摘要： 小波变换后，图像的小波系数中含有的大量零系数。文章根据图像的小波系数的这一统计特性提出了一种新的图像压缩编码方法，即方块填充法。利用图像小波系数的特征，对零系数比较密集的区域进行方块填充，并对各个方块进行标识，然后对填充后的小波系数编码，就可以用很少的比特描述比较多的零系数。实验结果表明，与零树编码方法相比，在大大节约运算量的同时，也获得了很大的数据压缩率。

关键词: 小波图像；图像压缩；图像编码；方块填充法

Abstract: There are lots of coefficients around zero in the wavelet coefficients of image data. In this paper, a novel image coding method based on this statistical properties of wavelet coefficients of image data is presented. It utilizes this property of wavelet coefficients of image data, filles the square areas of the wavelet coefficients where all the elements equal zero with zero filling of square block mechod, then markes these blocks. At last the remaining coefficients and the marks are coded. So the square area where the coefficients are the same can be described with only a few bit by zero filling of square block method. The experimental result shows that this method can save computing cost as well as obtain comparative compression ratio comparing with zerotree wavelet image coding.

Key words: Wavelet image; Image compression; Image coding; Zero filling of square block