| 摘要 |
<p>PROBLEM TO BE SOLVED: To realize encoding/decoding efficiency of almost the same degree as an arithmetic encoding system and to considerably improve encoding and decoding speed. SOLUTION: In the proposed data encoding method, '0' or '1' is defined as a superior symbol and the other as an inferior symbol when a binary bit string constituted of '0' and '1' is inputted, and only a right signal is outputted when n-pieces of superior symbols are predicted to continue and the prediction on a noticed group is right. At the time of executing a prediction-run-length encoding, an input bit string is separated with the number of separation bits, which is smaller than the rum number of n-pieces of prediction bits. When an inferior symbol is included in a separated pattern, the pattern of the superior symbols, which contain the pattern and which continue till then, are collectively encoded. At the time of decoding, decoding is executed with an opposite algorithm.</p> |
