JP3417933B2 - Variable length decoding method and apparatus - Google Patents

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a variable length decoding method and apparatus.

[0002]

2. Description of the Related Art A variable length code is a code having a short code length for frequently appearing symbols, based on the frequency of occurrence of symbols.
A symbol that rarely appears is a code system in which a code having a long code length is assigned so that the code length becomes short on average. Therefore, when the variable length code is used, the amount of data can be significantly compressed as compared with the data before encoding. As a method of constructing such a variable-length code, an optimal Huffman algorithm for a memoryless information source is known.

As a general problem of the variable length code, when an error is mixed in the encoded data due to a transmission path error or other reasons, the data after the error is mixed is decoded by the influence of the propagation. The point is that the device will not be able to decrypt correctly. Therefore, when there is a possibility that an error will occur in the transmission path, it is general to insert a synchronization code periodically at a certain interval to prevent the propagation of the error. A bit pattern that does not appear in a combination of variable length codes is assigned to the synchronization code. According to this method, even if an error occurs in the encoded data and the decoding cannot be performed, the propagation of the error can be prevented and the decoding can be correctly performed by finding the next synchronization code. However, even when the sync code is used in this way, the coded data from the point where an error occurs as shown in FIG. Can't do.

Therefore, the variable length code is changed from the normal configuration shown in FIG. 40 to that shown in FIG. 41 so that the variable length code is not only decodable from the normal forward direction as shown in FIG. A method of forming a code that can be decoded from the direction is known. Further, since such a code can read the encoded data from the reverse direction, it can be used for reverse reproduction in a storage medium such as a disk memory for storing the encoded data. Such a variable-length code that can be decoded not only in the forward direction but also in the reverse direction is referred to as a reversible code here. An example of a reversible code is
For example, it is described in JP-A-5-300027, "Reversible Variable Length Coding System". In the reversible code of this known example, as shown in FIG. 41, each code word has a longer code length at the end of the code word of the Huffman code which is a variable length code that can be decoded from the forward direction shown in FIG. It is a variable length code that can be decoded from the reverse direction by adding a bit under the condition that it does not match the end of another code word.

However, since this reversible code adds bits to the end of the code word of the variable length code that can be decoded only from the forward direction, the number of useless bits increases and the average code length increases. As a result, the coding efficiency is greatly reduced as compared with the variable length code that can be decoded only in the forward direction.

Another problem of the conventional reversible code is that when the number of information symbols is large, the amount of storage required by the variable length coding / decoding device is large, and therefore it is suitable for practical use. There was no point. As an example of a case where the number of information symbols is large, there is coding of DCT coefficients which is often used in video coding of moving images and still images. Normally, in image coding, 8 × 8 DCT (discrete cosine transform) is performed, and 8-bit linear quantization is performed on orthogonal transform coefficients called DCT coefficients obtained by this. Furthermore, the quantized DCT transform coefficient is subjected to zigzag scanning from the low frequency band, and variable length coding is performed by combining zero run and non-zero coefficients.

For example, ITU-T DRAFT Rec
Ommendation H.M. In H.263 (1995), the quantization index value is -127 to +127 (the index value 0 is a run, so the number of indexes is 25).
3), the number of zero runs is 63 at the maximum, and since the coding that distinguishes whether the last DCT coefficient of a block is a non-zero coefficient is adopted, the number of information symbols is 253 ×
It is a very large number of 64 × 2 = 32384.
Therefore, H. In H.263, DCT coefficients with a high appearance probability are coded with a variable length code, but DC with a low appearance probability is used.
For the T coefficient, a fixed-length code with a total of 15-bit code including a 1-bit code for distinguishing whether it is the last non-zero coefficient of a block, a 6-bit code for a zero run, and a 8-bit code for a quantization index. The fixed length code is added with a code called an escape code at the beginning.

Encoding / decoding is possible even if all codewords are stored as a codeword table. However, with the above-described code configuration, the 15-bit fixed-length code part has a variable length. Since encoding / decoding can be performed separately from the code, it is only necessary to prepare a code word table of DCT coefficients and escape codes with high appearance probability. Therefore, it is possible to reduce the storage capacity by reducing the code word table of the variable length code.

However, the conventional reversible code cannot add an escape code because it is necessary to add a bit to the end of a code word of a variable length code that can be decoded only in the forward direction as described above. Therefore, even when the number of information symbols is large, it is necessary to prepare the code words corresponding to all the information symbols, and the storage amount becomes enormous.

[0010]

As described above, the reversible code according to the prior art, that is, the variable length code that can be decoded in the forward direction and the backward direction is a code word of the variable length code that can be decoded only in the forward direction. Since a reversible code is constructed by adding a bit to the end of, the average code length becomes longer and the average code length becomes longer, and the coding efficiency decreases as compared with the variable length code that can be decoded only from the forward direction. However, in a variable length coding / decoding device using such a reversible code, it is necessary to prepare a codeword table having codewords corresponding to all information symbols in some form, If the number of information symbols is large as in image coding, there is a problem that a large amount of memory is required, so it is not suitable for practical use in any case. There is a problem in that.

It is an object of the present invention to provide a variable length decoding method and apparatus which can be decoded in the forward direction and the backward direction and which are highly practical.

More specifically, there is provided a variable length decoding method and apparatus which can be decoded in the forward direction and the backward direction and do not require a large storage amount even when the number of information symbols is large. The purpose is to

[0013]

In order to solve the above-mentioned problems, the present invention provides a specific code (escape code, for example, forward or backward decoding at the beginning and end of a fixed-length code) indicating the beginning and end. Variable to decode variable-length coded data that can be decoded in the forward direction or in the backward direction, including a code added with a specific possible code, a specific variable-length code that can be decoded in the forward direction or the backward direction. In the long decoding method, when decoding variable-length coded data from the forward direction or the backward direction, the first and second decoding value tables that store the decoding values of the specific codes indicating the first and last values are used. When the specific code indicating the end is decoded, the code following the specific code indicating the head is regarded as the fixed length code by referring to the second decoded value table storing the decoded value of the fixed length code. Characterized by decoding

Further, according to the present invention, at the beginning and the end of the fixed length code, a specific code (escape code,
For example, it can be decoded both forward and backward, including a specific code that can be decoded in the forward and backward directions, and a code with a specific variable length code that can be decoded in both the forward and backward directions. In a variable length decoding device for decoding variable length encoded data, a first decoded value table storing the decoded value of a specific code indicating the beginning and the end and a second decoded value table storing the decoded value of the fixed length code. For decoding the variable length coded data from the forward direction, a backward decoding means for decoding the variable length coded data from the reverse direction, and a fixed length code. Fixed-length code decoding means for decoding, wherein when decoding the variable-length coded data by the forward decoding means and the backward decoding means, the first and second decoding value tables are used for the start and end. Decoding a specific code that indicates When, referring to the second decoding value table, characterized in that decoded by the fixed-length code decoding means is regarded as a fixed length code the code following the specific code indicating the beginning.

