摘要： 为了实现基于内容的编码，在ＭＰＥＧ－４中引入了视频对象层（ＶＯＰ）的概念，ＳＡ－ＤＣＴ技术可以用于对ＶＯＰ中边缘块的编码．形状自适应区域分割方法可以提高ＳＡ－ＤＣＴ的编码效率，减少需要编码的块数，但相对较复杂。一种快速区域分割方法被引入，它直接对区域作出划分，实验表明，它在降低分割复杂度的同时，保持了区域的分割效率．

关键词: MPEG-4, 形状自适应, 区域分割, SA-DCT

Abstract: VOP concept is introduced in MPEG-4 for the content-based coding. The SA-DCT algorithm can be applied to the blocks which straddle the boundedes of VOP. The shape-adaptive region partitioning method can improve the coding efficiency of SA-DCT and reduce the number of coding blocks, but it has high compuboonal complexity. A fast region partitioning method is introduced, it partitions blocks directly and has low complexity. Computel simulations ho comparisons with results of other methods reveal that this technique is quite promising and competitive, it decreases the complexity greatly with high coding efficiency.

中图分类号: 

• TN919.8

• TN911.7