彩色图像三维矩阵变换压缩编码

本文提出了一种基于人类视觉的新的彩色图像压缩编码方法，即三维矩阵变换压缩编码，推导了三维矩阵混合变换理论及其实现方法，使得压缩编码更充分地利用极色图象在空间上及各分量之间的相关性，与其它变换压缩编码不同的是：它不是把彩色图像的各分量独立开来，而是将去除各分量内部及各分量之间的冗余同时进行，仿真结果表明，该方法的重建图象具有有较好的视觉效果并有较好的压缩比。

Compression Coding of Color Image via 3-D Matrix Transform

In this paper,a new technique of color image compression coding based on human vision, 3-D matrix transform compression coding,has been proposed. A mixed transform theory of3-D matrix and its accomplishment has been introduced to fully exploit spatial as well as spectral redundancies of the color image. Different from other transform coding methods, in which each component was compressed and encoded separately,it combined the intra and intercomponent compressions together. Simulution results demonstrate good quality reproduction and high compression ratio of this method.