Arithmetic coding and decoding method based on semantic information source, and related device

Abstract

Provided in the present disclosure is an arithmetic coding and decoding method based on a semantic information source. The method comprises: acquiring a preset coding interval of a semantic information source and a syntax symbol sequence to be coded, wherein said syntax symbol sequence includes at least one syntax symbol; for each syntax symbol, determining a pre-constructed target synonymous subset corresponding to the syntax symbol; using an arithmetic coding algorithm and the preset coding interval to code the target synonymous subset corresponding to each syntax symbol, so as to obtain a coding result sequence corresponding to said syntax symbol sequence; acquiring the sequence length of said syntax symbol sequence; and sending the sequence length and the coding result sequence to a decoder, such that the decoder decodes the coding result sequence on the basis of the sequence length, so as to obtain a reconstructed syntax symbol sequence. Further provided in the present disclosure is a device related to arithmetic coding and decoding based on a semantic information source.

Description

Arithmetic encoding and decoding method based on semantic information source and related equipment Technical Field

[0001] The present disclosure relates to the field of coding and decoding, and in particular to an arithmetic coding and decoding method based on a semantic information source and related devices. Background Art

[0002] Source coding is a crucial component in ensuring the effectiveness of communication systems. It compresses source data using lossless or limited-loss source coding methods. Arithmetic coding is a common and important method in lossless source coding. When the code sequence is sufficiently long, it can achieve compression efficiency approaching the limits of lossless source compression as described in classic Shannon information theory. Existing arithmetic coding methods, when only semantic information is preserved, suffer from low compression efficiency. Therefore, improving compression efficiency while preserving semantic information has become a crucial research issue.

[0003] Summary of the Invention

[0004] In view of this, the embodiments of the present disclosure propose an arithmetic coding method and related devices based on a semantic source, which can improve compression efficiency while ensuring that semantic information is not distorted.

[0005] An arithmetic coding method based on a semantic information source provided by an embodiment of the present disclosure is applied to an encoder, comprising: obtaining a preset coding interval of a semantic information source and a sequence of grammatical symbols to be encoded; wherein the sequence of grammatical symbols to be encoded includes at least one grammatical symbol; for each grammatical symbol, determining a pre-constructed target synonymous subset corresponding to each grammatical symbol; using an arithmetic coding algorithm and the preset coding interval, encoding the target synonymous subset corresponding to each grammatical symbol to obtain a coding result sequence corresponding to the sequence of grammatical symbols to be encoded; obtaining a sequence length of the sequence of grammatical symbols to be encoded, and sending the sequence length and the coding result sequence to a decoder, so that the decoder decodes the coding result sequence based on the sequence length to obtain a reconstructed grammatical symbol sequence.

[0006] An arithmetic coding method based on a semantic information source provided in an embodiment of the present disclosure is applied to a decoder, comprising: receiving a sequence length and an encoding result sequence sent by an encoder, obtaining a preset decoding interval and preset syntax symbols of the semantic information source; decoding the encoding result sequence using an arithmetic decoding algorithm and the preset decoding interval to obtain at least one reconstructed synonymous subset, wherein the number of the reconstructed synonymous subsets is the same as the sequence length; determining, for each reconstructed synonymous subset, semantic information corresponding to the reconstructed synonymous subset, and selecting, from the preset syntax symbols, a target reconstructed syntax symbol corresponding to the reconstructed synonymous subset based on the semantic information; and arranging the target reconstructed syntax symbols corresponding to each reconstructed synonymous subset to obtain a reconstructed syntax symbol sequence.

[0007] The semantic source-based arithmetic coding and decoding device provided in the embodiment of the present disclosure is provided in an encoder, and includes:

[0008] A data acquisition module is configured to acquire a preset coding interval and a sequence of grammatical symbols to be encoded of a semantic information source; wherein the sequence of grammatical symbols to be encoded includes at least one grammatical symbol;

[0009] a target synonym subset determination module configured to determine, for each grammatical symbol, a pre-built target synonym subset corresponding to each grammatical symbol;

[0010] The coding result sequence determination module is configured to use an arithmetic coding algorithm and the preset coding interval to perform coding processing on the target synonymous subset corresponding to each grammatical symbol to obtain a coding result sequence corresponding to the grammatical symbol sequence to be encoded;

[0011] The data sending module is configured to obtain the sequence length of the grammatical symbol sequence to be encoded, and send the sequence length and the encoding result sequence to the decoder, so that the decoder decodes the encoding result sequence based on the sequence length to obtain a reconstructed grammatical symbol sequence.

[0012] The semantic source-based arithmetic coding and decoding device provided in the embodiments of the present disclosure is provided in a decoder and includes:

[0013] The coding result sequence receiving module is configured to receive the sequence length and coding result sequence sent by the encoder, and obtain the preset decoding interval and preset grammatical symbols of the semantic information source;

[0014] a reconstructed synonymous subset determination module configured to decode the encoded result sequence using an arithmetic decoding algorithm and the preset decoding interval to obtain at least one reconstructed synonymous subset, wherein the number of the reconstructed synonymous subsets is the same as the length of the sequence;

[0015] a target reconstruction grammatical symbol determination module configured to determine, for each reconstructed synonymous subset, semantic information corresponding to the reconstructed synonymous subset, and select, from the preset grammatical symbols, a target reconstruction grammatical symbol corresponding to the reconstructed synonymous subset based on the semantic information;

[0016] The reconstructed grammatical symbol sequence determination module is configured to arrange the target reconstructed grammatical symbols corresponding to each reconstructed synonymous subset to obtain a reconstructed grammatical symbol sequence.

[0017] Based on the same inventive concept, an embodiment of the present disclosure also proposes an electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable by the processor, wherein the processor implements the above-described semantic source-based arithmetic coding method when executing the computer program.

[0018] Based on the same inventive concept, an embodiment of the present disclosure further proposes a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute the arithmetic coding method based on semantic information source as described above.

[0019] Based on the same inventive concept, an embodiment of the present disclosure further proposes a computer program product, including computer program instructions. When the computer program instructions are executed on a computer, the computer is caused to execute the arithmetic coding method based on semantic information source as described above.

[0020] As can be seen from the foregoing, the semantic source-based arithmetic coding method and related devices proposed in this disclosure can obtain a preset coding interval of the semantic source and a sequence of grammatical symbols to be encoded, wherein the sequence of grammatical symbols to be encoded includes at least one grammatical symbol. For each grammatical symbol, a pre-constructed target synonymous subset corresponding to each grammatical symbol is determined for subsequent encoding of the pre-constructed target synonymous subset corresponding to each grammatical symbol. Using an arithmetic coding algorithm and the preset coding interval, the target synonymous subset corresponding to each grammatical symbol is encoded to obtain a coded result sequence corresponding to the sequence of grammatical symbols to be encoded. This achieves compression of semantic information during the source compression process, while also improving data compression efficiency using the arithmetic coding algorithm. The sequence length of the sequence of grammatical symbols to be encoded is obtained, and the sequence length and the coded result sequence are sent to a decoder, which decodes the coded result sequence based on the sequence length to obtain a reconstructed grammatical symbol sequence. The semantic source-based arithmetic coding method and related devices described above ensure that the reconstructed grammatical symbol sequence contains the same semantic information as the original sequence to be encoded, achieving lossless transmission of semantic information. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] FIG1A is a flowchart of an arithmetic coding and decoding method based on a semantic information source according to an embodiment of the present disclosure;

[0023] FIG1B is a flowchart of a target coding interval selection process according to an embodiment of the present disclosure;

[0024] FIG2 is a flowchart of an arithmetic coding and decoding method based on a semantic information source according to another embodiment of the present disclosure;

[0025] FIG3A is a schematic diagram of dividing synonymous subsets according to another embodiment of the present disclosure;

[0026] FIG3B is a schematic diagram of an original edge texture and a reconstructed edge texture according to another embodiment of the present disclosure;

[0027] FIG4 is a structural block diagram of an arithmetic coding and decoding apparatus based on a semantic information source according to an embodiment of the present disclosure;

[0028] FIG5 is a structural block diagram of an arithmetic coding and decoding apparatus based on a semantic information source according to another embodiment of the present disclosure;

[0029] FIG6 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0030] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0031] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should have the usual meanings understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the described object changes, the relative position relationship may also change accordingly.

