基于小波变换的静止图像压缩编码技术研究

【作者】 黄维；

【导师】 罗桂娥；

【作者基本信息】 中南大学 ， 控制理论与控制工程， 2007， 硕士

【摘要】 随着图像应用范围和深度的不断扩大，不仅要求实现高压缩率、高保真度和快速性，还应满足诸如渐进传输等网络应用的需要。因此，图像／视频的压缩技术成为国际上热门的研究课题。本文在对图像压缩编码理论与实际应用技术的的现状和发展趋势进行简要分析的基础上，从一定理论深度较为详细阐述了基于小波变换的静止图像压缩编码技术，重点探讨了编码器结构、编码步骤、小波基的选取以及影响压缩编码性能因素等几个关键问题。对基于图像压缩性能的小波特性进行了分析与仿真，通过仿真比较了具有相似小波特性的正交与双正交小波的压缩性能。在基于小波变换的图像压缩方案中，嵌入式零树小波(EZW)编码能够很好地利用小波系数的特性使得输出的码流具有嵌入特性。论文在深入分析了EZW算法的原理、步骤、实现过程和优缺点的基础上，针对EZW算法计算复杂度高的缺陷，从探求算法中变换与编码同步实现的角度出发，提出了一种基于分块变换一融合的改进EZW算法。在上述理论工作的基础上，通过VC编程实现了该算法。实验结果表明，使用本算法，解码图像保持了与EZW算法相似的客观性能，但通过使用变换阶段的副产品一带峰值，可以简化零树扫描步骤，在一定程度上降低了算法执行时间，同时可实现真正意义上的子图像编码。更多还原

【Abstract】 With the extension of image application,some new image compression and coding techniques which can not only achive high compression ratio and definition but also meet such demands by Progressive Transmission,and other applications are needed. Therebefore, compression and coding technique in image and video becomes a hot point in image research field.In this paper, the digital image compression techniques based on wavelet transform are analyzed and summarized, the current state of the techniques is introduced and several directions of development are proposed. Some important problems such as the arichitecture of wavelet coder, the coding step and how to choose wavelet basis are discussed, some simulations with matlab wavelet toolbox is made to demonstrate. A conclusion that the difference in performance between orthnorgal wavelets and biorthnorgal wavelets were not as largr as we desired when they have familiar mathematical properties is also made.This paper proposes an improved EZW algorithm based on Partition-transform-Merge computation. Stronger coupling and pipelining between the transform and the coding stages is made possible. The algorithm computes the sub_band_max for each sub-band as a by-product of the Partition-transform-Merge computation to improve the zero-tree checking process and reduce the computing complexities.An extensive research on static image compression is presented in this paper. On the basis theoretical above, programming achieves corresponding algorithm, experimental results show that our new algorithm actually improves the execution time .更多还原