出一种基于小波变换的灰度图像数据压缩编码方法 ,基本思路是利用小波变换实现图像的多分辨率分解 ,用矢量量化 (VQ)对分解后的图像进行编码 ,在矢量量化 L BG算法的初始码书的选取中根据矢量中各分量的特性提出一种改进的随机选取法 ,避免了可能的胞腔不均现象 ,提高了码书的质量 ,而且重构的图像质量也有所提高。 A wavelet transform based gray image data compression coding method, the basic idea is to use wavelet transform to achieve multi-resolution image decomposition, the vector quantization (VQ) to decompose the image encoding, vector quantization in the L BG algorithm In the selection of initial codebook, an improved stochastic selection method is proposed according to the characteristics of each component in the vector, which avoids possible intra-cell inhomogeneity, improves the quality of the codebook, and improves the quality of the reconstructed image.