By doing so, for example, if an information symbol having a low occurrence probability is encoded as a fixed length code with an escape code added, the fixed length code portion can be decoded independently of the reversible code. Therefore, even if the number of information symbols that can be input increases, the maximum code length can be suppressed and the storage amount can be reduced.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a variable length coding / decoding device according to an embodiment of the present invention. The variable length coding / decoding apparatus of the present embodiment is roughly divided into a codeword table creating unit 101, a coding unit 114, a transmission system or storage system 104, and a decoding unit 115. First, the functions of these units will be briefly described. The codeword table creation unit 101 creates a codeword table based on the occurrence probability of information symbols, and the coding table 102 in the coding unit 114 and the decoding unit 115. To the forward codeword table 110 and the backward codeword table 112. The encoding unit 114 encodes the information symbol into a variable length code and outputs the variable length code as encoded data to the transmission system or the storage system 104. Decoding section 115 decodes the coded data input via transmission system or storage system 104 to reproduce the original information symbol.

Next, the detailed configuration and operation of each part of this embodiment will be described. In the encoding unit 114, the input information symbol is input to the encoder 103. On the other hand, the code word table 102 stores the information symbols previously created by the code word table creating unit 101 and the code words of the variable length code in association with each other. The encoder 103 selects a codeword corresponding to the input information symbol from the codewords stored in the codeword table 102 and outputs it as encoded data. This encoded data is sent to the decoding unit 115 via the transmission system or the storage system 104.
At this time, a synchronization code is inserted into the encoded data at regular intervals.

In the decoding unit 115, the synchronization code detecting unit 105 detects the synchronization code from the encoded data input from the transmission system or the storage system 104, and the encoded data between the synchronization code is stored in the buffer 106. accumulate. The forward decoding unit 110 starts decoding from the head of the encoded data accumulated in the buffer 106 based on the forward decoding tree from the forward decoding tree creating unit 111, and the backward decoding unit 108 Based on the backward decoding tree from the backward decoding tree creating unit 113, the decoding is started from the end of the encoded data accumulated in the buffer 106.

In the decoded value decision unit 109, the decoding result obtained by the forward decoder 107 (referred to as the forward decoding result) and the decoding result obtained by the backward decoder 108 (referred to as the backward decoding result). The decoded value is determined from and the final decoded result is output. That is, the decoded value determination unit 10
In No. 9, if there is an error in the encoded data, the existence of an error can be known because a bit pattern that does not appear in the decoding tree occurs, so the decoded value is determined from the forward decoding result and the backward decoding result as shown in FIG. . FIG. 2 shows a method of determining a decoded value between sync codes.

First, as shown in FIG. 2A, when the forward decoding result and the backward decoding result do not intersect the position of the code word where the error is detected (error detection position), no error is detected. Only the decoding result is used as the decoding value, and the decoding results of the two error detection positions are not used as the decoding value and are discarded.

Further, as shown in FIG. 2B, when the error detection positions of the forward decoding result and the backward decoding result intersect, the decoding result in which no error is detected is used as the decoding result value. . Also, in this case, the decoding result between the code words at the two error detection positions is not used as the decoded value,
Abandon.

Further, as shown in FIG. 2C, when an error is detected only in the unidirectional decoding result of the forward decoding result and the backward decoding result (in this example, only the forward decoding result has an error. Detected) discards the decoded value for the code word at the error detection position, and the decoded values for the code words thereafter use the decoding result in the reverse direction.

Further, when an error is detected in both the forward decoding result and the backward decoding result for the same codeword as shown in FIG. 2D, the error detection position is the same as in FIG. 2C. The decoded value for the code word of is discarded, and the decoded result in the reverse direction is used as the decoded value for the code words after that.

The code word table creating section 101 creates a code word table of codes that can be decoded in the forward direction and the backward direction based on the occurrence probability of the information symbol. The codeword table created by the codeword table creating unit 101 is
It is sent to the codeword table 102 in the encoding unit 114, and the forward codeword table 110 and the backward codeword table 112 in the decoding unit 115. Forward decoding tree creation unit 1
At 11, a forward decoding tree is created from the forward codeword table 110. Further, in the backward decoding tree creating unit 113,
A backward decoding tree is created from the backward codeword table 112.

FIG. 3 is a block diagram showing the configuration of the code word table creating section 101. The code selecting unit 21 receives the information of the occurrence probability of the information symbol, selects the one having the smallest average code length from the selectable code systems based on this information, and outputs the selection result to the code word forming unit 22. send. The code word constructing unit 22 composes the code word of the code selected by the code word selecting unit 21 and creates a code word table. The codeword table created by the codeword constructing unit 22 is the codeword table 102 and the decoding unit 11 in the encoding unit 114.
5 to the forward codeword table 110. The codeword table described in the reverse direction to the forward direction is sent to the backward codeword table 112 in the decoding unit 115.

In the code selection unit 21, the input information symbols are rearranged in the descending order of the occurrence probability, and S = {S1, S
, ..., Sn}, and from the set C of configurable codes based on the occurrence probability P = {p1, p2, ..., Pn} of these information symbols S,

[0028]

[Equation 1]

A code c satisfying the above condition is selected. Where L
i is a code length that can be calculated by the weight of the codeword given by the codeword construction unit 22. The weight of the codeword in this case refers to the number of "1" or "0" in the codeword.

The method of constructing the codeword in the codeword constructing section 22 will be described below.

FIG. 4 shows an example of how to create a binary code sequence with a constant weight, which is the basis of the variable length code according to this embodiment. As shown in FIG. 4A, from the start point to e
A grid-like directed graph reaching the nd point is formed. When tracing the route from the start point to the end point in this directed graph, "0" is made to correspond when the left route is selected, and "1" is made when the right route is selected. The binary code sequence shown in is generated.
In this example, the code length is 5 bits and the weight (the number of "1")
Generate a binary code sequence of 2. Generally, there are as many binary code sequences having a code length of n bits and a weight of w as the number of combinations for selecting w from n. In this case, there are 10 binary code sequences, which is the number of combinations for selecting 2 from 5.

FIG. 5 shows a first method of constructing a code word of a variable length code (hereinafter referred to as a reversible code) that can be decoded in the forward direction and the backward direction in the code word construction unit 22. First, as shown on the left side of FIG. 5, the weight (the number of “1”) is fixed by the method of FIG. In this case, one binary code sequence is created. Next, as shown on the right side of FIG. 5, a code word of the reversible code is constructed by adding "1" to the beginning and the end of each of the binary code sequences and assigning 0 to the shortest code. This reversible code always starts with a code “0” for the information symbol A and starts with a code “1” for the other information symbols B to J, and when three “1” appear in total. The code configuration ends the code, that is, the code delimiter (code length) is known.

The necessary and sufficient condition for the variable-length code to be a reversible code, that is, a code that can be decoded in the forward direction and the backward direction, is that all codewords are assigned to the leaves of the forward decoding tree and the leaves of the backward decoding tree. Is to be done. The leaf of the decryption tree is an end of the decryption tree, that is, a portion where nothing exists before that. For example, in the variable-length code in FIG. 5, the codewords corresponding to all the information symbols A to J are in the leaves of the forward decoding tree shown in FIG. 6A and in the leaves of the backward decoding tree in FIG. 6B. Since it is also assigned, it can be understood that decoding is possible in the forward direction and the backward direction.

