JP3446240B2 - Encoding method and encoding device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to orthogonal transform coding and coding.
And coding method for performing coding including variable-length coding
It relates to an encoding device .

[0002]

2. Description of the Related Art When compressing the data transmission amount of digital image data, two-dimensional cosine transform (Discrete Cosin conversion)
An encoding method using an orthogonal transform such as e Transform (hereinafter referred to as DCT) has been conventionally proposed.

In the DCT-based coding method, a television signal of one frame is divided into a plurality of small blocks each consisting of n pixels in the horizontal direction and m pixels in the vertical direction, and DCT is applied to each block. The obtained DC component coefficient data and a plurality of AC component coefficient data are converted into entropy codes having different bit lengths, for example, Huffman codes according to the appearance probability of each AC component, and then transmitted.

[0004]

By the way, it is conceivable to record the image data compressed as described above in a digital VTR. However, in this digital VTR, a reproduced image as good as possible can be obtained in a high speed search operation. Preferably.

At the time of this high-speed search, the rotary head scans over a plurality of recording tracks, so that the reproduced data can be obtained only intermittently. However, an entropy code such as a Huffman code has a variable bit length, and the generated Huffman code is continuously recorded. Therefore, it is difficult to decode the Huffman code at the time of high-speed search in which the reproduced data can be obtained only intermittently.

Further, since the Huffman code having a variable bit length is continuously recorded, when a bit error occurs in one code, the influence is propagated to the subsequent codes and the code identification is performed. Depending on the code that has become impossible and the error has occurred, the error may not be propagated to the relevant block but to other blocks.

As a means for solving the above problems, a method of recording a direct current component at a fixed position with a fixed length has already become common, but it has been proposed as a method of handling an alternating current component. There is a method of recording fixed-length fixed-length AC coefficient data of a high degree of importance in a predetermined number.

However, this method has a drawback that the coding efficiency is deteriorated because the data is fixed in length, and
In reality, it is impossible to record such a large amount of AC components, so that there is a drawback in that a very good image cannot be obtained during a high-speed search or when an error occurs.

Further, JP flat 4 according to the proposal of the present applicant
Japanese Patent No. 322591 discloses a variable length recording of a predetermined number of coefficient data of AC components of high importance at a predetermined position.

This method has the advantage that the amount of generated information does not increase at all, but like the method described above, many AC components cannot be recorded, so that it is very good both during high-speed search and when an error occurs. There was a problem that a good image could not be obtained.

Furthermore, it is possible to arrange and record the coefficient data of the AC component in order from the most important one at a predetermined position by a predetermined amount. With this method, the amount of generated information does not increase at all, and many AC components can be recorded. Therefore, a very good image can be obtained during high-speed search and when an error occurs.

However, when an error occurs in the AC component data having a high degree of importance, it is impossible to cut out all the AC component data thereafter, so that it is extremely difficult to conceal the block, and a good restored image is obtained. Is difficult to obtain.

An object of the present invention is to provide an encoding method and an encoding device that solve the above problems.

[0014]

SUMMARY OF THE INVENTION The present invention provides a plurality of pixels.
The coefficient obtained by orthogonally transforming the block consisting of
In a coding method for variable-length coding and transmitting numerical data , coefficient data of a DC component of coefficient data is a transmission unit.
Place from the beginning of the defined first area in the block
The first step and the remaining area of the transmission unit block intersect.
It is set as the second area in which the coefficient data of the flow component is arranged.
The maximum number of codes generated by variable-length coding
Of the coefficient data, it is divided by the third area that is longer than
Coefficients of relatively high importance among coefficient data of AC components of
Each of the data from the beginning of each of the third areas
In the second step of arranging and in the coefficient data of the AC component
Of the remaining coefficient data of the second unit in the transmission unit block
The encoding method is characterized by comprising a third step of arranging in an empty area inside .