[0032] Based on the above description, this embodiment proposes an arithmetic coding and decoding method based on semantic information sources applied to an encoder. As shown in FIG1A , the arithmetic coding and decoding method based on semantic information sources applied to an encoder includes the following steps.

[0033] Step 101: Obtain a preset coding interval of a semantic information source and a sequence of grammatical symbols to be encoded.

[0034] In an embodiment of the present disclosure, the above-mentioned grammar symbol sequence to be encoded includes at least one grammar symbol.

[0035] In a specific implementation, the coding interval of the semantic information source is initialized and the initialized coding interval is used as a preset coding interval. The preset coding interval can be regarded as a set of coding results of any information source sequence contained in the semantic information source. A grammatical symbol sequence to be encoded is obtained, wherein the grammatical symbol sequence to be encoded includes at least one grammatical symbol.

[0036] Exemplarily, the preset coding interval is [0, 1), and any source sequence contained in the semantic source can be mapped to a decimal interval in [0, 1).

[0037] Step 102: for each grammatical symbol, determine a pre-built target synonym subset corresponding to each grammatical symbol.

[0038] In a specific implementation, each grammatical symbol is a grammatical symbol included in a semantic information source. A corresponding synonymous subset may be pre-determined for each grammatical symbol, and the relationship between the grammatical symbol and the synonymous subset may be stored in association. For each grammatical symbol, a target synonymous subset corresponding to each grammatical symbol is determined from the associated stored grammatical symbols and synonymous subsets.

[0039] Step 103 : Using an arithmetic coding algorithm and the preset coding interval, encoding is performed on the target synonymous subset corresponding to each grammatical symbol to obtain a coding result sequence corresponding to the grammatical symbol sequence to be encoded.

[0040] In specific implementation, arithmetic coding is a commonly used and important coding method in lossless source coding. Its basic idea is to divide the set coding interval into sub-intervals according to the probability ratio of the symbols to be encoded, and select a specific sub-interval as the new coding interval based on the current symbol to be encoded. In this way, based on the recursive division of sub-intervals, the sequence to be encoded is uniquely mapped to a sub-interval, and the shortest binary sequence that can represent any numerical value in this sub-interval is used as the compression result.

[0041] The target synonymous subset corresponding to each grammatical symbol is coded using an arithmetic coding algorithm and the preset coding interval to obtain a coding result sequence.

[0042] Step 104: Obtain the sequence length of the grammatical symbol sequence to be encoded.

[0043] Step 105: Send the sequence length and the encoding result sequence to a decoder, so that the decoder performs decoding processing on the encoding result sequence based on the sequence length to obtain a reconstructed syntax symbol sequence.

[0044] In a specific implementation, the to-be-encoded grammar symbol sequence is traversed to obtain a sequence length of the to-be-encoded grammar symbol sequence, and the sequence length and the encoding result sequence are sent to a decoder. The decoder decodes the encoding result sequence to obtain a reconstructed grammar symbol sequence having a length equal to the sequence length. The semantic information of the reconstructed grammar symbol sequence is the same as the semantic information of the to-be-encoded grammar symbol sequence.

[0045] Through the above scheme, a preset coding interval of the semantic information source and a sequence of grammatical symbols to be encoded are obtained, wherein the sequence of grammatical symbols to be encoded contains at least one grammatical symbol. For each grammatical symbol, a pre-constructed target synonymous subset corresponding to each grammatical symbol is determined for subsequent encoding of the pre-constructed target synonymous subset corresponding to each grammatical symbol. Using an arithmetic coding algorithm and the preset coding interval, the target synonymous subset corresponding to each grammatical symbol is encoded to obtain a coding result sequence corresponding to the sequence of grammatical symbols to be encoded, thereby achieving compression of semantic information during the information source compression process, and at the same time, using an arithmetic coding algorithm to improve data compression efficiency. The coding result sequence is sent to a decoder for the decoder to decode the coding result sequence to obtain a reconstructed grammatical symbol sequence, thereby ensuring that the semantic information of the reconstructed grammatical symbol sequence is the same as that of the original sequence to be encoded, thereby achieving distortion-free transmission of semantic information.

[0046] In some embodiments, step 102 specifically includes: obtaining target symbol information of the grammatical symbol, and selecting an initial synonymous subset corresponding to the target symbol information from all pre-constructed initial synonymous subsets associated with the grammatical symbol as the target synonymous subset corresponding to the grammatical symbol.

[0047] In a specific implementation, each grammatical symbol includes target symbol information, and each grammatical symbol is associated with at least one initial synonymous subset. For each grammatical symbol, an initial synonymous subset corresponding to the target symbol information is determined from all associated initial synonymous subsets based on the target symbol information, and is used as the target synonymous subset corresponding to the grammatical symbol.

[0048] The construction process of the initial synonymous subset specifically includes: first, for each grammatical symbol, obtaining multiple preset symbol information corresponding to the grammatical symbol and semantic information corresponding to each preset symbol information; then, dividing the multiple preset symbol information according to the semantic information, and dividing all the preset symbol information corresponding to the same semantic information into the same set to obtain at least one initial synonymous subset; wherein the multiple preset symbol information includes the target symbol information.

[0049] In specific implementation, each grammatical symbol corresponds to multiple preset symbol information, each preset symbol information corresponds to semantic information, the semantic information corresponding to any two preset symbol information may be the same or different, and the multiple preset symbol information corresponding to any two grammatical symbol information may be the same or different.

[0050] For each grammatical symbol, all preset symbol information corresponding to the grammatical symbol is divided according to the semantic information, and the preset grammatical symbols with the same semantic information are divided into the same set to obtain at least one initial synonymous subset.

[0051] The plurality of preset grammatical symbol information include the target symbol information, and the initial synonymous subset corresponding to the target symbol information is used as the target synonymous subset.

[0052] For example, assuming that the grammatical symbol is u1, the preset symbol information corresponding to u1 is 1, 2, 3, 4, 5, 6, and 7, and the target symbol information corresponding to u1 is 3, the preset symbol information is divided according to the semantic information, and the initial synonymous subsets corresponding to u1 are synonymous subset A, synonymous subset B, and synonymous subset C. The preset symbol information contained in synonymous subset A is 1 and 2, the preset symbol information contained in synonymous subset B is 3 and 4, and the preset symbol information contained in synonymous subset C is 5, 6, and 7. According to the target symbol information corresponding to u1, synonymous subset B is selected as the target synonymous subset.

[0053] In some embodiments, step 103 specifically includes the following process.

[0054] First, for each grammatical symbol, the probability value of each preset symbol information corresponding to each initial synonymous subset associated with the grammatical symbol is summed up to obtain the probability value of each initial synonymous subset.

[0055] Then, a first arrangement order of all grammatical symbols in the grammatical symbol sequence to be encoded is obtained, and an arithmetic coding algorithm is used to select a target coding interval from the preset coding interval based on the first arrangement order, the target synonymous subset corresponding to each grammatical symbol, and the probability values ​​of all initial synonymous subsets associated with each grammatical symbol.

[0056] Finally, the shortest binary sequence is determined in the target coding interval, and the shortest binary sequence is used as the coding result sequence corresponding to the grammatical symbol sequence to be encoded.

[0057] In a specific implementation, for any initial synonymous subset corresponding to each grammatical symbol, the calculation process of the probability value of the initial synonymous subset specifically includes:

[0058] The preset symbol information corresponding to the initial synonymous subset is obtained, a probability value of each preset symbol information is determined, and the quantity of the preset symbol information is obtained.

[0059] If the number of the preset symbol information is one, the probability value of the preset symbol information is used as the target probability value corresponding to the initial synonymous subset.

[0060] If there are multiple preset symbol information, the probability values ​​of each preset symbol information corresponding to the initial synonymous subset are summed up to obtain the target probability value corresponding to the initial synonymous subset.

[0061] For example, the initial synonymous subset A includes the preset symbol information of 1, 2, and 3, and the corresponding probability values ​​are 0.1, 0.2, and 0.1, respectively. Then, the target probability value corresponding to the initial synonymous subset A is 0.4.

[0062] In some embodiments, based on the consideration of correlation, the probability of the initial synonymous subset can also be calculated based on conditional probability, that is, for the grammatical symbols of other orders except the grammatical symbols of the first order in the first arrangement order, when calculating the probability value of the initial synonymous subset corresponding to the grammatical symbols of other orders, it is necessary to use the target synonymous subset corresponding to the previous one or more grammatical symbols that have been encoded as a condition to determine the probability value of the preset symbol information corresponding to the initial synonymous subset, and add up the probability values ​​of the preset symbol information corresponding to the initial synonymous subset to obtain the probability value of the initial synonymous subset.