The parameter of the method of constructing the code word of the variable length code of FIG.
By setting the above natural number w, it is possible to cope with the occurrence probabilities of various information symbols. Of course, the same applies when the bits are inverted and the number of “0” is considered as the weight.

FIG. 7 shows a second method of constructing the codeword of the reversible code in the codeword constructing unit 22. First, as shown on the left side of FIG. 7, only 2 (information symbols / 2) whose code lengths are ascending in order of the weight (the number of “1”) (one in this case) by the method of FIG. Create a binary code sequence. Next, as shown in the center of FIG. 7, “1” is added to the beginning and the end of each of the binary code sequences, and as shown on the right side of FIG. Add a word.

In this variable length code, the code length can be known by counting the number of symbols at the beginning of each code. In the example of FIG. 7, the first symbol of each code is 4
The code configuration is such that the code ends when one appears, that is, the code delimiter (code length) is known.

In the variable-length code of FIG. 7, the code words corresponding to all the information symbols A to J are in the leaves of the decoding tree in the forward direction shown in FIG. 8A, but in the backward direction of FIG. 8B. Since it is also assigned to the leaves of the tree, it can be seen that it can be decoded in the forward direction and the backward direction.

FIG. 9 shows a third method of constructing a codeword of a reversible code in the codeword constructing section 22. This reversible code has a code configuration in which the code length of the code word can be known by the same number of "0" s and "1" s. That is, considering the grid-like directed graph shown in FIG. 9A, “0” is selected when the left route is selected from the start point in this directed graph, and “1” is selected when the right route is selected. Correspond, sta
A path reaching a black circle point on a diagonal line passing through the rt point is generated as a code word.

In this case, the codewords corresponding to all the information symbols A to J are assigned to both the leaves of the forward decoding tree shown in FIG. 10A and the leaves of the backward decoding tree shown in FIG. 10B. From this, it can be seen that decoding can be performed in the forward direction and the backward direction.

Next, a method of shortening a reversible code according to the present invention will be described. Since the number of information symbols, that is, the number of codes of the variable length code is finite, it is possible to shorten a part of the codes of the variable length code. Here, shortening means reducing the code lengths of some codes without increasing the code lengths of other codes. FIG. 11 shows an example of a method of shortening a reversible code.

For example, the reversible code shown in FIG. 7 is decoded from both the forward and reverse directions even if the last 1 bit of the code word corresponding to the information symbols G, H, I and J is deleted. It is possible. By utilizing this, in FIG. 11, the code lengths of the four code words corresponding to the information symbols G, H, I, and J are shortened.

The variable length code shown in FIG. 11 is
The code words corresponding to all the information symbols A to J are shown in FIG.
Since it is assigned to both the leaves of the forward decoding tree shown in (a) and the leaves of the backward decoding tree of FIG. 12 (b), it can be seen that decoding is possible in both the forward direction and the backward direction.

Next, the method for extending the reversible code according to the present invention will be described with reference to FIG. The extension of the reversible code can be realized by adding an equal-length code to at least one of the beginning and the end of each code to increase the number of code words having the same code length. In the example of FIG. 13, a 2-bit equal-length code is added to the end of each reversible code shown on the left side. In this case, the code length of each reversible code is increased by 2 bits as a whole, but the number of code words having the same code length is expanded four times. Generally, when an n-bit equal-length code is added, the code length is increased by n bits as a whole, but the number of code words having the same code length can be increased by 2n. Since the equal-length code is obviously a reversible code, even if the equal-length code is added to the beginning or the end of the reversible code, it is still a reversible code.

FIG. 14 shows a known example of a reversible code obtained by performing variable length coding according to the present invention on an information symbol which is an English alphabet (Japanese Patent Laid-Open No. 5-300).
This is shown in comparison with the reversible code disclosed in Japanese Patent Publication No. 027). The reversible code according to the present invention shown in FIG. 14 is a code obtained by expanding the code when the weight of the code word is 0 in the method shown in FIG. 5 by adding a 2-bit equal-length code to the end. Is.

The reversible code according to the present invention is
Although it is inferior to the Huffman code, which is the optimum variable-length code that can be decoded only in the forward direction, it can be seen that the average code length is shorter than that of the reversible code disclosed in the known example and that it has excellent performance. This is because the known reversible code has a bit added to the end of a Huffman code that is a variable length code that can be decoded only in the forward direction, whereas the present invention does not add an extra bit. This is because the reversible code is formed from the beginning, and there are few wasted bit patterns.

Next, another embodiment of the present invention will be described.
FIG. 15 is a block diagram illustrating variable length coding / coding according to another embodiment of the present invention.
It is a block diagram which shows the structure of a decoding device. The variable-length coding / decoding device of the present embodiment roughly includes a codeword table creation unit 201, a coding unit 213, a transmission system or storage system 205, and a decoding unit 214.

The function of each of these units will be briefly described. The code word table creating unit 201 creates a code word table based on the occurrence probability of information symbols, and the coding unit 21.
3 to the code word table 202 in the H. No. 3, and further sends the parameter of the code word of the created code word table to the decoding graph / decoded value table creating unit 208 in the decoding unit 214. The encoding unit 213 encodes the information symbol into a variable length code and outputs the variable length code as encoded data to the transmission system or the storage system 205. The decoding unit 214 decodes the coded data input via the transmission system or the storage system 205 to reproduce the original information symbol.

Next, the detailed configuration and operation of each part of this embodiment will be described.

In the coding section 213, the input information symbols are input to the encoder 203. The code word table 202 stores the information symbols previously created by the code word table creating unit 201 and the code words of the variable length code in association with each other. However, the codeword table 202 does not store codewords corresponding to all the information symbols that can be input, but stores only codewords corresponding to some information symbols having a relatively high appearance frequency. Be present. The encoder 203 selects a codeword corresponding to the input information symbol from the codewords stored in the codeword table 202 and outputs it as encoded data.

On the other hand, the code word table 2 is supplied to the encoding unit 213.
When an information symbol having no code word corresponding to 02, that is, an information symbol having a relatively low appearance frequency is input, a fixed-length code corresponding to the information symbol is created by the fixed-length code encoding unit 204, and further, Codeword table 20
The escape code in 2 is added to the beginning and end of the fixed length code to form a code word, which is output as encoded data.

Thus, the encoder 20 in the encoder 213 is
The encoded data output from No. 3 is sent to the decoding unit 214 via the transmission system or the storage system 205. At this time, a synchronization code is inserted into the encoded data at regular intervals.

The decoding unit 214 has a synchronization code detecting unit 20.
6, buffer 207, decoding graph / decoding value table creating unit 208, forward decoding unit 209, fixed length decoding unit 21
0, a backward decoder 211, and a decoded value determination unit 212. First, the synchronization code detection unit 206 detects a synchronization code from the encoded data input from the transmission system or storage system 205, and detects the synchronization code and the synchronization code. The coded data in between are accumulated in the buffer 207. The forward decoder 209 starts decoding from the beginning of the encoded data stored in the buffer 207, and the backward decoder 211 starts decoding from the end of the encoded data stored in the buffer 207. In the decoded value determination unit 212, the forward decoder 209
Decoding result obtained by (called forward decoding value),
The decoding value is determined from the decoding result (referred to as the backward decoding value) obtained by the backward decoding unit 211, and the final decoding result is output.