[0015]

The DC component of the coefficient data and the AC component of the AC component are arranged in a predetermined position in order from the most important one by a predetermined amount (not in number) for each predetermined transmission unit. The coefficient data of some AC components that have been transmitted (for example, recorded) and have high importance are arranged at independent positions. As a result, these coefficient data can be extracted and decoded at the high speed search of the digital VTR, for example. Further, since these coefficient data are arranged at the determined positions, they can be decoded without being affected by the error. These coefficient data are important components in the image data in block units, and if these can be decoded, a good restored image can be obtained, and the image quality of the restored image at the time of high-speed search or when an error occurs improves.

Particularly, the coefficient data of some AC components having a high degree of importance are recorded in independent areas, so that the data can be independently cut out regardless of the error of the previous data. Therefore, more data can be cut out than in the past, so that concealment is facilitated and the quality of the restored image is improved.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the high efficiency coded data transmission method according to the present invention is applied to a digital VTR recording system will be described below with reference to the drawings.

FIG. 1 shows an embodiment of a digital VTR recording system. That is, the raster scanning format image signal input through the input terminal 11 is supplied to the A / D converter 12 and one pixel sample is converted into, for example, an 8-bit digital image signal. This digital image signal is supplied to the blocking circuit 13.

The blocking circuit 13 has a memory having a capacity capable of recording a digital image signal for one frame, and is, for example, an area of (8 pixels in the horizontal direction) × (8 pixels (line) in the vertical direction). As one image block,
A digital image signal of one frame (one screen) is divided into a plurality of image blocks. In this case, one image block contains 64 pixel samples.

Further, in the blocking circuit 13, the blanking period in which the data in the input image signal does not exist is removed, and the effective data is made continuous.
A data lack period is formed in the data series.

The data of each image block from the blocking circuit 13 is supplied to the shuffling circuit 14. In the shuffling circuit 14, the image block unit is 1
All the image blocks in the frame are rearranged according to a predetermined rule in one frame. This shuffling processing is executed by address control of the memory.

The output signal of the shuffling circuit 14 is supplied to the DCT conversion circuit 21 of the encoding section 20. This D
In the CT conversion circuit 21, a DCT conversion process is performed for each image block, and from this DCT conversion circuit 21, a plurality of, for example, 8 × 8 coefficient data corresponding to the block size is obtained. As shown in FIG. 2A, the coefficient data includes DC component coefficient data DC and a plurality of AC component coefficient data ACi (i = 1 to 63).

The coefficient data from the DCT conversion circuit 21 is supplied to the block scanning circuit 22. From the block scanning circuit 22, the coefficient data for each block is changed from the frequency of the direct current component DC to the frequency of the alternating current component as shown in FIG. 2B. It is output in a state of performing zigzag scanning toward a relatively high direction. Figure 2
In B, the numerical values described as 0, 1, 2, ... Show the output order. Generally, in the DCT coefficient, the low-frequency component is visually more important than the high-frequency component, and in the block scanning circuit 22, the AC component coefficient is rearranged in the descending order of importance.

The coefficient data from the block scanning circuit 22 is supplied to the requantization circuit 23. In the requantization circuit 23, the coefficient data is stored in the buffer control circuit 27.
Is quantized with the quantization step width from.

The output signal of the requantization circuit 23 is supplied to the variable length coding circuit 25. This variable length coding circuit 25
Encodes the AC component of the coefficient data with a variable length code such as Huffman code. The output data of the variable length coding circuit 25 is supplied to the buffer memory 26. The buffer memory 26 is provided to perform rate conversion so that the coefficient data does not exceed a predetermined transmission rate, that is, in the case of the digital VTR of this example, the transmission rate of the tape recording / reproducing conversion system. That is, the data rate on the input side of the buffer memory 26 is variable, but the data rate on the output side is substantially constant.

In the buffer memory 26,
A change in the amount of transmitted data is detected, and the detected output is supplied to the buffer control circuit 27. The buffer control circuit 27 controls the quantization step width of the requantization circuit 23 and controls the transmitted data output from the variable length coding circuit 25 to have a predetermined data amount. For example, when the amount of transmitted data is too large, the quantization step width is increased and the amount of transmitted data is reduced. Not only the feedback control as in this example, but also the feedforward type buffering process for controlling the quantization step width of the requantization circuit 23 to control the amount of generated data in a predetermined period may be performed. good.