[0063] Exemplarily, the semantic information source is a first-order Markov information source, and the calculation based on conditional probability means that for the grammatical symbols of other orders except the grammatical symbol of the first order in the first arrangement order, when calculating the probability values ​​of the initial synonymous subsets corresponding to the grammatical symbols of other orders, it is necessary to determine the target synonymous subset corresponding to the grammatical symbol of the previous order, and use the target synonymous subset corresponding to the grammatical symbol of the previous order as a condition to determine the probability value of the preset symbol information corresponding to the initial synonymous subset, and add the probability values ​​of the preset symbol information corresponding to the initial synonymous subset to obtain the probability value of the initial synonymous subset.

[0064] The probability value of the preset symbol information corresponding to the initial synonymous subset needs to consider the first-order Markov characteristic. At this time, the probability value of the initial synonymous subset is expressed by the formula:

[0065] Among them, U i-1 is the coded synonym subset, m is the number of grammatical symbols, i is the i-th rank in the first arrangement order, U i,k is the kth initial synonymous subset of the i-th order, N i,k is the set of serial numbers of all preset symbol information in the kth initial synonym subset corresponding to the grammatical symbol at the i-th order, u i,n is the probability value of the nth preset symbol information at the i-th position.

[0066] In another example, the semantic information source is a second-order Markov information source, and calculation based on conditional probability means that when calculating the probability value of the initial synonymous subset corresponding to the second-rank grammatical symbol in the first arrangement order, it is necessary to determine the target synonymous subset corresponding to the first-rank grammatical symbol, and use the target synonymous subset corresponding to the first-rank grammatical symbol as a condition to determine the probability value of the preset symbol information corresponding to the initial synonymous subset, and add the probability values ​​of the preset symbol information corresponding to the initial synonymous subset to obtain the probability value of the initial synonymous subset.

[0067] For grammatical symbols of other orders except the first and second orders in the first arrangement order, when calculating the probability values ​​of the initial synonymous subsets corresponding to the grammatical symbols of other orders, it is necessary to determine the target synonymous subsets corresponding to the first two orders of grammatical symbols, and use the target synonymous subsets corresponding to the first two orders of grammatical symbols as conditions to determine the probability value of the preset symbol information corresponding to the initial synonymous subset, and sum the probability values ​​of the preset symbol information corresponding to the initial synonymous subset to obtain the probability value of the initial synonymous subset.

[0068] The probability value of the preset symbol information corresponding to the initial synonymous subset needs to consider the second-order Markov characteristic. At this time, the probability value of the initial synonymous subset is expressed by the formula:

[0069] In some embodiments, the probabilities of the initial synonymous subsets can be transmitted to the decoder as header information. Alternatively, the codec can synchronize the initial synonymous subsets within the codec in the form of pre-agreed a priori information to prevent the probability information from occupying the length of the encoded sequence and reducing compression efficiency. Alternatively, the probabilities of the initial synonymous subsets can be obtained through adaptive probability estimation. In this case, there is no need to transmit the initial synonymous subset probabilities or synchronize them within the codec in the form of pre-agreed a priori information. Instead, the same adaptive probability estimation method needs to be used at both codecs.

[0070] In a specific implementation, the probability value of each preset symbol information included in each initial synonymous subset is obtained, wherein the probability value of each preset symbol information is determined in at least one of the following ways: by statistical method, by prior probability method, and by table lookup method.

[0071] Exemplarily, the probability value of each symbol information is pre-calculated by a table lookup method, the symbol information and the probability value are associated and stored in a database, and the storage format in the database is a table. According to the preset symbol information, the probability value corresponding to the preset symbol information is searched in the database.

[0072] A target probability value of the target synonymous subset is determined according to the probability value of each preset symbol information included in the target synonymous subset.

[0073] Obtain a first arrangement order of all grammatical symbols in a sequence of grammatical symbols to be encoded, and select a target coding interval from the preset coding interval using an arithmetic coding algorithm based on the first arrangement order, the target synonymous subset corresponding to each grammatical symbol, and the probability values ​​of all initial synonymous subsets associated with each grammatical symbol.

[0074] A shortest binary sequence is determined in the target coding interval, the shortest binary sequence being the shortest binary coding sequence that can represent the value in the target coding interval, and the shortest binary sequence is used as the coding result sequence. The codeword corresponding to the shortest binary sequence is within the target coding interval.

[0075] In some embodiments, if the preset coding interval is [0, 1), the shortest binary sequence is the binary decimal sequence corresponding to the codeword, which satisfies the following relationship: c = b1·2 -1 +b2·2 -2 +…b l 2 -l ; Where b is the shortest binary sequence, where b=(b1,b2,…,b l ), c is the codeword, and l is the length of the shortest binary sequence.

[0076] In some embodiments, the step of obtaining a first arrangement order of all grammatical symbols in the sequence of grammatical symbols to be encoded, and selecting a target coding interval from the preset coding interval using an arithmetic coding algorithm based on the first arrangement order, the target synonymous subset corresponding to each grammatical symbol, and the probability values ​​of all initial synonymous subsets associated with each grammatical symbol specifically includes: using the preset coding interval as a candidate coding interval, using the first rank in the first arrangement order as the target rank, and performing at least one round of iterative operations on the preset coding interval, wherein each round of iterative operations is performed as follows:

[0077] In the candidate coding intervals, the candidate coding intervals are divided according to the probability values ​​of all initial synonymous subsets corresponding to the grammatical symbols of the target sequence, to obtain at least one initial coding interval corresponding to each initial synonymous subset;

[0078] Taking an initial coding interval corresponding to a target synonymous subset corresponding to a grammatical symbol of a target priority in at least one initial coding interval as an updated candidate coding interval, and taking the next priority of the target priority as the target priority of the next iteration according to the first arrangement order; until there is no next priority of the target priority according to the first arrangement order, then exiting the iteration operation;

[0079] The candidate coding interval after at least one round of iterative operation is used as the target coding interval.

[0080] In a specific implementation, the preset coding interval is used as a candidate coding interval, the first rank in the first arrangement sequence is used as a target rank, and at least one round of iterative operation is performed on the preset coding interval. Each round of iterative operation is performed as follows:

[0081] Among the candidate coding intervals, a corresponding interval is selected based on the target synonym subset corresponding to the grammatical symbol at the target order as the updated candidate coding interval. The next interval after the target order in the first arrangement sequence is used as the target order for the next iteration. The iteration operation is terminated until no next interval exists for the target order according to the first arrangement sequence. The candidate coding interval after at least one round of iteration is used as the target coding interval.

[0082] Specifically, the process of determining the updated candidate coding interval is as follows:

[0083] Obtain probability values ​​for all initial synonymous subsets corresponding to the grammatical symbols of the target sequence, where the sum of the probabilities of all initial synonymous subsets is 1. Use the probability values ​​to divide the candidate coding intervals to obtain at least one initial coding interval corresponding to each initial synonymous subset, each initial coding interval corresponding to a probability value of the initial synonymous subset. Select, from the at least one initial coding interval, the initial coding interval corresponding to the target synonymous subset corresponding to the grammatical symbols of the target sequence as an updated candidate coding interval.

[0084] For example, the grammatical symbol of the target sequence is u1, and all initial synonymous subsets corresponding to u1 are obtained as synonymous subset A, synonymous subset B, and synonymous subset C. The probability value of synonymous subset A is 0.2, the probability value of synonymous subset B is 0.3, and the probability value of synonymous subset C is 0.5. Based on the probability values ​​of all initial synonymous subsets corresponding to the grammatical symbol of the target sequence, the candidate coding interval is divided, and the initial coding interval corresponding to synonymous subset A accounts for 0.2 of the candidate coding interval, the initial coding interval corresponding to synonymous subset B accounts for 0.3 of the candidate coding interval, and the initial coding interval corresponding to synonymous subset C accounts for 0.5 of the candidate coding interval. If the target synonymous subset corresponding to the grammatical symbol of the target sequence is B, the initial coding interval corresponding to synonymous subset B is used as the updated candidate coding interval.

