一种新的基于小波变换的图像压缩方法  

A New Image Compression Scheme Based on Wavelet Transform

作　　者：王孝国[1] 王曙光[2] 

机构地区：[1]解放军理工大学通信工程学院研究生2队,江苏南京210016 [2]解放军理工大学通信工程学院信号和信息处理研究所

出　　处：《军事通信技术》2001年第3期5-8,共4页Journal of Military Communications Technology

摘　　要：提出一种基于小波变换的图像数据的压缩编码方法 ,基本思想是利用小波变换实现图像的多分辨率分解 ,采用了结合矢量量化 (VQ)的混合编码技术对分解后的图像进行编码 ,考虑图像的边缘轮廓对视觉质量的影响 ,利用了小波图像的方向选择性和低频子带的边缘信息组织码本 ,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 。Based on wavelet transform and vector quantization, a new scheme for image data compression is presented. The image is decomposed in multiresolution by wavelet transform, then it is encoded by a mixed encoding technology combining with vector quantization(VQ). Considering the influence of the vision′s quality caused by the edge and outline information of the image, we organize the coding by making use of the direction selection of the wavelet image and the edge information of low frequency. Based on some characteristics of components in the vector, a new method is put forward which is used to get the initial codebook of LBG algorithm in VQ. The simulation results manifest that possible cavity disproportionate phenomenon is avoided, the quality of the codebook is improved, and the quality of reconstructed image is also improved at low bit rate.

关 键 词：图像压缩 小波变换 预测编码 矢量量化 四叉树 图像编码 

分 类 号：TN919.81[电子电信—通信与信息系统]