The output signal from the buffer memory 26 is supplied to the framing circuit 28, a sync block is formed for each predetermined data amount, and the coefficient data is rearranged into a data array having a frame structure in which the sync blocks are continuous. . The present invention is characterized by the processing performed by the framing circuit 28.

FIG. 3 shows an example of the structure of the sync block SB. The sync block SB is composed of a plurality of transmission unit blocks BL. In this example, the transmission unit block BL is composed of 17 bytes (= 136 bits) as shown in FIG. 4, and has a threshold value TH representing a quantization step width.
Then, the coefficient data DC of the DC component and the coefficient data AC of the AC component are arranged in the determined position of the transmission unit block BL, that is, in the first area.

In the example of FIG. 4, the threshold TH is 6 bits,
10 bits are allocated to the coefficient data DC of the DC component. Also, 15 bytes (= 120 bytes) are allocated as the second area for the AC component coefficient data AC. Furthermore, the second area for the AC component is divided into a third area of equal length to the longest of the Huffman codes used. In the example of FIG. 4, the maximum length of the Huffman code used is 24 bits (= 3 bytes), and the second area for the AC component is divided into 5 by 3 bytes, and the third area for the AC components AC1 to AC5. It is said that.

The data transmission method that is a feature of the present invention, that is, the processing performed by the framing circuit 28 will be described below with reference to FIGS. The following example is 1
This is an example in which two transmission unit blocks are included in a sync block and buffering is performed in units of one sync block.

The threshold value and the DC component data, which are fixed length data, are arranged in a defined first area of each transmission unit block BL, for example, in the leading 2-byte section. The other area of the transmission unit block BL is the second area for the AC component coefficient data.

The AC component data of relatively high importance (AC1 to AC5 in this case) are arranged in each of the third areas prepared in advance. Since each area is prepared for the longest Huffman code to be used as described above, each coefficient data of AC1 to AC5 can be arranged in the prepared area without fail. Shows the this state is shown in FIG 5.

With respect to the AC component data of low importance (AC6 to AC6 in this case), after the AC component data of high importance is recorded, it is sequentially inserted into the vacant area in the transmission unit block BL and inserted. Is placed. This state is shown in FIG.

Since buffering is not performed for each transmission unit block, there may occur a transmission unit block in which all AC component data cannot be arranged. Figure 6
In the column of, the capacity in the transmission unit block 2 is insufficient,
Part of AC24 and EOB are not placed.

By the above procedure, the data which cannot be arranged is sequentially inserted and arranged in the remaining area in the buffering unit (equal to the sync block in this example). This state is shown in FIG. Earlier,
A part of the AC 24 of the transmission unit block 2 and the EOB, which could not be arranged, are arranged in the empty area of the transmission unit block 1. A flowchart of the data arrangement process is shown in FIG.

As shown in FIG. 8, in the first step,
The threshold value and the coefficient data of the DC component are arranged in the defined first area in the transmission unit block. In the next second step, the data of relatively high importance in the coefficient data of the alternating current, AC1 to AC5 in the above example, are set in the defined third region of the second region for the alternating current component. Will be placed. In the third step, the remaining AC component coefficient data is packed into the empty area in the transmission unit block.

Between the second step and the third step,
There is a step of determining whether or not the arrangement of all data has been completed. At the time when the third step is completed, if there is data that cannot be packed in the transmission unit block, coefficient data of the unallocated AC component is packed in the empty area generated in the entire sync block. In this way, the framing process is completed.

Returning to FIG. 1, the output signal of the framing circuit 28 is supplied to the parity generating circuit 15 and, for example, the error correcting code having the product code structure is encoded, and the parity data is generated and added. To be done. The compressed image data to which the parity data is added is supplied to the digital modulation circuit 16 and digitally modulated. Then, the output signal of the digital modulation circuit 16 is supplied to the parallel-to-serial conversion circuit 17 to be a serial data recording signal. The serial recording signals from the parallel-to-serial conversion circuit 17 are magnetically recorded on the tape by the rotary head as, for example, four diagonal tracks per frame of data.

