Abstract: The basic theory of KPD is introduced, and an image coding method based on wavelet and KPD is proposed. The image is decomposed by the Kronecker product decomposition, and the aim of the first compression will be achieved. According to the character of each small decomposed block, these small blocks will be classified into high frequency information and low frequency information. For the low frequency information, the method of SPECK code is selected, while for the high frequency information, the wavelet decomposition is used firstly, and then the maxing frequency information is coded, so the aim of  recompression will be reached. The experiment results show that this approach is simple and effective.

Key words: Kronecker Product Decomposition（KPD）, wavelet transform, image compression

摘要： 介绍了KPD的基本理论，提出了一种基于小波和KPD的图像编码方法。对图像进行Kronecker积分解，达到第一次压缩的目的。根据Kronecker积分解后的各个分解小块固有的性质，把这些小块分成两类：高频信息和低频信息。对于低频信息选取SPECK小波编码方式，而对于高频信息先进行小波分解，再对其中的混频信息进行编码，从而达到再次压缩的目的。实验证明该算法具有计算简便，压缩比较高的特点。

关键词: Kronecker积分解（KPD）, 小波变换, 图像压缩