[0085] Exemplarily, the preset coding interval is [L0, H0), the first arrangement order is (u1, u2, u3), u1 is taken as the target order, and the corresponding interval in [L0, H0) is selected as the updated candidate coding interval [L1, H1) according to the target synonymous subset corresponding to u1. u2 is taken as the target order, and the corresponding interval in [L1, H1) is selected as the updated candidate coding interval [L2, H2) according to the target synonymous subset corresponding to u2. u3 is taken as the target order, and the corresponding interval in [L2, H2) is selected as the updated candidate coding interval [L3, H3) according to the target synonymous subset corresponding to u3. If there is no next order for the target order, the iterative operation is exited, and [L3, H3) is taken as the target coding interval.

[0086] In another example, the preset coding interval is [L0, H0), and the target coding interval is [L m ,H m ), b is the shortest binary sequence, u i is the grammatical symbol at the i-th position, U i The target synonymous subset corresponding to the grammatical symbol at the i-th order is determined, and a target coding interval is selected from the preset coding interval according to the target synonymous subset at each order using an arithmetic coding algorithm. The schematic diagram of the selection process is shown in FIG1B .

[0087] In some embodiments, after the updated candidate coding interval is determined, the updated candidate coding interval may be amplified to ensure calculation accuracy.

[0088] The method of the amplification process is to set the updated candidate coding interval as [L i ,H i ), select a value s to satisfy L i -ε<s≤L i and s<H i , and [Li ,H i ) is updated to [α(L i -s),α(H i -s)), where ε is a smaller value, indicating that the value of s should be close to L i α is the magnification factor, and its value range should be greater than or equal to 1 and can make the updated interval meet the specific value of the calculation accuracy.

[0089] In some embodiments, step 103 may specifically include the following process.

[0090] First, the arrangement order of all grammatical symbols in the grammatical symbol sequence to be encoded is obtained.

[0091] Then, an arithmetic coding algorithm and the preset coding interval are used to sequentially encode the synonymous subsets corresponding to each grammatical symbol according to the arrangement order to obtain a coding subsequence corresponding to each grammatical symbol.

[0092] Finally, all the coding subsequences are integrated according to the arrangement order to obtain a coding result sequence corresponding to the grammatical symbol sequence to be encoded.

[0093] In specific implementation, the encoding result sequence does not need to be converted after the encoding is completed, that is, part of the binary subsequences can be gradually output during the encoding process, so that after the encoding process is completed, all the binary subsequences are merged as the complete encoding result.

[0094] In some embodiments, before step 103, the method further includes: determining whether the to-be-encoded grammar symbol sequence satisfies a first preset condition. The first preset condition includes: multiple preset symbol information corresponding to any two grammar symbols in the to-be-encoded grammar symbol sequence are identical, initial synonymous subsets corresponding to any two grammar symbols are identical, and probability values ​​of the initial synonymous subsets corresponding to any two grammar symbols are identical.

[0095] If it is determined that the to-be-encoded syntax symbol sequence meets the first preset condition, step 103 specifically includes:

[0096] First, a target synonym subset sequence is constructed according to the target synonym subset corresponding to each grammatical symbol.

[0097] Then, the target synonymous subset sequence is coded using an arithmetic coding algorithm and the preset coding interval to obtain a coding result sequence corresponding to the grammatical symbol sequence to be coded.

[0098] In a specific implementation, the encoder may also directly encode the synonymous subset sequence. When the grammatical symbol sequence to be encoded meets the first preset condition, a target synonymous subset sequence is constructed based on the target synonymous subset corresponding to each grammatical symbol. The target synonymous subset sequence is encoded using an arithmetic coding algorithm and the preset coding interval to obtain an encoding result sequence corresponding to the grammatical symbol sequence to be encoded.

[0099] Exemplarily, a sequence of grammatical symbols to be encoded includes grammatical symbol A and grammatical symbol B. The preset symbol information corresponding to grammatical symbol A is 1, 2, and 3, and the preset symbol information corresponding to grammatical symbol B is 1, 2, and 3. In this case, the multiple preset symbol information corresponding to any two grammatical symbols is the same. The initial synonymous subset a corresponding to grammatical symbol A contains preset symbol information of 1 and 3, and the initial synonymous subset b corresponding to grammatical symbol A contains preset symbol information of 2. The initial synonymous subset a corresponding to grammatical symbol B contains preset symbol information of 1 and 3, and the initial synonymous subset b corresponding to grammatical symbol B contains preset symbol information of 2. In this case, the initial synonymous subsets corresponding to any two grammatical symbols are the same. The probability value corresponding to the initial synonymous subset a corresponding to grammatical symbol A is 0.3, the probability value corresponding to the initial synonymous subset b corresponding to grammatical symbol A is 0.2, the probability value corresponding to the initial synonymous subset a corresponding to grammatical symbol B is 0.3, and the probability value corresponding to the initial synonymous subset b corresponding to grammatical symbol B is 0.2. In this case, the probability values ​​of the initial synonymous subsets corresponding to any two grammatical symbols are the same.

[0100] Specifically, if the sequence of grammatical symbols to be encoded satisfies the first preset condition, a target synonymous subset corresponding to each grammatical symbol is determined to construct a target synonymous subset sequence. Specifically, the target synonymous subset corresponding to grammatical symbol A is a, and the target synonymous subset corresponding to grammatical symbol B is b. The target synonymous subset sequence is then constructed. The target synonymous subset sequence is encoded using an arithmetic coding algorithm and the preset coding interval to obtain a coding result sequence corresponding to the sequence of grammatical symbols to be encoded.

[0101] Through the above scheme, for the grammatical symbol sequence to be encoded, under the conditions that the multiple preset symbol information corresponding to any two grammatical symbols in the grammatical symbol sequence to be encoded is the same, the initial synonymous subsets corresponding to any two grammatical symbols are the same, and the probability values ​​of the initial synonymous subsets corresponding to any two grammatical symbols are the same, and under the condition that the reconstructed grammatical symbol sequence and the grammatical symbol sequence at the encoding end meet the semantic distortion-free condition, higher compression efficiency is achieved compared to arithmetic coding based on Shannon information theory, and the implementation difficulty of the semantic source arithmetic coding and decoding algorithm is further simplified.

[0102] Another embodiment of the present disclosure proposes an arithmetic coding method based on a semantic information source, as shown in FIG2 , which is applied to a decoder. The method includes:

[0103] Step 201: Receive the sequence length and encoding result sequence sent by the encoder, and obtain the preset decoding interval and preset grammatical symbols of the semantic information source.

[0104] In specific implementations, the sequence length (i.e., the encoded sequence) sent by the encoder is received, and the predetermined syntax symbols and predetermined decoding intervals contained in the semantic information source are determined. This corresponds to the initial encoding interval during the encoding process, which can be considered a set of codewords corresponding to any binary sequence to be decoded.

[0105] Step 202: Decode the encoding result sequence using an arithmetic decoding algorithm and the preset decoding interval to obtain at least one reconstructed synonymous subset.

[0106] In an embodiment of the present disclosure, the number of the reconstructed synonymous subsets is the same as the length of the sequence.

[0107] In a specific implementation, based on an arithmetic decoding algorithm and a preset decoding interval, the received coding result sequence is decoded to obtain at least one reconstructed synonymous subset corresponding to the coding result sequence. The number of the reconstructed synonymous subsets is the same as the sequence length, and the semantic information corresponding to the reconstructed synonymous subset is the same as the semantic information corresponding to the target synonymous subset during encoding.

[0108] Step 203: for each reconstructed synonymous subset, determine the semantic information corresponding to the reconstructed synonymous subset.

[0109] Step 204 : selecting a target reconstructed grammatical symbol corresponding to the reconstructed synonymous subset from the preset grammatical symbols according to the semantic information.

[0110] In a specific implementation, for each reconstructed synonymous subset, semantic information corresponding to the reconstructed synonymous subset is determined, and at least one initial reconstructed grammatical symbol corresponding to the reconstructed synonymous subset is selected from the preset grammatical symbols according to the semantic information.

