating the erroneous pixel data by the surrounding pixel data, and the spatial interpolation means, combined by the above-mentioned spatial interpolation means, by the correct pixel data at the same position in the previous frame. A time direction interpolation means for replacing pixel data, wherein when the error pixel data is in a still region, the interpolation by the time direction interpolation means is prioritized over the interpolation by the spatial interpolation means. Decoding device for block conversion code.