IOE OpenIR  > 光电技术研究所博硕士论文
Thesis Advisor雷泽宽
Degree Grantor中国科学院光电技术研究所
Place of Conferral中国科学院光电技术研究所
Keyword信噪比 小波变换 重复字符计数混合编码 双门限降阶系数分解
Abstract本文提出一种基于小波分析的低损失率(高信噪比)、高压缩比图像压缩算法,可以用于大多数同时需要高信噪比和高压缩比的图像压缩与存储领域。算法基本组成:首先对图像做三层双正交小波分解,分解后的系数是浮点数,为了便于压缩处理,只保留系数的整数部队分,丢弃小数部分。对包含图像基本信息的低频子图像系数(LL3系数)保留,对包含高频信息的子图像做如下无损压缩编码:重复字符记数混合编码,双门限降阶系数分解,Huffman编码:获得的码流和LL3系数合并,形成输出码流。仿真实验表明,算法压缩比达到6.8:1以上,压缩图像的PSNR ≥ 75dB。算法提出了一种新思路,为了在保持高信噪比的前提下获得高压缩比,先利用小波变换等先进数学工具把图像分解为一些系数,然后使用可逆编码技术压缩。针对系数的特征,算法中提出了一种新的可逆压缩方法,获得了较高的压缩比。为了验证算法用于嵌入式系统中的性能,采用TMS320C50数字信号处理器设计硬件电路,做了DSP压缩实验。实验结果证明算法用于嵌入式系统是可行的。
Other AbstractThis paper presents a wavelet analysis-based image compression algorithm with low information loss ratio and high compression ratio. It can be applied to most image compression and storage fields which need high signal-noise ratio and high compression ratio simultaneously. The base flow of the algorithm is: Firstly, the image is decomposed to coefficients by three level dual-orthogonal wavelet transform. These coefficients are float numbers, for the purpose of processing conveniently, only their integer part are saved and their fraction part are discarded. Sequentially, the coefficients of low frequency sub-image (namely LL3 coefficients) are hold, and the coefficients of high frequency sub-image are compressed by lossless compression process whose base flow is: mixed repeat character counting coding, phase descending coefficient decomposing with two threshold, Huffman coding, then a data stream is produced. Sequentially, combining the data stream with LL3 coefficients to form the final code stream. The emulational experiment result proves that the algorithm's compression ratio is larger than 6.8:1. The peak value signal-noise ratio (PSNR) is lager than 75dB. The algorithm presents a new kind of method that, in order to gain high compression ratio under the precondition of keeping high signal-noise ratio, firstly using advanced mathematics tool such as wavelet-analysis to decompose the image into some coefficients, then using reversible coding technology to compress them. Aiming at characteristics of these coefficients, a new kind of reversible compression method in the algorithm is developed to gain high compression ratio. In order to test the algorithm's performance using in the general embedded system, a hardware circuit was factured using digital signal processor TMS320C50 and did the compression experiment. The experiment result shows that the algorithm is feasible to be used in embedded system.
Document Type学位论文
Recommended Citation
GB/T 7714
杨洪. 基于小波分析的图像低损高压缩比技术研究[D]. 中国科学院光电技术研究所. 中国科学院光电技术研究所,1999.
Files in This Item:
File Name/Size DocType Version Access License
LW000617.pdf(2197KB) 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[杨洪]'s Articles
Baidu academic
Similar articles in Baidu academic
[杨洪]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杨洪]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.