[0111] The number of the initial reconstructed grammatical symbols is obtained. If the number of the initial reconstructed grammatical symbols is one, the initial reconstructed grammatical symbol is used as the target reconstructed grammatical symbol. If the number of the initial reconstructed grammatical symbols is multiple, one of the initial reconstructed grammatical symbols is randomly selected or selected according to a certain rule as the target reconstructed grammatical symbol. The selecting of an initial reconstructed grammatical symbol as the target reconstructed grammatical symbol according to a certain rule includes determining the target reconstructed grammatical symbol by a preset prior probability value, or determining the target reconstructed grammatical symbol based on relevant background knowledge of semantic information corresponding to the reconstructed synonymous subset, or determining the target reconstructed grammatical symbol by a trained neural network model, wherein the input of the neural network model is multiple reconstructed synonymous subsets and the output is the target reconstructed grammatical symbol.

[0112] Step 205 : Arrange the target reconstructed grammatical symbols corresponding to each reconstructed synonymous subset to obtain a reconstructed grammatical symbol sequence.

[0113] During specific implementation, the arrangement order of the reconstructed synonymous subsets is obtained, the target reconstructed grammatical symbol corresponding to each reconstructed synonymous subset is determined, and the target reconstructed grammatical symbol corresponding to each reconstructed synonymous subset is arranged according to the arrangement order of the reconstructed synonymous subsets to obtain a reconstructed grammatical symbol sequence.

[0114] In some embodiments, step 202 specifically includes:

[0115] First, the encoding result sequence is converted to obtain a target codeword.

[0116] During specific implementation, in a preset decoding interval, the coding result sequence corresponds to the codeword one-to-one, so the coding result sequence is converted to obtain the target codeword.

[0117] In some embodiments, when the preset decoding interval is [0, 1), the corresponding codeword can be calculated according to the relationship between the encoding result sequence and the codeword.

[0118] The relationship is expressed by the formula: c = b1·2 -1 +b2·2 -2 +…b l 2 -l .

[0119] Then, a reconstructed symbol sequence is constructed according to the sequence length, and the number of elements in the reconstructed symbol sequence is the same as the sequence length.

[0120] Next, a second arrangement order of elements in the reconstructed symbol sequence is determined.

[0121] Furthermore, for each element in the reconstructed symbol sequence, at least one initial reconstructed synonymous subset corresponding to the element and a probability value of each initial reconstructed synonymous subset are obtained.

[0122] Furthermore, according to the second arrangement order, the probability value of each initial reconstructed synonymous subset corresponding to each element and the target codeword, an arithmetic decoding algorithm is used to determine the target decoding interval corresponding to each element from the preset decoding interval.

[0123] Finally, for each element, a reconstructed synonymous subset corresponding to the element is determined according to the target decoding interval and all initially reconstructed synonymous subsets.

[0124] In a specific implementation, a reconstructed symbol sequence is constructed based on the sequence length, the number of elements in the reconstructed symbol sequence is the same as the sequence length, a second arrangement order of the elements in the reconstructed symbol sequence is determined, and for each element in the reconstructed symbol sequence, at least one initial reconstructed synonymous subset corresponding to the element and a probability value of each initial reconstructed synonymous subset are obtained.

[0125] The at least one initially reconstructed synonymous subset corresponding to the element and the probability value of each initially reconstructed synonymous subset may be sent from the encoder to the decoder, or may be obtained from pre-agreed prior information, or may be obtained by using an adaptive probability estimation method.

[0126] Based on the second permutation order, the probability values ​​of each initially reconstructed synonymous subset corresponding to each element, and the target codeword, an arithmetic decoding algorithm is used to determine a target decoding interval corresponding to each element from the preset decoding interval. For each element, a reconstructed synonymous subset corresponding to the element is determined based on the target decoding interval and all initially reconstructed synonymous subsets.

[0127] In some embodiments, the step of determining the target decoding interval corresponding to each element from the preset decoding interval using an arithmetic decoding algorithm based on the second arrangement order, the probability value of each initially reconstructed synonymous subset corresponding to each element, and the target codeword may specifically include: using the preset decoding interval as a candidate decoding interval, using the first rank in the second arrangement order as the target rank, and performing at least one round of iterative operations on the preset decoding interval, wherein each round of iterative operations is performed as follows:

[0128] Dividing the candidate decoding intervals according to the probability values ​​of each initial reconstructed synonymous subset corresponding to the elements of the target sequence to obtain at least one initial decoding interval, wherein the initial decoding intervals correspond one-to-one to the probability values ​​of the initial reconstructed synonymous subsets;

[0129] Selecting an initial decoding interval containing the target codeword from a plurality of initial decoding intervals as a target decoding interval corresponding to an element of a target sequence, using the target decoding interval as an updated candidate decoding interval, and using the next sequence of the target sequence as a target sequence for a next iteration according to the second arrangement order; until no next sequence of the target sequence exists according to the second arrangement order, then exiting the iteration operation;

[0130] Get the target decoding interval corresponding to each element.

[0131] In a specific implementation, the preset decoding interval is used as a candidate decoding interval, the first rank in the second arrangement order is used as a target rank, and at least one round of iterative operation is performed on the preset decoding interval. Each round of iterative operation is performed as follows:

[0132] The candidate decoding intervals are divided according to the probability value of each initial reconstructed synonymous subset corresponding to the elements of the target sequence to obtain at least one initial decoding interval, each initial decoding interval having a one-to-one correspondence with the probability value of the initial reconstructed synonymous subset, and thus the initial decoding intervals have a one-to-one correspondence with the initial reconstructed synonymous subset.

[0133] An initial decoding interval containing the target codeword is selected from multiple initial decoding intervals as the target decoding interval corresponding to the element of the target sequence, the target decoding interval is used as the updated candidate decoding interval, and the next sequence of the target sequence is used as the target sequence of the next iteration according to the second arrangement order. Until there is no next sequence of the target sequence according to the second arrangement order, the iterative operation is exited to obtain the target decoding interval corresponding to each element.

[0134] Exemplarily, the initial decoding interval is [L0, H0), the second arrangement order is (u1, u2), and u1 is taken as the target sequence. Each initial reconstructed synonymous subset corresponding to u1 is obtained as reconstructed synonymous subset A, reconstructed synonymous subset B, and reconstructed synonymous subset C. The probability value of each initial reconstructed synonymous subset corresponding to u1 is obtained, that is, the probability value of reconstructed synonymous subset A is 0.2, the probability value of reconstructed synonymous subset B is 0.6, and the probability value of reconstructed synonymous subset C is 0.2. [L0, H0) is divided to obtain three initial decoding intervals. The initial decoding interval containing the target codeword is determined to be the initial decoding interval corresponding to the reconstructed synonymous subset B. The initial decoding interval corresponding to the reconstructed synonymous subset B is used as the updated candidate decoding interval [L1, H1).

[0135] Taking u2 as the target sequence, obtain each initial reconstructed synonymous subset corresponding to u2 as reconstructed synonymous subset D and reconstructed synonymous subset E. Obtain a probability value for each initial reconstructed synonymous subset corresponding to u2, i.e., the probability value of reconstructed synonymous subset D is 0.1, and the probability value of reconstructed synonymous subset E is 0.9. Then, partition [L1, H1) to obtain two initial decoding intervals. The initial decoding interval containing the target codeword is determined to be the initial decoding interval corresponding to reconstructed synonymous subset E. The initial decoding interval corresponding to reconstructed synonymous subset E is used as the updated candidate decoding interval [L2, H2]. If there is no next sequence for the target sequence, exit the iteration and use [L2, H2) as the target decoding interval.

[0136] In some embodiments, before step 102, the method further includes: determining whether the preset grammar symbol satisfies a second preset condition. The second preset condition is that the plurality of preset symbol information corresponding to any two preset grammar symbols are identical, the division method of the initial synonymous subsets corresponding to any two preset grammar symbols is identical, and the third probability values ​​of the initial synonymous subsets corresponding to any two preset grammar symbols are identical.

[0137] If it is determined that the preset syntax symbol meets the second preset condition, step 202 includes: using an arithmetic decoding algorithm and the preset decoding interval to decode the encoded result sequence to obtain a reconstructed synonymous subset sequence, wherein the reconstructed synonymous subset sequence includes multiple reconstructed synonymous subsets.

[0138] In a specific implementation, if the multiple preset symbol information corresponding to any two of the preset grammar symbols is identical, the division method of the initial synonymous subsets corresponding to any two of the preset grammar symbols is identical, and the third probability values ​​of the initial synonymous subsets corresponding to any two of the preset grammar symbols are identical, then the encoding result sequence can be directly decoded using the arithmetic decoding algorithm and the preset decoding interval to obtain a reconstructed synonymous subset sequence.