FIG. 16 is a diagram showing a method of constructing a code word by adding an escape code to the beginning and the end of a fixed length code in the code word table creating unit 201 in FIG. As shown in FIG. 16A, in the codeword created by the codeword table creating unit 201, when the information symbol is other than “A”, the number of “1” appears three times, and the code delimiter is separated. It is based on the known reversible code. This reversible code will be referred to as code (1).

On the other hand, reference numeral (2) shown in FIG.
Code word "1" corresponding to information symbol "C" of code (1)
011 "is an escape code for indicating the beginning and the end of the fixed length code, and a code word configured by adding to the beginning and the end of the 3-bit fixed length code is newly added to the code (1). However, the reversible code part corresponding to the code (1) in the code (2) is the code word “101” used as the escape code in the code (2).
1 "is not used, and only the code words corresponding to the information symbols A to I that are one less by the code (1) are used.
As is clear from comparison between (a) and (b), the code (2) has the number of codeable information symbols of “10” of the code (1).
Has increased from 17 to.

In order to decode such a code (2), the code (1) is always used for decoding both in the forward direction and in the reverse direction as can be seen from FIGS. 16 (a) and 16 (b). Since the fixed-length code can be decoded from both the forward direction and the backward direction, if the code (1) can be decoded from the forward direction and the backward direction, the code (2) can also be decoded. Decoding is possible from both directions.

In the case of the conventional reversible code, it is necessary to prepare all codewords corresponding to the information symbols that can be input as a codeword table in the encoding section 213 and the decoding section 214. On the other hand, the code (2)
According to the above, for the part of the fixed length code, 3 bits of 2
Since the fixed-length code encoding unit 204 and the fixed-length code decoding unit 210 can be separately prepared as a binary code, only the codeword of the code (1) needs to be stored in the codeword table 202. Therefore, compared with the conventional reversible code, it is possible to significantly reduce the storage amount of the variable length coding / decoding device.

Decoding graph / decoding value table creation unit 208
Creates a decoding graph based on the Sharkwijk algorithm and a decoding value table, and the forward decoder 2
09 and the backward decoder 211 perform decoding based on these decoding graph and decoding value table.

FIG. 17 shows a method of counting binary code sequences with constant weight. As an algorithm for counting binary code sequences with constant weight, JPMShalkwijk:
“Analgorithm for source coding”, IEEE Trans.Infr
om Theory, vol.IT-18, no.3, pp.395-399, May 1972.
It is well known as a halkwijk algorithm. According to the algorithm of Sharkwijk, FIG.
As shown in (a), a lattice-like directed graph from the START point to the END point is formed based on the "Pascal's triangle". The numerical value at each node is ST
The value of the node at the ART point is 1, and it is the total value of the values of the nodes at the starting points of the two arrows coming into each node.
That is, the value of each node is determined by Pascal's triangle.

Then, in the directed graph of FIG. 17A, when the route from the START point to the END point is traced according to the input information symbol, when the left side route (arrow) is selected, the code word "0", When the path (arrow) on the right side is selected, the binary code sequence with the weight of 2 can be generated by associating the code words “1” with each other. Here, when “0” is selected, the sum of the values obtained by subtracting the value of the node at the starting point of the arrow from the value of the node at the ending point of the arrow is the codeword of the same code length of the binary code sequence. Is the ordinal value of. For example, if the input information symbol is “01001”, it is the sum of the differences of the node values with respect to three “0” s (1-1) + (3-
2) + (4-3) = 2 is the ordinal value. Here, the sequential value is a value for ordering different code words having the same code length. FIG. 17B shows the relationship between the information symbol and the order value in this case.

An example in which the Sharkwijk algorithm is applied to encoding is publicly known, and in that case, encoding is performed by converting a value corresponding to the above-mentioned order value into a binary code. On the other hand, in the present embodiment, this algorithm is used for decoding as described below.

That is, the decoding graph / decoding value table creating section 208 creates a directed graph as a decoding graph based on the above-described Sharkwijk algorithm. That is, this decoding graph is a directed graph in which the values determined by the Pascal's triangle are the values of the nodes and the codewords "1" and "0" are arrows, and an example thereof is shown in FIG.
It shows in (a). FIG. 18B shows an example of a method of calculating a decoded value in the forward decoder 208 and the backward decoder 211 using this decoding graph.

The code (1) shown in FIG. 16 (a) is a reversible code in which the delimiter can be recognized at the point where three "1" s appear when the head is other than "0". When calculating the decoded value of the reversible code of this code (1), the decoded value is set to 0 if the head of the reversible code is "0" in the forward direction, and "1" at the head and the tail otherwise. After deleting, the ordinal value is calculated by the algorithm of Sharkwijk. Now, the number of codewords having a shorter code length than the codeword to be decoded is EN in FIG.
Since the sum of the values of the right diagonal nodes at the point D is +1, the sum of the values to the sequence value becomes the decoded value.

On the other hand, in the reverse direction, as in the case of the above-mentioned forward direction, first, if the beginning of the reversible code is "0", the decoded value is set to 0, and in other cases, the leading and trailing "1" are deleted. Then, after inverting the rest, S
The ordinal value is calculated by the algorithm of halkwijk.
Then, the sum of this order value and the sum of the nodes on the diagonally right upper side of the END point +1, is taken as the decoded value.

More specifically, referring to FIGS. 18 (a) and 18 (b), for example, if the codeword to be decoded by the reversible code is "10101", first the leading and trailing "1" s are deleted and "1" is deleted. 010 ". Sh for this "010"
When the ordinal value is calculated by the alkwijk algorithm, 2
For a certain “0”, (1-1) + (3-2) = 1. In this case, the sum of the values of the nodes diagonally upper right of the END point is 1 + 2 = 3, so this value +1 (= 4) is added to the sequence value (= 1), and the decoded value is obtained as 5.

When decoding is performed using such a decoding graph, the storage amount can be greatly reduced. FIG. 19
(A) and (b) are a decoding method using a conventional decoding tree,
FIG. 11 is a diagram showing a difference in storage amount required in the decoding method using the decoding graph according to the present embodiment. Comparing the two by the number of nodes, the number of arrows coming out from each node is the same, but the number of arrows coming into each node is always one in the case of the decoding tree of FIG. 19A. . On the other hand, in the case of the decoding graph of FIG. 19B, since two arrows enter each node, the total number of nodes required for the same codeword decreases accordingly. In this example, the decoding tree in FIG. 19 (a) requires 15 nodes, whereas the decoding graph in FIG. 19 (b) requires only 8 nodes. Therefore, the decoding method of the present embodiment using the decoding graph has an advantage that the required storage amount is smaller than that of the conventional method using the decoding tree.

Further, in this embodiment, in order to decode the code (2) to which the escape code shown in FIG. 16B is added, the decoded value table of the code (1) shown in FIG. Two of the fixed-length code decoded value tables shown in 20 (b) are prepared, and when the escape code is decoded in the decoded value table of code (1), the fixed-length code decoded value table is read. For example, the code word is “10110011
In the case of 011, when the leading "1011", that is, the escape code, is decoded, the next 3-bit "001" is regarded as a fixed length code and is decoded by the fixed length code decoding unit 210. In this case, Since 1 is obtained as the decoded value of "001", the decoding result becomes the information symbol "K" from the decoded value table of the fixed length code in FIG.

