摘要： 在研究非对称逆布局模式表示模型(NAM)的基础上，针对多值图像表示中子模式的数量优化问题，提出一种基于位平面分解的优化策略，利用位面间的相关性，改变位面内的扫描算法，进一步减少数据冗余，实现更高的压缩率，并进行理论分析和实验验证。结果表明，该优化策略可以大幅减少子模式数量，优化后的子模式个数减少7.1%~27.1%。

关键词: 模式表示, 非对称逆布局模式表示模型, 优化策略, 位平面分解

Abstract: On basis of analyzing Non-symmetry Anti-packing pattern representation Model(NAM), an optimization strategy based on Bit Plane Decomposition(BPD) is proposed for the optimization problem on the number of sub-patterns in image representation for multi-valued images. The pertinence between bit planes is employed and the scanning algorithm is changed, thus data redundancy is reduced and a higher compression ratio is achieved which is analyzed theoretically and validated experimentally. Experimental results show the optimization strategy can sharply decrease the number of sub-patterns which is reduced by 7.1% to 27.1% after the optimization.

Key words: pattern representation, Non-symmetry Anti-packing pattern representation Model(NAM), optimization strategy, Bit Plane Decomposition(BPD)

中图分类号: 

• TP391