[0139] In some embodiments, before step 205C, the method further includes:

[0140] Normalization is performed on each candidate decoding interval and the target codeword. The normalized candidate decoding intervals are then divided according to the probability value of each initial reconstructed synonymous subset corresponding to the element of the target sequence to obtain at least one initial decoding interval. An initial decoding interval containing the normalized target codeword is selected from the multiple initial decoding intervals as the target decoding interval corresponding to the element of the target sequence.

[0141] Through the above solution, by normalizing the candidate decoding intervals, problems such as decoding errors caused by excessive recursive depth of the candidate decoding intervals and insufficient calculation accuracy are avoided.

[0142] Based on the same inventive concept, another embodiment of the present disclosure shows an arithmetic coding result of a semantic information source for an edge texture image information source of a natural image, specifically including:

[0143] The edge texture maps of natural images used as semantic information sources come from the public dataset BIPEDv2d. Each edge texture map is a 01 black and white image with a resolution of 720*1280, where black represents the non-edge texture part of the image (objects and background) with a value of 0, and white represents the edge texture part of the image (boundaries between objects and objects, and between objects and background) with a value of 1.

[0144] In order to perform lossless compression of semantic information sources, this embodiment uses non-overlapping 2*2 image blocks as 1 syntax symbol, which corresponds to 16 preset symbol information with a value range of , where each preset symbol information corresponds to a 2*2 image block type. During the encoding process, each syntax symbol satisfies the conditions of having the same possible value range, synonymous mapping method, and the same synonymous subset probability. The synonymous subsets obtained according to the semantic information are shown in Figure 3A. According to the texture form (i.e., semantic information), these 16 preset symbol information can be divided into 11 synonymous subsets, where the synonymous subset represents the texture in the upper right (lower left) direction and contains 3 preset symbol information, the synonymous subset represents the texture in the upper left (lower right) direction and contains 4 preset symbol information, and the other synonymous subsets each contain 1 preset symbol information. During the decoding process, the synonymous rule uses a random selection method to randomly select a preset symbol information from the synonymous subset for output.

[0145] Figure 3B is a diagram showing the effect comparison and compression effect of the original edge texture and the reconstructed edge texture corresponding to the embodiment. In particular, the original edge texture and the reconstructed edge texture are both displayed with dark borders in the original image to demonstrate the semantic accuracy of the reconstructed edge texture.

[0146] As can be seen from the figure, through calculation, it can be seen that when standard arithmetic coding is used, the edge texture in this example can be compressed to 205990 bits, and when the semantic source arithmetic coding method disclosed in this application is adopted, the edge texture image in this example can be compressed to 204003 bits, which is a relative saving of 1987 bits.

[0147] It should be noted that the method of the embodiments of the present disclosure can be performed by a single device, such as a computer or server. The method of the embodiments of the present disclosure can also be applied in a distributed scenario, where multiple devices cooperate to perform the method. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiments of the present disclosure, and the multiple devices will interact with each other to complete the method.

[0148] It should be noted that the above description is limited to some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0149] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure further provides an arithmetic coding and decoding device based on a semantic information source.

[0150] 4 , which shows an arithmetic coding and decoding apparatus based on a semantic source according to an embodiment, and is provided in an encoder, including:

[0151] The data acquisition module 401 is configured to acquire a preset coding interval and a sequence of grammatical symbols to be encoded of a semantic information source, wherein the sequence of grammatical symbols to be encoded includes at least one grammatical symbol;

[0152] The target synonym subset determination module 402 is configured to determine, for each grammatical symbol, a pre-built target synonym subset corresponding to each grammatical symbol;

[0153] The coding result sequence determination module 403 is configured to use an arithmetic coding algorithm and the preset coding interval to perform coding processing on the target synonymous subset corresponding to each grammatical symbol to obtain a coding result sequence corresponding to the grammatical symbol sequence to be encoded;

[0154] The data sending module 404 is configured to obtain the sequence length of the grammatical symbol sequence to be encoded, and send the sequence length and the encoding result sequence to the decoder, so that the decoder decodes the encoding result sequence based on the sequence length to obtain a reconstructed grammatical symbol sequence.

[0155] In some embodiments, the target synonymous subset determination module 402 is configured to:

[0156] Obtain target symbol information of the grammatical symbol, and select an initial synonymous subset corresponding to the target symbol information from all pre-constructed initial synonymous subsets associated with the grammatical symbol as a target synonymous subset corresponding to the grammatical symbol;

[0157] The process of constructing the initial synonymous subset includes:

[0158] For each grammatical symbol, obtaining a plurality of preset symbol information corresponding to the grammatical symbol and semantic information corresponding to each preset symbol information;

[0159] The plurality of preset symbol information are divided according to the semantic information, and all preset symbol information corresponding to the same semantic information are divided into the same set to obtain at least one initial synonymous subset; wherein the plurality of preset symbol information include the target symbol information.

[0160] In some embodiments, the encoding result sequence determination module 403 specifically includes:

[0161] a probability value determining unit configured to, for each grammatical symbol, sum the probability values ​​of each preset symbol information corresponding to each initial synonymous subset associated with the grammatical symbol to obtain a probability value of each initial synonymous subset;

[0162] a target coding interval determining unit configured to obtain a first arrangement order of all grammatical symbols in the grammatical symbol sequence to be encoded, and select a target coding interval from the preset coding interval using an arithmetic coding algorithm based on the first arrangement order, a target synonymous subset corresponding to each grammatical symbol, and probability values ​​of all initial synonymous subsets associated with each grammatical symbol;

[0163] The encoding result sequence determining unit is configured to determine the shortest binary sequence in the target encoding interval, and use the shortest binary sequence as the encoding result sequence corresponding to the grammatical symbol sequence to be encoded.

[0164] In some embodiments, the target coding interval determination unit is specifically configured to:

[0165] The preset coding interval is used as a candidate coding interval, the first rank in the first arrangement sequence is used as a target rank, and at least one round of iterative operation is performed on the preset coding interval. Each round of iterative operation is performed as follows:

[0166] In the candidate coding intervals, the candidate coding intervals are divided according to the probability values ​​of all initial synonymous subsets corresponding to the grammatical symbols of the target sequence, to obtain at least one initial coding interval corresponding to each initial synonymous subset;

[0167] Taking the initial coding interval corresponding to the target synonymous subset corresponding to the grammatical symbol of the target sequence in at least one initial coding interval as the updated candidate coding interval, and taking the next sequence of the target sequence as the target sequence of the next iteration according to the first arrangement order;

[0168] Until there is no next rank for the target rank according to the first arrangement order, then exit the iterative operation;

[0169] The candidate coding interval after at least one round of iterative operation is used as the target coding interval.

[0170] In some embodiments, the device also includes a conditional judgment module, which is specifically configured to judge whether the grammatical symbol sequence to be encoded meets a first preset condition; the first preset condition includes: multiple preset symbol information corresponding to any two grammatical symbols in the grammatical symbol sequence to be encoded is the same, the initial synonymous subsets corresponding to any two grammatical symbols are the same, and the probability values ​​of the initial synonymous subsets corresponding to any two grammatical symbols are the same.

[0171] If it is determined that the to-be-encoded grammatical symbol sequence meets the first preset condition, the encoding result sequence determination module is specifically configured to:

[0172] Construct a target synonym subset sequence according to the target synonym subset corresponding to each grammatical symbol;

[0173] The target synonymous subset sequence is coded using an arithmetic coding algorithm and the preset coding interval to obtain a coding result sequence corresponding to the grammatical symbol sequence to be coded.

[0174] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure further provides an arithmetic coding and decoding device based on a semantic information source.

[0175] 5 , which shows an arithmetic encoding and decoding device based on a semantic source according to an embodiment, and is provided in a decoder, including:

[0176] The encoding result sequence receiving module 501 is configured to receive the sequence length and encoding result sequence sent by the encoder, and obtain the preset decoding interval and preset grammatical symbols of the semantic information source;

[0177] a reconstructed synonymous subset determining module 502 configured to decode the encoded result sequence using an arithmetic decoding algorithm and the preset decoding interval to obtain at least one reconstructed synonymous subset, wherein the number of the reconstructed synonymous subsets is the same as the length of the sequence;

[0178] The target reconstruction grammar symbol determination module 503 is configured to determine, for each reconstructed synonymous subset, semantic information corresponding to the reconstructed synonymous subset, and select, from the preset grammar symbols, a target reconstruction grammar symbol corresponding to the reconstructed synonymous subset according to the semantic information;

[0179] The reconstructed grammar symbol sequence determining module 504 is configured to arrange the target reconstructed grammar symbols corresponding to each reconstructed synonymous subset to obtain a reconstructed grammar symbol sequence.