Next, an embodiment in which the present invention is applied to a moving picture encoder / decoder will be described. FIG. 21 is a block diagram showing a schematic configuration of a moving picture encoder / decoder in which the above-mentioned variable length encoder / decoder is incorporated.

The moving picture encoder 709 shown in FIG.
In the above, the data coded by the information source encoder 702 is subjected to variable length coding, channel coding and multiplexing in the moving picture multiplexing unit 703, and further the transmission buffer 704.
After the transmission rate has been smoothed by, the data is sent to the transmission system or storage system 705 as encoded data. The encoding control unit 701 considers the buffer amount of the transmission buffer 704, and the information source encoder 702 and the moving image multiplexing unit 70.
3 is controlled.

On the other hand, in the moving picture decoder 710 shown in FIG. 21 (b), the coded data from the transmission system or storage system 705 is temporarily stored in the reception buffer 706, and is coded by the moving picture demultiplexing unit 707. After demultiplexing the encoded data, channel code decoding and variable length code decoding,
It is sent to the information source decoder 708, and finally the moving image is decoded.

FIG. 22 shows the syntax of the moving picture coding system in the moving picture encoder 709 and the moving picture multiplex / separation unit 707 in FIG. In the picture layer shown in FIG. 22A, macroblock mode information, motion vector information, and information other than DCT coefficients such as INTRA DC are arranged in the upper layer, and DCT coefficient information is arranged in the lower layer. Therefore, the reversible code according to the present invention is applied to the lower layer portion.

FIGS. 23A and 23B are block diagrams showing more detailed structures of the moving picture multiplexing unit 703 and the moving picture multiplexing separating unit 707 in FIG. FIG. 23 (a)
In the moving picture multiplexing unit 703 shown in FIG. 21, among the encoded data from the information source encoder 702 in FIG. 21, macroblock mode information, motion vector information, and INTRAD
Although information other than the DCT coefficient, such as C, is used as an upper layer, normal variable length coding is performed by the upper layer variable length encoder 901, and then redundancy is further increased by the upper layer communication channel encoder 902. The channel is coded by an error correction detection code having a high correction capability and sent to the multiplexing unit 905.

On the other hand, among the encoded data from the information source encoder 702, the DCT coefficient is the lower layer variable length encoder 9
It is encoded into a reversible code at 03, and further, is channel-coded at the lower layer communication channel encoder 904 by an error correction detection code having a low degree of redundancy, and then sent to the multiplexing unit 905. The multiplexing unit 905 multiplexes the encoded data of the upper layer and the encoded data of the lower layer,
Send to 4.

The moving picture multiplexing / separating unit 7 shown in FIG. 23 (b).
In 07, first, the coded data from the reception buffer 706 is separated by the demultiplexing unit 906 into an upper layer and a lower layer. The encoded data of the upper layer is decoded by the upper layer channel decoder 907, and the decoding result is sent to the upper layer variable length decoder 909. The encoded data of the lower layer is decoded by the lower layer channel decoder 908, and the decoding result is sent to the lower layer variable length decoder 910.

The lower layer variable length decoder 910 decodes the reversible code and outputs the decoding result to the information source decoder 70.
8 and higher layer variable length decoder 909. The upper layer variable length decoder 909 decodes the variable length code that is the encoded data of the upper layer, and rewrites the coding result based on the decoding result of the lower layer variable length decoder 910.

Here, the lower layer variable length coder 903 in FIG. 23 (a) is transmitted to the coder 213 in FIG.
The lower layer variable length decoder 910 in (b) corresponds to the decoding unit 214 in FIG. 15, respectively.

As seen in the moving picture encoder / decoder of this embodiment, when the encoding system has the syntax shown in FIG. 22, only the variable length code itself can be decoded from both directions. Not only that, it is also necessary to enable decoding from both directions in terms of syntax. In the present embodiment, this request is realized by using the codeword table described below.

24 to 28 show an example of a code word table of variable length codes of DCT coefficients used in the lower layer variable length encoder 903. FIG. 29 shows a codeword table of escape codes.

In the information source encoder 702, the quantized 8
For a block of DCT coefficients of × 8, scan inside the block is performed, and LAST (0: non-zero coefficient not at the end of block, 1: non-zero coefficient at the end of block), RUN
(Number of zero runs up to non-zero coefficient) and LEVEL (absolute value of coefficient) are obtained and sent to the moving picture encoder 709.

The lower layer variable length coder 903 in the moving picture coder 709 has the non-last coefficients of INTRA and INTER shown in FIGS. 24 to 25, RUN and LEVEL.
To the reversible code (VLC_CODE) for the non-LAST coefficient codeword table (first codeword table), and INTRA and INT shown in FIGS.
It has a codeword table (second codeword table) of LAST coefficients in which reversible codes (VLC_CODE) are associated with LAST coefficients of ER, RUN, and LEVEL. Then, based on the mode information, the codeword table of the INLAST non-LAST coefficient and the LAST coefficient is selected in the case of INTRA, and the codeword table of the INTER non-LAST coefficient and the LAST coefficient is selected in the case of INTER, respectively. 26 to 27, VLC_
The last bit "S" of CODE represents the sign of LEVEL, and when "S" is "0", the sign of LEVEL is positive,
It is negative when "1".

As for coefficients not existing in this codeword table, absolute values of RUN and LEVEL are encoded into a fixed length code as shown in FIG. 29, and escape codes are added to the head and end of this fixed length code. The last bit of the escape code is encoded so that the sign of the LAST coefficient and LEVEL can be distinguished. FIG. 28 is a code word table of this escape code, and the last bit “t” of VLC_CODE used as an escape code indicates whether or not it is a LAST coefficient when added to the head of the fixed length code,
When it is "0", it is a non-LAST coefficient, and when it is "1", it is a LAST coefficient. Further, "t" represents the LEVEL code when added to the end of the fixed-length code, and when it is "0", the LEVEL code is positive and when it is "1", it is negative.

FIG. 30 shows an example of encoded data in this embodiment. As shown in the figure, when decoding is performed in the forward direction in the lower layer, D of 8 × 8 pixels is used.
Since the code of the LAST coefficient always exists at the end of the block of CT coefficients, the end of the block can be determined. On the other hand, when decoding is performed from the reverse direction, the head of the block can be determined by the appearance of the code of the LAST coefficient of the block whose address is one before. This code is LA
Since the ST coefficient is shared between the INTRA mode and the INTER mode, the head of the block can be determined even if the mode exists for each macroblock.

For the first block in the lower layer, the lower layer variable length encoder 903 encodes a dummy LAST coefficient at the beginning of the lower layer in advance, or the lower layer variable length decoder 910 The number of bits of the lower layer is calculated in advance from the number of bits up to the synchronization code of the next frame, and this is compared with the number of decoded bits, or the lower layer variable length decoder 910 uses a buffer in the lower layer. By adopting a method of inserting a dummy LAST coefficient at the beginning of the, it is possible to determine the beginning of the lower layer when decoding in the reverse direction.

