Efficient coding method and its decoding method

Claim 1

A method of encoding data, comprising the steps of: (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j, where i=1, 2, . . . , s, and j=1, 2, . . . , m, s and m are integers not less than one, and i consecutively increases from 1 to s; (b) predicting the input data Di,j to obtain estimates Pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the input data Di,1, Di,2, . . . , Di,j-of the block, and additional codes of the block; (c) evaluating predictive errors Si,j which are the difference between the input data Di,j and estimates Pi,j thereof for respective values of j specified in step (b) in each block i; (d) determining a maximum value SXi of the predictive errors Si,j and a minimum value SNi of the predictive errors Si,j for each block i; (e) finding a divisor data OUi which is larger than the difference between the maximum and minimum values (SXi-SNi) of the predictive errors for each block i; (f) dividing the input data Di,j for respective values of j by the divisor data OUi to find remainders Ei,j in each block i; (g) encoding the remainders Ei,j for respective values of j in each block i; and (h) outputting the additional codes respectively representing the maximum value SXi, the minimum value SNi and the divisor data OUi of each block i.

Claim 2

A method of encoding data, comprising the steps of: (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j, where i=1, 2, . . . , s, and j=1, 2, . . . , m, s and m are integers not less than one, and i consecutively increases from 1 to s; (b) predicting the input data Di,j to obtain estimates Pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the input data Di,1, Di,2, . . . , Di,j-of the block, and additional codes of the block; (c) evaluating predictive errors Si,j which are the difference between the input data Di,j and estimates Pi,j thereof for respective values of j specified in step (b) in each block i; (d) determining a maximum value SXi of the predictive errors Si,j and a minimum value SNi of the predictive errors Si,j for each block i; (e) finding a divisor data OUi which is larger than the difference between the maximum and minimum values (SXi-SNi) of the predictive errors for each block i; (f) dividing the input data Di,j for respective values of j by the divisor data OUi to find remainders Ei,j in each block i; (g) encoding the remainders Ei,j for respective values of j in each block i; and (h) outputting the additional codes respectively representing the maximum value SXi, the minimum value SNi and the divisor data OUi of each block i.

Claim 9

A method of encoding data, comprising the step of: (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j (i=1, 2, . . . , s, j=1, 2, . . . , m) where s and m are integers not less than one, and i consecutively increases from 1 to s; (b) converting the input data Di,j to be encoded by a predetermined conversion method of obtaining converted data di,j for respective values of j in each block; (c) predicting the converted data di,j to be encoded to obtain estimates pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the converted data di,1, di,2, . . . , di,j-1 of the block, and a set of additional codes of the block; (d) evaluating predictive errors si,j which are the differences between the converted data di,j and the estimates pi,j thereof for respective values of j specified in step (c) in each block i; (e) determining a maximum value sxi of the predictive errors si,j and a minimum value sni of the same for each block i; (f) finding a divisor data oui which is larger than the difference between the maximum and minimum values (sxi-sni) of the predictive errors for each block i; (g) dividing the converted data di,j for respective values of j by the divisor data oui to find remainders ei,j in each block i; (h) encoding the remainders ei,j for respective values of j in each block i; and (i) outputting the additional codes respectively representing the maximum value sxi, the minimum value sni, the divisor data oui and a conversion information showing the conversion method used in the step (b), of each block i.

Claim 10

A method of encoding data, comprising the step of: (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j (i=1, 2, . . . , s, j=1, 2, . . . , m) where s and m are integers not less than one, and i consecutively increases from 1 to s; (b) converting the input data Di,j to be encoded by a predetermined conversion method of obtaining converted data di,j for respective values of j in each block; (c) predicting the converted data di,j to be encoded to obtain estimates pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the converted data di,1, di,2, . . . , di,j-1 of the block, and a set of additional codes of the block; (d) evaluating predictive errors si,j which are the differences between the converted data di,j and the estimates pi,j thereof for respective values of j specified in step (c) in each block i; (e) determining a maximum value sxi of the predictive errors si,j and a minimum value sni of the same for each block i; (f) finding a divisor data oui which is larger than the difference between the maximum and minimum values (sxi-sni) of the predictive errors for each block i; (g) dividing the converted data di,j for respective values of j by the divisor data oui to find remainders ei,j in each block i; (h) encoding the remainders ei,j for respective values of j in each block i; and (i) outputting the additional codes respectively representing the maximum value sxi, the minimum value sni, the divisor data oui and a conversion information showing the conversion method used in the step (b), of each block i.