[0180] In some embodiments, the reconstructed synonymous subset determination module 502 specifically includes:

[0181] a target codeword determining unit, configured to perform conversion processing on the encoding result sequence to obtain a target codeword;

[0182] a reconstructed symbol sequence construction unit, configured to construct a reconstructed symbol sequence according to the sequence length, wherein the number of elements in the reconstructed symbol sequence is the same as the sequence length;

[0183] an arrangement order determining unit, configured to determine a second arrangement order of elements in the reconstructed symbol sequence;

[0184] a probability value determining unit configured to obtain, for each element in the reconstructed symbol sequence, at least one initial reconstructed synonymous subset corresponding to the element and a probability value of each initial reconstructed synonymous subset;

[0185] a target decoding interval determining unit configured to determine a target decoding interval corresponding to each element from the preset decoding interval using an arithmetic decoding algorithm based on the second arrangement order, a probability value of each initially reconstructed synonymous subset corresponding to each element, and the target codeword;

[0186] The reconstructed synonymous subset determining unit is configured to determine, for each element, a reconstructed synonymous subset corresponding to the element according to the target decoding interval and all the initial reconstructed synonymous subsets.

[0187] In some embodiments, the target decoding interval determination unit is configured to:

[0188] The preset decoding interval is used as a candidate decoding interval, and the first rank in the second arrangement order is used as a target rank. At least one round of iterative operation is performed on the preset decoding interval, and each round of iterative operation is performed as follows:

[0189] Dividing the candidate decoding intervals according to the probability values ​​of each initial reconstructed synonymous subset corresponding to the elements of the target sequence to obtain at least one initial decoding interval, wherein the initial decoding intervals correspond one-to-one to the probability values ​​of the initial reconstructed synonymous subsets;

[0190] Selecting an initial decoding interval containing the target codeword from a plurality of initial decoding intervals as a target decoding interval corresponding to an element of a target sequence, using the target decoding interval as an updated candidate decoding interval, and using the next sequence of the target sequence as a target sequence for a next iteration according to the second arrangement order;

[0191] Until there is no next position to the target position according to the second arrangement order, then exit the iterative operation;

[0192] Get the target decoding interval corresponding to each element.

[0193] In some embodiments, the device also includes a conditional judgment module, which is configured to: judge whether the preset grammatical symbol meets a second preset condition; the second preset condition is that the multiple preset symbol information corresponding to any two preset grammatical symbols are the same, the division method of the initial synonymous subsets corresponding to any two preset grammatical symbols is the same, and the third probability value of the initial synonymous subsets corresponding to any two preset grammatical symbols is the same.

[0194] If it is determined that the preset grammatical symbol meets the second preset condition, the reconstructed synonymous subset determination module is specifically configured to:

[0195] The encoding result sequence is decoded using an arithmetic decoding algorithm and the preset decoding interval to obtain a reconstructed synonymous subset sequence, wherein the reconstructed synonymous subset sequence includes a plurality of reconstructed synonymous subsets.

[0196] For the convenience of description, the above devices are described as being functionally divided into various modules. Of course, when implementing the present disclosure, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0197] The apparatus of the above embodiment is used to implement the corresponding semantic source-based arithmetic coding and decoding method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0198] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the arithmetic coding method based on the semantic signal source described in any of the above embodiments is implemented.

[0199] FIG6 shows a more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other within the device via the bus 1050.

[0200] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0201] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 1020 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0202] The input / output interface 1030 is used to connect input / output modules to implement information input and output. The input / output modules can be configured as components within the device (not shown in the figure) or can be externally connected to the device to provide corresponding functions. Input devices may include a keyboard, mouse, touch screen, microphone, various sensors, etc., and output devices may include a display, speaker, vibrator, indicator light, etc.

[0203] The communication interface 1040 is used to connect to a communication module (not shown) to enable communication between the device and other devices. The communication module can communicate via a wired method (such as USB, network cable, etc.) or a wireless method (such as mobile network, WiFi, Bluetooth, etc.).

[0204] The bus 1050 comprises a path for transmitting information between the various components of the device (eg, the processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 ).

[0205] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, it will be understood by those skilled in the art that the above device may only include the components necessary to implement the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0206] The electronic device of the above embodiment is used to implement the corresponding semantic source-based arithmetic coding and decoding method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0207] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the semantic source-based arithmetic coding method as described in any of the above embodiments.

[0208] The computer-readable media of this embodiment include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0209] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the arithmetic coding method based on semantic information source as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0210] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0211] For example, in response to a user's active request, a prompt message is sent to the user to clearly inform the user that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the electronic device, application, server, storage medium, or other software or hardware that performs the operation of the disclosed technical solution based on the prompt message.

[0212] As an optional but non-limiting implementation, in response to a user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. Furthermore, the pop-up window may also contain a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0213] It is understandable that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0214] Based on the same concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a computer program product, including computer program instructions. When the computer program instructions are run on a computer, the computer executes the arithmetic encoding and decoding method based on the semantic signal source as described in any of the above embodiments, which has the beneficial effects of the corresponding method embodiments and will not be repeated here.

[0215] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0216] For example, in response to a user's active request, a prompt message is sent to the user to clearly inform the user that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the electronic device, application, server, storage medium, or other software or hardware that performs the operation of the disclosed technical solution based on the prompt message.

[0217] As an optional but non-limiting implementation, in response to a user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. Furthermore, the pop-up window may also contain a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0218] It is understandable that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0219] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Within the scope of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of simplicity.

[0220] In addition, to simplify the description and discussion, and so as not to obscure the embodiments of the present disclosure, known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, devices may be shown in the form of block diagrams to avoid obscuring the embodiments of the present disclosure, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present disclosure are to be implemented (i.e., these details should be fully within the purview of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present disclosure, it will be apparent to those skilled in the art that the embodiments of the present disclosure may be implemented without these specific details or with variations in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0221] Although the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0222] The embodiments of the present disclosure are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. An arithmetic encoding and decoding method based on a semantic information source, applied to an encoder, comprising: Obtaining a preset encoding interval of the semantic information source and a sequence of syntax symbols to be encoded; wherein, the sequence of syntax symbols to be encoded includes at least one syntax symbol; For each syntax symbol, determining a pre-constructed target synonymous subset corresponding to each syntax symbol; Using an arithmetic coding algorithm and the preset encoding interval, performing encoding processing on the target synonymous subset corresponding to each syntax symbol to obtain an encoded result sequence corresponding to the sequence of syntax symbols to be encoded; Obtaining the sequence length of the sequence of syntax symbols to be encoded; Sending the sequence length and the encoded result sequence to a decoder for the decoder to perform decoding processing on the encoded result sequence based on the sequence length to obtain a reconstructed syntax symbol sequence.

2. The method according to claim 1, wherein, The determining a pre-constructed target synonymous subset corresponding to each syntax symbol includes: obtaining target symbol information of the syntax symbol, and selecting an initial synonymous subset corresponding to the target symbol information from all the initially constructed synonymous subsets associated with the syntax symbol as the target synonymous subset corresponding to the syntax symbol.

3. The method according to claim 2, wherein, The process of constructing the initial synonymous subset includes: For each syntax symbol, obtaining a plurality of preset symbol information corresponding to the syntax symbol and semantic information corresponding to each preset symbol information; Dividing the plurality of preset symbol information according to the semantic information, and dividing all the preset symbol information corresponding to the same semantic information into the same set to obtain at least one initial synonymous subset; wherein, the target symbol information is included in the plurality of preset symbol information.

4. The method according to claim 2, wherein The using an arithmetic coding algorithm and the preset encoding interval to perform encoding processing on the target synonymous subset corresponding to each syntax symbol to obtain an encoded result sequence corresponding to the sequence of syntax symbols to be encoded includes: For each syntax symbol, performing a summation process on the probability values of each preset symbol information corresponding to each initially constructed synonymous subset associated with the syntax symbol to obtain the probability value of each initially constructed synonymous subset; Obtaining a first permutation order of all the syntax symbols in the sequence of syntax symbols to be encoded, and using an arithmetic coding algorithm to select a target coding interval from the preset encoding interval according to the first permutation order, the target synonymous subset corresponding to each syntax symbol, and the probability values of all the initially constructed synonymous subsets associated with each syntax symbol; Determining the shortest binary sequence in the target coding interval, and using the shortest binary sequence as the encoded result sequence corresponding to the sequence of syntax symbols to be encoded.