In the case of the reversible code stored in the code word table shown in FIGS. 24 to 27, in the forward direction, if the leading 1 bit is "0", it is read until two "0" s appear, and " If it is "1", it is understood that two "1" s are read until the next one bit, and the next 1 bit is the final bit representing the positive or negative of LEVEL. In the reverse direction, the first 1 bit represents the positive / negative of LEVEL, and the next 1 bit is “0”.
If so, read until two "0" appear, and if "1", read until two "1" appear.

FIG. 31 shows a decoding graph for decoding a reversible code applied to the variable length code of the DCT coefficient used in the lower layer variable length encoder 903 in this embodiment. The decoded value can be calculated by this decoding graph for the encoded data excluding the first bit and the last 2 bits of the code word when viewed from the forward direction.

In this decoding graph, the first bit is "0".
Then, if "0" appears, proceed with the right arrow, and if "1" appears, proceed with the left arrow. If the first bit is "1", if the "1" appears, the right arrow is advanced, and if the "0" appears, the left arrow is advanced.

In order to obtain the sequence value for the binary code sequences having the same code length, according to the previously described Sharkwijk algorithm, when the left arrow is selected, the value of the end point of the arrow is changed to the start point of the arrow. The sum of the values obtained by subtracting the values may be the ordinal value for that code length.

On the other hand, in order to obtain the decoded value, if the leading bit is "0", a value obtained by adding twice the sum of the values of the nodes diagonally to the right of the terminal node to the sequence value is the decoded value. do it.
If the first bit is "1", the decoded value may be twice the sum of the values of the nodes diagonally to the right of the terminal node + the value of the terminal node plus the sequence value.

For example, the code string to be decoded is "01101".
If it is 01 ”, first, looking from the forward direction, the first 1 bit and the last 2 bits are removed and set to“ 1101. ”Since the first bit is“ 0 ”, if“ 0 ”appears in the decoding graph, the right arrow is displayed. Proceed and if the "1" appears, proceed to the left sign.
In this case, the difference between the nodes when the left arrow is advanced is (1-
1) + (1-1) + (4-3) = 1, so the ordinal value is 1
Is. Since the most significant bit is “0”, it is twice the sum of the nodes diagonally to the right of the end node (1 + 2 + 3) × 2 = 1
If order 1 is added to 2, it becomes 13, and the decoded value of this code word is obtained as 13.

32 to 34 are examples of the decoded value table used in the lower layer variable length decoder 910.
33. FIG. 33 shows non-LAST coefficients of INTRA and INTER, non-LAS in which decoded values correspond to RUN and LEVEL.
34A and 34B are decoded value tables of T coefficients, and FIGS. 34A and 34B show LAST coefficients of INTRA and INTER, RUN,
It is a decoded value table of the LAST coefficient which matched the decoded value with LEVEL. Further, FIG. 34C is a decoded value table of escape codes.

In this example, for example, the decoded value 13 is shown in FIG.
As can be seen from (a), it is found that the LAST number is 3, the absolute value of LEVEL is 1, and the least significant bit is "1", so LEVEL is negative. . Furthermore, as can be seen from FIG. 34C, when the decoded value 41 is decoded, it is determined that the escape code has been decoded. The last bit of this escape code is used to determine whether the coefficient is the LAST coefficient or the non-LAST coefficient, and the fixed-length code decoding unit 2 determines the subsequent 13 bits.
After decoding the absolute values of RUN and LEVEL in 10, the escape code is decoded again, and the sign of LEVEL is determined by the last bit.

As an example, the code string to be encoded is "11".
0000010000111000101111000
Consider the case of 010 ”. In this case, since the last bit of the escape code“ 11000010 ”at the beginning is“ 0 ”, it is a non-LAST coefficient, and the fixed length code of the subsequent 13 bits is calculated. Since 6 bits represent RUN, "000111" means RUN is 7,
Since the lower 7 bits represent the absolute value of LEVEL, "00"
With 01011, the absolute value of LEVEL is 12, and the last bit of the escape code "11000010" at the end is "0".
Therefore, LEVEL can be decoded as positive.

In the decoded value determination unit 212, the decoding result thus obtained by the forward decoding unit 209 (referred to as forward decoding result) and the decoding result obtained by the backward decoding unit 211 (reverse direction decoding). The decoded value is determined from the (decoded value) and the final decoded result is output. That is, if the decoded value determination unit 212 has an error in the encoded data, the decoding result of the lower layer communication channel decoder 908 or a bit pattern that does not exist as a code word occurs, so that the existence of the error is known. Therefore, the decoded value is determined from these forward decoding result and backward decoding result as shown in FIG. Note that FIG. 36 shows a method for determining the decoded value of the lower layer.

First, as shown in FIG. 36 (a), no error was detected when the positions of macroblocks in which an error was detected (error detection positions) did not intersect in the forward decoding result and the backward decoding result. Only the decoding result of the macroblock is used as the decoding value, and the decoding results of the two error detection positions are not used as the decoding value. Then, based on the decoding result of the mode information of the upper layer, the INTE macroblock is displayed so that the previous frame is displayed as it is.
For the R macroblock, the decoding result of the upper layer is rewritten so that the R macroblock is displayed only by motion compensation from the previous frame.

Further, as shown in FIG. 36B, when the forward decoding result and the backward decoding result intersect at the error detection position, the decoding result in which no error is detected is used as the decoding result value. To do. Further, in this case, the decoding result between the code words at the two error detection positions is not used as a decoding value, and the IN result is determined based on the decoding result of the mode information of the upper layer.
The decoding result of the upper layer is rewritten so that the previous frame is displayed as it is for the TRA macroblock and the motion compensation is displayed only for the INTER macroblock from the previous frame.

Further, as shown in FIG. 36 (c), when an error is detected only in the unidirectional decoding result of the forward decoding result and the backward decoding result (in this example, only the forward decoding result has an error. Has been detected), the macro block at the error detection position is not used as a decoded value, and the previous frame is displayed as it is for the INTRA macro block based on the decoding result of the mode information of the upper layer.
For the INTER macroblock, the decoding result of the upper layer is rewritten so that the INTER macroblock is displayed only by motion compensation from the previous frame. The decoding value for the subsequent macroblock uses the decoding result in the reverse direction.

Further, as shown in FIG. 36D, when an error is detected in both the forward decoding result and the backward decoding result in the same macroblock, the decoded value of the macroblock at the error detection position is abandoned. And do not use it as a decrypted value,
Based on the decoding result of the upper layer mode information, INTR
The decoding result of the upper layer is rewritten so that the previous frame is displayed as it is for the A macroblock and the motion compensation is displayed only for the INTER macroblock from the previous frame. Use the decoding result of.

Regarding the order of encoding / decoding of macroblocks, a method other than the above embodiment may be used. For example, when an error occurs, the more important screen central portion of the decoded image is relieved. A method in which the order is as shown in FIG. 37 may be used. That is, assuming that decoding is performed from both directions, there is a high probability that the first part and the last part of the encoded data between the synchronization code will be correctly decoded. The order is decided so that is in the center.

Although the present invention is applied to the variable length coding of DCT coefficients in the above embodiment, it goes without saying that the same variable length coding can be applied to other information symbols.

