基于小波变换的多层次图像增强算法  

The algorithm of multi-level image enhancement based on wavelet transform

作　　者：陈志坚[1] 成培[1] 李峰[1] 

机构地区：[1]长沙理工大学计算机与通信工程学院,湖南长沙410076

出　　处：《长沙理工大学学报（自然科学版）》2006年第2期69-72,共4页Journal of Changsha University of Science and Technology:Natural Science

基　　金：湖南省自然科学基金资助项目(05JJ30123);湖南省教育厅科研资助项目(05C246)

摘　　要：提出了基于小波变换的多层次图像增强算法.首先对图像进行4级小波分解,得到尺度系数和多个层次的小波系数;然后对不同层次的小波系数采用不同的增强算法进行处理,并对图像的尺度系数采用多尺度方法进行处理;最后用得到的小波系数和尺度系数进行逆变换得到增强后的图像.试验结果证明,该算法具有优于传统增强算法的增强效果和抗噪性能.An algorithm of multi-level image enhancement based on wavelet transform was presented. First, the image was decomposed into four-level wavelet, the scale coefficients treat and the multi-level wavelet coefficients can be obtained. Then, different enhancement algorithms were used to treat different levels of wavelet coefficients, and scale coefficients were processed by multiscale retinex. At last, an enhanced image was obtained after inverse wavelet transform using the processed wavelet and scale coefficients were given. All the results show that the algorithm can improve the image enhanlement and avoid a noise compared to traditional enhancement algorithm.

关 键 词：小波变换 图像增强 尺度系数 小波系数 逆变换 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]