Although not shown, the parity generation circuit 1
5 and the digital modulation circuit 16 between the block identification signal ID (for example, 2 bytes) and the block synchronization signal SYN.
C (for example, 2 bytes) is added. With this block identification signal ID, a threshold value TH arranged in a predetermined area of the sync block SB, coefficient data DC of a DC component,
The position of the coefficient data AC of the AC component is known.

As described above, the DC component and the AC component of the coefficient data are recorded in the areas determined in the order of importance, and the AC components of particularly high importance are recorded in the independent areas. Therefore, the coefficient data of these DC components and important AC components can be extracted even if the reproduced data can be obtained only intermittently during high-speed search, and it is possible to obtain good results by decoding these coefficient data. A restored image can be obtained.

Further, when an error that cannot be corrected occurs during normal reproduction, a certain bit error has conventionally been propagated to other blocks and the reproduced image may be extremely deteriorated. For example, when an error occurs in the block end code EOB. On the other hand, in the present invention, an error that occurs in a certain block does not participate in cutting out the coefficient data of the DC component and the important AC component of the other block. Therefore, it is possible to obtain a reproduced image which is more resistant to an error than the conventional one and whose image quality is improved.

Further, in the conventional variable length recording method, when an error occurs in the AC component coefficient of high importance,
The data after that could not be cut out. For example,
If an error occurs in AC1, AC2, AC3, AC
No data could be cut out for 4, AC5, ...
On the other hand, in the present invention, even if there is an error in AC1, the coefficient recorded in the independent area can be cut out without being influenced by the error propagation (AC2 to AC5 are Can be cut out). Therefore, concealment is much easier than in the conventional method, and as a result, a good reproduced image can be obtained.

The point of the present invention is to record the coefficient data of the AC component in a predetermined area as much as possible in terms of capacity.
Furthermore, the coefficient data of some of the most important AC components are recorded in independent areas.

The previously proposed method of recording fixed-length fixed-length AC coefficient data of a high degree of importance at a predetermined position, and the important method previously proposed by the present inventors The method of recording variable-length alternating-current component coefficient data at a predetermined position in a predetermined number is limited in the number of alternating-current component coefficient data to be recorded. It is not possible to obtain a very good restored image only by decoding with the data of only these fixed recording areas.

Further, the method recently proposed by the inventors of the present invention, in which the coefficient data of the AC component is recorded in a predetermined area as much as possible in capacity, solves many of the above problems. , If an error occurs in the AC component data of high importance, subsequent AC component data of the same block becomes impossible to cut out, and it is difficult to obtain a good restored image for that block, That problem remained.

On the other hand, according to the method of the present invention, the AC component coefficient data AC records the AC component coefficient data of the corresponding block. When the amount of the AC component coefficient data of a block is smaller than the area of the AC component coefficient data AC, all the data of the block is recorded in the fixed area portion, so that the original restored image is displayed at both the high speed search and the error. You can get the image as it is. Also,
When the amount of the AC component coefficient data of a block is larger than the area of the AC component coefficient data AC, a part of the data of the block will not be recorded in the fixed recording area. Since more AC component coefficient data is recorded in the fixed recording area than in the proposed method, it is possible to obtain a better restored image than in the conventional method.

Further, in the method of recording a fixed number of coefficient data of the above-mentioned AC component having a high degree of importance at a predetermined position, in general, this format increases the amount of information. The method of the present invention does not increase the amount of information at all.

Further, since the coefficient data having the highest importance is recorded in an independent area, the data can be extracted without being affected by the propagation of an error that has occurred in other data. As for the block in which the coefficient data having the highest importance has an error, a restored image better than the conventional one can be obtained.

Although the description has been given for the DCT, the present invention is not limited to the DCT and is applicable to general orthogonal transform coding.

Further, for the sake of simplicity, the explanation has been given only in the case of using the one-dimensional Huffman, but the same applies to the case of using the two-dimensional Huffman.