Claim 14

A method of encoding and decoding data, comprising: (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j, where i=1, 2, . . . , s, and j=1, 2, . . . , m, s and m are integers not less than one, and i consecutively increases from 1 to s; (b) predicting the input data Di,j to obtain estimates Pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the input data Di,1, Di,2, . . . , Di,j-1 of the block, and additional codes of the block; (c) evaluating predictive errors Si,j which are the difference between the input data Di,j and estimates Pi,j thereof for respective values of j specified in step (b) in each block i; (d) determining a maximum value SXi of the predictive errors Si,j and a minimum value SNi of the predictive errors Si,j for each block i; (e) finding a divisor data OUi which is larger than the difference between the maximum and minimum values (SXi-SNi) of the predictive errors for each block i; (f) dividing the input data Di,j for respective values of j by the divisor data OUi to find remainders Ei,j in each block i; (g) encoding the remainders Ei,j for respective values of j in each block i; (h) outputting the additional codes respectively representing the maximum value SXi, the minimum value SNi and the divisor data OUi of each block i; and (i) decoding said encoded remainders Ei,j and said additional codes by the following steps: (j) increasing i consecutively from 1 to s; (k) finding a divisor OUi for each block i by use of the additional codes of the block; (l) recovering remainders Ei,j for respective values of j in each block i by decoding the additional codes representing remainder Ei,j of the block; (m) predicting data Di,j to obtain estimates Pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of data Di,1, Di,2 . . . , Di, j-1 of the block which have already been decoded, and a set of the additional codes of the block; (n) finding offsets Fi,j for respective values of j specified in step (m) in each block i, by use of the additional codes, the recovered remainders Ei,J and the estimates Pi,j of the block, wherein the offsets Fi,j are Nj times the divisor OUi, where Nj are uniquely determinable integers; (o) adding the offsets Fi,j and the recovered remainders Ei,j to obtain decoded data Di,j for respective values of j specified in the step (m) in each block i; and (p) deblocking the blocks to which steps (j) to (o) have been applied.

Claim 15

A method of encoding and decoding data, comprising: (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j, where i=1, 2, . . . , s, and j=1, 2, . . . , m, s and m are integers not less than one, and i consecutively increases from 1 to s; (b) predicting the input data Di,j to obtain estimates Pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the input data Di,1, Di,2, . . . , Di,j-1 of the block, and additional codes of the block; (c) evaluating predictive errors Si,j which are the difference between the input data Di,j and estimates Pi,j thereof for respective values of j specified in step (b) in each block i; (d) determining a maximum value SXi of the predictive errors Si,j and a minimum value SNi of the predictive errors Si,j for each block i; (e) finding a divisor data OUi which is larger than the difference between the maximum and minimum values (SXi-SNi) of the predictive errors for each block i; (f) dividing the input data Di,j for respective values of j by the divisor data OUi to find remainders Ei,j in each block i; (g) encoding the remainders Ei,j for respective values of j in each block i; (h) outputting the additional codes respectively representing the maximum value SXi, the minimum value SNi and the divisor data OUi of each block i; and (i) decoding said encoded remainders Ei,j and said additional codes by the following steps: (j) increasing i consecutively from 1 to s; (k) finding a divisor OUi for each block i by use of the additional codes of the block; (l) recovering remainders Ei,j for respective values of j in each block i by decoding the additional codes representing remainder Ei,j of the block; (m) predicting data Di,j to obtain estimates Pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of data Di,1, Di,2 . . . , Di, j-1 of the block which have already been decoded, and a set of the additional codes of the block; (n) finding offsets Fi,j for respective values of j specified in step (m) in each block i, by use of the additional codes, the recovered remainders Ei,J and the estimates Pi,j of the block, wherein the offsets Fi,j are Nj times the divisor OUi, where Nj are uniquely determinable integers; (o) adding the offsets Fi,j and the recovered remainders Ei,j to obtain decoded data Di,j for respective values of j specified in the step (m) in each block i; and (p) deblocking the blocks to which steps (j) to (o) have been applied.