In the above embodiment, only binary codes have been discussed, but it is easy to extend to multi-level codes and similar effects can be obtained.

Finally, as an application example of the present invention, an embodiment of an image transmitting / receiving apparatus to which the variable length coding / decoding apparatus according to the present invention is applied will be described with reference to FIG. The image signal input from the camera 1002 provided in the personal computer (PC) 1001 is encoded by the variable length encoding device incorporated in the PC 1001. The information symbol in this case is not limited to this,
For example, it may be DCT coefficient data obtained by performing a discrete cosine transform on the input image signal or a difference between the input image signal and the predicted image signal by the DCT circuit, and further quantizing the signal by the quantization circuit. The coded data output from the variable-length coding device is multiplexed with other voice and data information, and then wirelessly transmitted by the wireless device 1003 and received by the other wireless device 1004. Radio 1
The signal received at 004 is decomposed into encoded data of an image signal and information of voice and data. Of these,
The encoded data of the image signal is the workstation (EW
S) Decoded by the variable length code decoding device incorporated in 1005, and displayed on the display of EWS 1005.

On the other hand, the image signal input from the camera 1006 provided in the EWS 1005 is the EWS 100
It is coded in the same manner as above using the variable length coding device incorporated in H.6. The encoded data is multiplexed with other audio and data information, wirelessly transmitted by the wireless device 1004, and received by the wireless device 1003. Radio 100
The signal received by 3 is decomposed into encoded data of an image signal and information of voice and data. Of these, the coded data of the image signal is decoded by the variable length code decoding device incorporated in the PC 1001, and the PC 10
01 display.

The variable-length coding / decoding apparatus of the present invention has a code length according to the frequency of appearance of information symbols, has less wasteful bit patterns than the conventional one, is efficient, and has a forward direction. Since a reversible variable-length code that can be decoded in the reverse direction can be configured, it is particularly effective in a transmitting / receiving device using a transmission path with many errors such as the wireless transmission path shown in FIG.

As described above, the variable-length coding / decoding device of the present invention can decode with a smaller storage amount as compared with the conventional one, and the information symbol which cannot be applied conventionally. It can also be applied to coding with a large number. Specifically, it can be applied to a moving picture coding / decoding apparatus, and an error resistant moving picture coding / decoding apparatus can be provided.

[0104]

As described above, according to the present invention, it is possible to perform decoding in the forward direction and the backward direction, and the variable length is practical in terms of the coding efficiency and the storage amount required for the device. A decoding method and device can be provided.

That is, an information symbol having a low appearance probability is encoded as a fixed length code with an escape code added,
Since the fixed-length code part can be encoded / decoded independently of the reversible code, an extremely large number of information symbols such as quantized orthogonal transform coefficients in the image encoder / decoder are encoded. Even in such a case, the maximum code length can be suppressed to effectively reduce the storage amount.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a variable length coding / decoding device according to an embodiment of the present invention.

FIG. 2 is a diagram for explaining the operation of a decoded value determination unit in FIG.

FIG. 3 is a block diagram of a codeword table creation unit in FIG.

FIG. 4 is a diagram showing a method of creating a binary code sequence with a constant weight, which is a basis of a variable length code according to the present invention.

5 is a diagram showing a first codeword constructing method in the codeword constructing unit in FIG. 3;

FIG. 6 is a diagram showing forward and backward decoding trees created from codewords configured by the first code configuration method.

FIG. 7 is a diagram showing a second codeword constructing method in the codeword constructing unit in FIG. 3;

FIG. 8 is a diagram showing forward and backward decoding trees created from codewords constructed by a second code construction method.

9 is a diagram showing a third code structuring method in the code word structuring unit in FIG. 3;

FIG. 10 is a diagram showing forward-direction and backward-direction decoding trees created from codewords configured by the third code configuration method.

FIG. 11 is a diagram showing a code shortening method in the code word construction unit in FIG. 3;

FIG. 12 is a diagram showing forward and backward decoding trees created from codewords shortened by a code shortening method.

FIG. 13 is a diagram showing a code enlargement method in the code word configuration unit in FIG.

FIG. 14 is a diagram showing characteristics of a variable-length code configured in the present embodiment and a known variable-length code in comparison.

FIG. 15 is a block diagram illustrating variable length coding / coding according to another embodiment of the present invention.
Block diagram showing the configuration of the decoding device

16 is a diagram showing a code configuration method in a codeword configuration section in FIG.

FIG. 17 is a diagram showing a method of counting binary sequences with constant weights.

FIG. 18 is a diagram showing operations of a forward decoder and a backward decoder in FIG.

FIG. 19 is a diagram showing a comparison between a conventional decoding tree and a decoding graph according to the present embodiment.

FIG. 20 is a diagram for explaining a decoded value table according to the present embodiment.

FIG. 21 is a block diagram showing a schematic configuration of a moving picture coding / decoding device in which a variable length coding / decoding device according to the present invention is incorporated.

FIG. 22 is a diagram showing the syntax of encoded data in the moving image encoder / decoder according to the embodiment.

23 is a block diagram showing a configuration of a moving image multiplexing unit and a moving image multiplexing / separating unit in FIG. 21.

FIG. 24 is an INTRA and INT in the same embodiment.
The figure which shows a part of codeword table of the non-LAST coefficient of ER.

FIG. 25 is an INTRA and INT in the same embodiment.
The figure which shows the other part of the codeword table of the non-LAST coefficient of ER.

FIG. 26 is an INTRA and INT in the same embodiment.
The figure which shows a part of codeword table of the last coefficient of ER.

FIG. 27 is an INTRA and INT in the same embodiment.
The figure which shows the other part of the codeword table of the last coefficient of ER.

FIG. 28 is a diagram showing a codeword table of escape codes in the same embodiment.

FIG. 29 is a diagram showing an encoding method of a codeword not included in the codeword table in the embodiment.

FIG. 30 is a diagram showing an example of encoded data in the same embodiment.

FIG. 31 is a diagram showing a decoding graph in the same embodiment.

FIG. 32 is an INTRA and INT in the same embodiment.
The figure which shows a part of decoding value table of the non-LAST coefficient of ER.

FIG. 33 is an INTRA and INT in the same embodiment.
The figure which shows the other part of the decoding value table of the non-LAST coefficient of ER.

FIG. 34 is an INTRA and INT in the same embodiment.
The figure which shows the decoding value table of the LAST coefficient of ER.

FIG. 35 is a diagram showing a decoded value table of escape codes in the same embodiment.

36 is a diagram for explaining the operation of the decoded value determination unit in FIG.

FIG. 37 is a diagram showing an encoding / decoding order of macroblocks in the embodiment.

FIG. 38 is a diagram showing an example of an apparatus in which the variable length coding / decoding apparatus according to the present invention is incorporated.

FIG. 39 is a diagram showing a general decoding method for reversible codes.

FIG. 40 is an explanatory diagram of a normal variable length code.

FIG. 41 is an explanatory diagram of a conventional reversible code.