The present invention is applicable not only to digital VTRs but also to various transmission lines.

[0052]

As described above, according to the present invention, since important information is arranged and transmitted at a predetermined position, for example, when the present invention is applied to a digital VTR, a high speed search is performed. Even if the reproduced data is obtained only intermittently, these important data can be extracted, and a good reproduced image can be obtained by decoding these coefficient data.

When an error that cannot be corrected occurs during normal reproduction, the error generated in one block may have a fatal effect on other blocks.
In the present invention, since a large amount of information is recorded in a predetermined area, important coefficient data can be extracted even in such a case, and a reproduced image with improved image quality can be obtained even if there is an error. Can be obtained, and the error resistance is improved.

[0054] Further, with respect to the importance degree of a high AC component coefficient data, because they are recorded in a separate area of the predetermined position of the second region, the errors that occurred to other data propagation effect can data extraction carried out without receiving, therefore, also with respect to block an error to a higher coefficient data of importance degree is generated, a good restored image can be obtained than the conventional.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration example of a digital VTR for implementing a transmission method according to the present invention.

FIG. 2 is a schematic diagram for explaining coefficient data of DCT conversion output.

FIG. 3 is a schematic diagram illustrating a configuration example of a sync block to be recorded.

FIG. 4 is a schematic diagram showing a part of a data array of a main part of transmission data.

FIG. 5 is a diagram for explaining an example of a framing operation of the present invention.

FIG. 6 is a diagram for explaining an example of a framing operation of the present invention.

FIG. 7 is a diagram for explaining an example of a framing operation of the present invention.

FIG. 8 is a flow chart for explaining the framing operation of the present invention.

[Explanation of symbols]

21 DCT conversion circuit 23 Requantization circuit 25 Variable Length Coding Circuit 28 framing circuit

Continuation of the front page (56) Reference JP-A-4-322591 (JP, A) JP-A-6-209462 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) G11B 20 / 12 H03M 7/30

Claims (4)

(57) [Claims] 1. A block composed of a plurality of pixels as a unit
Variable-length coding the coefficient data obtained by
In the encoding method for transmission , the coefficient data of the DC component in the above coefficient data is transmitted by the transmission unit.
Place from the beginning of the defined first area in the block
In the first step and the remaining area of the transmission unit block, the coefficient decoupling of the AC component is performed.
The second area where the data is placed, and the entire second area
The area is greater than or equal to the maximum code length generated by the variable length coding
Is divided by the third region of, and is compared within the coefficient data of the AC component of the above coefficient data.
Each of the coefficient data with high
The second step of arranging from the top of each area and the remaining coefficient data of the above-mentioned AC component coefficient data
In the empty area in the second area in the transmission unit block
And a third step of arranging . 2. The data amount of encoded data of a plurality of the transmission unit blocks according to claim 1.
Is controlled to be substantially constant, and in the above third step,
Even if the remaining coefficient data is placed in the empty area, it remains
In some cases, the surplus coefficient data is blocked by the other above-mentioned transmission unit blocks.
A coding method characterized by arranging it in a free area . 3. A block composed of a plurality of pixels as a unit
Variable-length coding the coefficient data obtained by
In the transmitting encoding device, the coefficient data of the DC component of the above coefficient data is transmitted as a transmission unit.
The block is arranged from the beginning of the first area defined in the block, and the remaining area of the transmission unit block is the coefficient data of the AC component.
The second area where the data is placed, and the entire second area
The area is greater than or equal to the maximum code length generated by the variable length coding
Is divided by the third region of, and is compared within the coefficient data of the AC component of the above coefficient data.
Each of the coefficient data with high
Place from the beginning of each area, and add the remaining coefficient data of the coefficient data of the AC component above.
In the empty area in the second area in the transmission unit block
An encoding device characterized by being arranged . 4. The data amount of encoded data of a plurality of the transmission unit blocks according to claim 3.
Is controlled to be almost constant, and the remaining coefficient data is
If the remaining coefficient data is left in the
An encoding device, wherein the encoding device is arranged in an empty area of another transmission unit block .