Claim 19

A method of encoding and decoding data, comprising (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j (i=1, 2, . . . , s, j=1, 2, . . . , m) where s and m are integers not less than one, and i consecutively increases from 1 to s; (b) converting the input data Di,j to be encoded by a predetermined conversion method of obtaining converted data di,j for respective values of j in each block; (c) predicting the converted data di,j to be encoded to obtain estimates pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the converted data di,1, di,2, . . . , di,j-1 of the block, and a set of additional codes of the block; (d) evaluating predictive errors si,j which are the differences between the converted data di,j and the estimates pi,j thereof for respective values of j specified in step (c) in each block i; (e) determining a maximum value sxi of the predictive errors si,j and a minimum value sni of the same for each block i; (f) finding a divisor data oui which is larger than the difference between the maximum and minimum values (sxi-sni) of the predictive errors for each block i; (g) dividing the converted data di,j for respective values of j by the divisor data oui to find remainders ei,j in each block i; (h) encoding the remainders ei,j for respective values of j in each block i; (i) outputting the additional codes respectively representing the maximum value sxi, the minimum value sni, the divisor data oui and a conversion information showing the conversion method used in the step (b), of each block i; and (j) decoding said encoded remainders Ei,j and the additional codes by the following steps: (k) increasing i consecutively from 1 to s; (l) finding a divisor oui for each block i by use of the additional codes of the block; (m) recovering remainders ei,j for respective values of j in each block i by decoding the additional codes representing remainders ei,j of the block; (n) predicting data di,j to obtain estimates pi,j thereof for consecutively increasing j in each block i, by use of at least one of a set of data di,1, di,2, . . . , di,j-1 which have already been decoded, and the additional codes of the block; (o) finding an offset fi,j for respective values of j specified in step (n) in each block i, by use of the additional codes, the recovered remainders ei,j and the estimates pi,j of the block, wherein the offsets fi,j are nj times the divisor oui, where nj are uniquely determinable integers; (p) adding the offsets fi,j and the recovered remainders ei,j to obtain decoded data di,j for respective values of j specified in step (n) in each block i; (q) re-converting the decoded data di,j to obtain data Di,j for respective values of j in each block i, by way of a reversal operation of converting step (b) according to conversion information given in one of the additional codes; and (r) deblocking the blocks to which the steps (k) to (q) have been applied.

Claim 20

A method of encoding and decoding data, comprising (a) punctuating an input signal to be encoded into s blocks, each block having m input data Di,j (i=1, 2, . . . , s, j=1, 2, . . . , m) where s and m are integers not less than one, and i consecutively increases from 1 to s; (b) converting the input data Di,j to be encoded by a predetermined conversion method of obtaining converted data di,j for respective values of j in each block; (c) predicting the converted data di,j to be encoded to obtain estimates pi,j thereof for consecutively increasing j in each block i, by use of at least one of a partial set of the converted data di,1, di,2, . . . , di,j-1 of the block, and a set of additional codes of the block; (d) evaluating predictive errors si,j which are the differences between the converted data di,j and the estimates pi,j thereof for respective values of j specified in step (c) in each block i; (e) determining a maximum value sxi of the predictive errors si,j and a minimum value sni of the same for each block i; (f) finding a divisor data oui which is larger than the difference between the maximum and minimum values (sxi-sni) of the predictive errors for each block i; (g) dividing the converted data di,j for respective values of j by the divisor data oui to find remainders ei,j in each block i; (h) encoding the remainders ei,j for respective values of j in each block i; (i) outputting the additional codes respectively representing the maximum value sxi, the minimum value sni, the divisor data oui and a conversion information showing the conversion method used in the step (b), of each block i; and (j) decoding said encoded remainders Ei,j and the additional codes by the following steps: (k) increasing i consecutively from 1 to s; (l) finding a divisor oui for each block i by use of the additional codes of the block; (m) recovering remainders ei,j for respective values of j in each block i by decoding the additional codes representing remainders ei,j of the block; (n) predicting data di,j to obtain estimates pi,j thereof for consecutively increasing j in each block i, by use of at least one of a set of data di,1, di,2, . . . , di,j-1 which have already been decoded, and the additional codes of the block; (o) finding an offset fi,j for respective values of j specified in step (n) in each block i, by use of the additional codes, the recovered remainders ei,j and the estimates pi,j of the block, wherein the offsets fi,j are nj times the divisor oui, where nj are uniquely determinable integers; (p) adding the offsets fi,j and the recovered remainders ei,j to obtain decoded data di,j for respective values of j specified in step (n) in each block i; (q) re-converting the decoded data di,j to obtain data Di,j for respective values of j in each block i, by way of a reversal operation of converting step (b) according to conversion information given in one of the additional codes; and (r) deblocking the blocks to which the steps (k) to (q) have been applied.