# US 9,521,433 B2 · Video encoding device, video decoding device, video encoding method, video decoding method, video encoding or decoding program > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 9,521,433 B2 - Document ID: US9521433B2 - Status: Granted - Kind: B2 - Application number: 12/676,207 - Filed: 2008-09-05 - Published: 2010-07-15 - Granted: 2016-12-13 - Earliest claimed: 2007-09-06 - Assignees: NEC Corporation - Applicants: Keiichi Chono; Yoshihiro Miyamoto - Inventors: Keiichi Chono; Yoshihiro Miyamoto - Web page: https://patentgrep.com/patent/US9521433B2 ## CPC - H04N 19/93: Run-length coding - H04N 19/136: Incoming video signal characteristics or properties - H04N 19/196: being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters (processing of motion vectors H04N19/513) - H04N 19/21: with binary alpha-plane coding for video objects, e.g. context-based arithmetic encoding [CAE] - H04N 19/91: Entropy coding, e.g. variable length coding [VLC] or arithmetic coding - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract Intended is to obtain a video encoding device, a video encoding method, and a video encoding program which enable to prevent reduction in compression efficiency caused by drastic changes in symbol occurrence probabilities in context adaptive coding, and a video decoding device, a video decoding method and a decoding program corresponding thereto. The video data 101 is input into video encoding device on a macroblock basis, and after quantization, the PCM determination unit 139 determines whether the coded data 123 is PCM mode or not. Even when the coded data 123 is PCM mode, the context updating unit 301 executes context updating processing so as to improve efficiency of binary arithmetic coding. According to the determination of PCM mode or non PCM mode, the third switch 148 supplies either binary arithmetic coding output or PCM coding output as its output data. ## Claims ### Claim 1 A video encoding device, comprising: a first encoding unit for subjecting transformed and quantized values which are obtained by applying frequency-transform and quantization to an image block, to binary arithmetic coding which is entropy coding based on contexts, and updating said contexts based on data of being target to encode; a second encoding unit for encoding said image block by an encoding method which is different from the method of said first encoding unit and is non-entropy encoding; an output data selection unit for selecting either first output data output by said first encoding unit or second output data output by said second encoding unit; and a context updating unit for, when the output data selection unit selects said second output data encoded by the non-entropy coding, updating said contexts for use in the binary arithmetic coding of said transformed and quantized values by increasing least-probable-symbol occurrence probabilities of said contexts with respect to least-probable-symbol so that symbol occurrence probabilities of the context coincide occurrence probabilities of bin, which is the least-probable-symbol, in the coded data before the next entropy coding being execute. ### Claim 2 (depends on 1) The video encoding device according to claim 1, wherein said context updating unit is a unit which updates contexts for use in the binary arithmetic coding of said transformed and quantized values so as to decrease the number of contexts associated with low least-probable-symbol occurrence probabilities. ### Claim 3 (depends on 1) The video encoding device according to claim 1, wherein said context updating unit is a unit which increases occurrence probabilities of least-probable-symbols of contexts for use in the binary arithmetic coding of said transformed and quantized values. ### Claim 4 (depends on 1) The video encoding device according to claim 1, wherein said context updating unit is a unit which writes contexts associated with predetermined certain values of symbol occurrence probabilities over contexts for use in the binary arithmetic coding of said transformed and quantized values. ### Claim 5 (depends on 4) The video encoding device according to claim 4, wherein said predetermined certain values of symbol occurrence probabilities are the value of equal probability. ### Claim 6 (depends on 1) The video encoding device according to claim 1, wherein the encoding method different from the method of said first encoding unit is PCM coding. ### Claim 7 A video encoding method, comprising: a first encoding step of subjecting transformed and quantized values which are obtained by applying frequency-transform and quantization to an image block to binary arithmetic coding which is entropy coding based on contexts, and updating said contexts based on data of being target to encode; a second encoding step of encoding said image block by an encoding method which is different from the method of said first encoding step and is non-entropy encoding; an output data selection step of selecting either first output data output at said first encoding step or second output data output at said second encoding step; and a context updating step of, when at the output data selection step, said second output data encoded by the non-entropy coding is selected, updating said contexts for use in the binary arithmetic coding of said transformed and quantized values by increasing least-probable-symbol occurrence probabilities of said contexts with respect to least-probable-symbol so that symbol occurrence probabilities of the context coincide occurrence probabilities of bin, which is the least-probable-symbol, in the coded data before the next entropy coding being execute. ### Claim 8 (depends on 7) The video encoding method according to claim 7, wherein the coding method different from the method of said first encoding step is PCM coding. ### Claim 9 (depends on 7) The video encoding method according to claim 7, wherein at said context updating step, updating contexts for use in the binary arithmetic coding of said transformed and quantized values so as to decrease the number of contexts having low least-probable-symbol occurrence probabilities. ### Claim 10 (depends on 7) The video encoding method according to claim 7, wherein at said context updating step, increasing occurrence probabilities of least-probable-symbols of contexts for use in the binary arithmetic coding of said transformed and quantized values. ### Claim 11 (depends on 7) The video encoding method according to claim 7, wherein at said context updating step, writing contexts associated with predetermined certain values of symbol occurrence probabilities over contexts for use in the binary arithmetic coding of said transformed and quantized values. ### Claim 12 (depends on 11) The video encoding method according to claim 11, wherein said predetermined certain values of symbol occurrence probabilities are the value of equal probability. ### Claim 13 (depends on 1) The video encoding device according to claim 1, wherein said context updating unit updates contexts for use in the binary arithmetic coding of said transformed and quantized values by setting most-probable-symbol occurrence probabilities of contexts to low probability. ### Claim 14 (depends on 7) The video encoding method according to claim 7, wherein at said context updating step, updating contexts for use in the binary arithmetic coding of said transformed and quantized values is by setting most-probable-symbol occurrence probabilities of contexts to low probability.