[Explanation of symbols]

101 ... Codeword table creation unit 102 ... Codeword table 103 ... Encoder 104 ... Transmission system or storage system 105 ... Sync code detection unit 106 ... buffer 107 ... Forward decoder 108 ... Inverse decoder 109 ... Decoded value determination unit 110 ... Forward decoding word table 111 ... Forward decoding tree creation unit 112 ... Reverse codeword table 113 ... Reverse decoding tree creation unit 114 ... Encoding unit 115 ... Decoding unit 21 ... Code selection unit 22 ... Codeword constructing section 201 ... Codeword table creation unit 202 ... Code word table 203 ... Encoder 204 ... Fixed-length code encoder 205 ... Transmission system or storage system 206 ... Sync code detector 207 ... buffer 208 ... Decoding graph / decoding value table creation unit 209 ... Forward decoder 220 ... Fixed-length code decoding unit 211 ... Inverse decoder 212 ... Decoded value determination unit 213 ... Encoding unit 214 ... Decoding unit 701 ... Encoding control unit 702 ... Information source encoder 703 ... Moving image multiplexing unit 704 ... Transmission buffer 705 ... Transmission system or storage system 706 ... Receive buffer 707 ... Moving image multiplexing / separating unit 708 ... Information source decoder 709 ... Moving image encoder 710 ... Video decoder 901 ... Lower layer variable length decoder 902 ... Upper layer communication channel encoder 903 ... Lower layer variable length encoder 904 ... Lower layer communication channel encoder 905 ... Multiplexing unit 906 ... Multiplexing / separating unit 907 ... Upper layer communication channel decoder 908 ... Lower layer channel decoder 909 ... Upper layer variable length decoder 910 ... Lower layer variable length decoder 1001 ... Personal computer 1002 ... work station 1003 ... radio 1004 ... radio 1005 ... camera 1006 ... camera

Continuation of the front page (56) Reference JP-A-5-300027 (JP, A) JP-A-4-213988 (JP, A) JP-A-6-21830 (JP, A) JP-A-6-237184 (JP , A) Y. Takashima, et al, Reversible Variable length Codes, IEE TRANSACTIONS ON COMMUNICATIONS, USA, February 1995, Vol. 43, Issue 162, p. ⁷ , DB name) H03M 7/42 H04N 7/30

Claims (8)

(57) [Claims] 1. A variable length for decoding variable length encoded data which includes a fixed length code having a head and a tail added with a specific code indicating the head and the tail and which can be decoded in a forward direction and a backward direction. A decoding method, wherein when decoding the variable-length coded data from a forward direction or a backward direction, using the first decoded value table storing decoded values of specific codes indicating the head and the tail, When the specific code indicating the head and the end is decoded, the code following the specific code indicating the head is fixed length code by referring to the second decoded value table storing the decoded value of the fixed length code. A variable length decoding method, which is characterized in that 2. Decoding in a forward direction or a backward direction, which includes a code in which a specific code that indicates the beginning and the end of a fixed-length code and that can be decoded in the forward direction or the backward direction is added A variable-length decoding method for decoding variable-length coded data, wherein when decoding the variable-length coded data from a forward direction or a backward direction, the forward and backward directions indicating the head and the tail are shown. When a specific code that can be decoded in the forward direction or the backward direction indicating the head and the end is decoded using the first decoded value table that stores the decoded value of the specific code that can be decoded,
By referring to the second decoded value table that stores the decoded value of the fixed length code, the code following the specific code that can be decoded in the forward direction or the backward direction indicating the head is regarded as the fixed length code and decoded. A variable length decoding method comprising: 3. A forward code or a backward code, which includes a code in which a specific variable-length code that can be decoded in the forward direction or the backward direction indicating the head and the end of the fixed-length code is added. A variable-length decoding method for decoding decodable variable-length coded data, wherein when decoding the variable-length coded data from a forward direction or a backward direction, the forward direction and the backward direction indicating the beginning and the end are also reversed. Stored the decoded value of a specific variable-length code that can also be decoded by
When a specific variable-length code that can be decoded in the forward direction or the backward direction indicating the head and the tail is decoded using the decoded value table of No. 2, the second decoding that stores the decoded value of the fixed-length code A variable-length decoding method, which refers to a value table and decodes a code following a specific variable-length code that can be decoded in the forward direction or the backward direction indicating the beginning as a fixed-length code. 4. The variable length decoding method according to claim 1, wherein the specific code or the specific variable length code is an escape code. 5. A variable length for decoding variable length coded data which includes a fixed length code having a head and a tail added with a specific code indicating the head and the tail and which can be decoded in the forward direction and the backward direction. A decoding device, which is configured to create a first decoded value table storing a decoded value of a specific code indicating the start and the end, and a second decoded value table storing a decoded value of the fixed length code. A forward decoding means for decoding the variable length coded data from the forward direction, a backward decoding means for decoding the variable length coded data from the reverse direction, and a fixed length code decoding means for decoding a fixed length code. And, when decoding the variable length coded data by the forward decoding means and the backward decoding means, a specific code indicating the head and the tail is decoded using the first decoded value table. When the second Referring to issue value table, the variable length decoding apparatus characterized by decoding by the fixed-length code decoding means codes following the specific code is regarded as a fixed length code indicating the beginning. 6. A forward- or backward-direction code including a code in which a fixed-length code is added to the start and end of the fixed-length code and a specific code that indicates the start and end can be decoded in the forward or reverse direction. Variable-length decoding apparatus for decoding variable-length coded data, which is a first decoded value table storing decoded values of a specific code that can be decoded in the forward direction and the backward direction indicating the start and end. And means for creating a second decoded value table storing decoded values of the fixed-length code, forward decoding means for decoding the variable-length coded data from the forward direction, and reverse direction for the variable-length coded data. Backward decoding means for decoding from, and fixed length code decoding means for decoding fixed length code, wherein when decoding the variable length coded data by the forward decoding means and the backward decoding means, Using the first decoded value table When a specific code that can be decoded in the forward and backward directions indicating the beginning and the end is decoded, the second decoded value table is referred to and the forward or backward direction indicating the beginning is detected. A variable length decoding device characterized in that a code following a specific decodable code is regarded as a fixed length code and is decoded by said fixed length code decoding means. 7. A forward code or a backward code, which includes a code in which a specific variable-length code that can be decoded in the forward and backward directions indicating the head and the end of the fixed-length code is added. A variable-length decoding method for decoding decodable variable-length coded data, which stores a decoded value of a specific variable-length code that can be decoded in a forward direction indicating the head and an end and a backward direction. Means for creating a second decoded value table storing the decoded value table and the fixed-length code decoded value, forward decoding means for decoding the variable-length coded data from the forward direction, and the variable-length coding It has a backward decoding means for decoding the data from the backward direction and a fixed length code decoding means for decoding a fixed length code, and decodes the variable length coded data by the forward decoding means and the backward decoding means. When the first decoded value table When a specific variable-length code that can be decoded in the forward and backward directions indicating the start and end is decoded using, the second decoded value table is referred to, and the forward direction indicating the start is indicated. A variable length decoding device characterized in that a code following a specific variable length code that can be decoded in the opposite direction is regarded as a fixed length code and is decoded by the fixed length code decoding means. 8. The variable length decoding device according to claim 5, wherein the specific code or the specific variable length code is an escape code.