5. The method according to claim 4, wherein, The using an arithmetic coding algorithm to select a target coding interval from the preset encoding interval according to the first permutation order, the target synonymous subset corresponding to each syntax symbol, and the probability values of all the initially constructed synonymous subsets associated with each syntax symbol includes: Taking the preset encoding interval as a candidate coding interval, taking the first position in the first permutation order as the target position, and performing at least one round of iterative operation on the preset encoding interval. Each round of iterative operation is performed as follows: In the candidate coding interval, according to the probability values of all initial synonymous subsets corresponding to the grammar symbol of the target order, the candidate coding interval is partitioned to obtain at least one initial coding interval corresponding to each initial synonymous subset; The initial coding interval corresponding to the target synonymous subset corresponding to the grammar symbol of the target order in at least one initial coding interval is used as the updated candidate coding interval, and the next order of the target order is used as the target order for the next round of iteration according to the first arrangement order; Until there is no next order for the target order according to the first arrangement order, the iterative operation is exited; The candidate coding interval after at least one round of iterative operation is used as the target coding interval.

6. The method according to claim 1, wherein The encoding process for the target synonymous subset corresponding to each grammar symbol by using the arithmetic coding algorithm and the preset coding interval includes: in response to determining that the to-be-encoded grammar symbol sequence meets the first preset condition, constructing a target synonymous subset sequence according to the target synonymous subset corresponding to each grammar symbol; using the arithmetic coding algorithm and the preset coding interval to perform encoding processing on the target synonymous subset sequence to obtain a coding result sequence corresponding to the to-be-encoded grammar symbol sequence.

7. The method according to claim 6, wherein The first preset condition includes: the multiple preset symbol information corresponding to any two grammar symbols in the to-be-encoded grammar symbol sequence is the same, the initial synonymous subsets corresponding to any two grammar symbols are the same, and the probability values of the initial synonymous subsets corresponding to any two grammar symbols are the same.

8. An arithmetic encoding and decoding method based on a semantic information source, applied to a decoder, includes: Receiving the sequence length and the coding result sequence sent by an encoder; Obtaining the preset decoding interval and the preset grammar symbols of the semantic information source; Using the arithmetic decoding algorithm and the preset decoding interval to perform decoding processing on the coding result sequence to obtain at least one reconstructed synonymous subset; wherein, the number of the reconstructed synonymous subsets is the same as the sequence length; For each reconstructed synonymous subset, determining the semantic information corresponding to the reconstructed synonymous subset; Selecting a target reconstructed grammar symbol corresponding to the reconstructed synonymous subset from the preset grammar symbols according to the semantic information; Arranging the target reconstructed grammar symbols corresponding to each reconstructed synonymous subset to obtain a reconstructed grammar symbol sequence.

9. The method according to claim 8, wherein The using the arithmetic decoding algorithm and the preset decoding interval to perform decoding processing on the coding result sequence to obtain at least one reconstructed synonymous subset includes: Performing conversion processing on the coding result sequence to obtain a target codeword; Constructing a reconstructed symbol sequence according to the sequence length, and the number of elements in the reconstructed symbol sequence is the same as the sequence length; Determining the second arrangement order of the elements in the reconstructed symbol sequence; For each element in the reconstructed symbol sequence, obtaining at least one initial reconstructed synonymous subset corresponding to the element and the probability value of each initial reconstructed synonymous subset; According to the second arrangement order, the probability value of each initial reconstructed synonymous subset corresponding to each element, and the target codeword, using the arithmetic decoding algorithm to determine the target decoding interval corresponding to each element from the preset decoding interval; For each element, determine the reconstructed synonymous subset corresponding to the element according to the target decoding interval and all initial reconstructed synonymous subsets. The reconstructed synonymous subset corresponding to the element.

10. The method according to claim 9, wherein, The determining of the target decoding interval corresponding to each element from the preset decoding interval by using the arithmetic decoding algorithm according to the second permutation order, the probability value of each initial reconstructed synonymous subset corresponding to each element, and the target codeword includes: Taking the preset decoding interval as the candidate decoding interval, taking the first position in the second permutation order as the target position, and performing at least one round of iterative operation on the preset decoding interval. Each round of iterative operation is performed as follows: Dividing the candidate decoding interval according to the probability values of each initial reconstructed synonymous subset corresponding to the element at the target position to obtain at least one initial decoding interval, where the initial decoding intervals correspond one-to-one to the probability values of the initial reconstructed synonymous subsets; Selecting the initial decoding interval containing the target codeword from the multiple initial decoding intervals as the target decoding interval corresponding to the element at the target position, taking the target decoding interval as the updated candidate decoding interval, and taking the next position of the target position in the second permutation order as the target position for the next round of iteration; Until there is no next position for the target position according to the second permutation order, the iterative operation is exited; Obtaining the target decoding interval corresponding to each element.

11. The method according to claim 9, wherein, The decoding process of the encoded result sequence by using the arithmetic decoding algorithm and the preset decoding interval to obtain the reconstructed synonymous subset includes: in response to determining that the preset syntax symbol satisfies the second preset condition, using the arithmetic decoding algorithm and the preset decoding interval to perform decoding processing on the encoded result sequence to obtain a sequence of reconstructed synonymous subsets, where the sequence of reconstructed synonymous subsets contains multiple reconstructed synonymous subsets.

12. The method according to claim 11, wherein, The second preset condition includes: the multiple preset symbol information corresponding to any two preset syntax symbols is the same, the partitioning methods of the initial synonymous subsets corresponding to any two preset syntax symbols are the same, and the third probability values of the initial synonymous subsets corresponding to any two preset syntax symbols are the same.

13. An arithmetic coding and decoding device based on a semantic source, disposed in an encoder, includes: A data acquisition module, configured to acquire a preset coding interval of a semantic source and a sequence of syntax symbols to be encoded; Wherein the sequence of syntax symbols to be encoded contains at least one syntax symbol; A target synonymous subset determination module, configured to determine, for each syntax symbol, a pre-constructed target synonymous subset corresponding to each syntax symbol; An encoded result sequence determination module, configured to perform encoding processing on the target synonymous subset corresponding to each syntax symbol by using the arithmetic coding algorithm and the preset coding interval to obtain an encoded result sequence corresponding to the sequence of syntax symbols to be encoded; A data sending module, configured to acquire the sequence length of the sequence of syntax symbols to be encoded, and send the sequence length and the encoded result sequence to a decoder for the decoder to perform decoding processing on the encoded result sequence based on the sequence length to obtain a sequence of reconstructed syntax symbols.

14. An arithmetic coding and decoding device based on a semantic source, disposed in a decoder, includes: A coded result sequence receiving module, configured to receive the sequence length and the coded result sequence sent by an encoder, and obtain a preset decoding interval and preset syntax symbols of a semantic information source; A reconstructed synonymous subset determining module, configured to perform decoding processing on the coded result sequence by using an arithmetic decoding algorithm and the preset decoding interval to obtain at least one reconstructed synonymous subset, where the number of the reconstructed synonymous subsets is the same as the sequence length; A target reconstructed syntax symbol determining module, configured to, for each reconstructed synonymous subset, determine the semantic information corresponding to the reconstructed synonymous subset, and select a target reconstructed syntax symbol corresponding to the reconstructed synonymous subset from the preset syntax symbols according to the semantic information; A reconstructed syntax symbol sequence determining module, configured to arrange the target reconstructed syntax symbols corresponding to each reconstructed synonymous subset to obtain a reconstructed syntax symbol sequence.

15. An electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the program, the arithmetic encoding and decoding method based on a semantic information source as claimed in claim 1 or 8 is implemented.

16. A non-transitory computer-readable storage medium, where the non-transitory computer-readable storage medium stores computer instructions for causing a computer to execute the arithmetic encoding and decoding method based on a semantic information source as claimed in claim 1 or 8.

17. A computer program product, comprising: Computer program instructions, when the computer program instructions run on a computer, cause the computer to execute the arithmetic encoding and decoding method based on a semantic information source as claimed in claim 1 or 8.

Images