摘要
传统直方图均衡化算法在增强图像的同时也丢失了图像细节、增强了图像的噪声信号,导致信息熵下降。结合小波变换多尺度、多分辨率的特点和直方图均衡的优势,提出一种基于小波分频和直方图均衡的高亮度图像增强算法。首先利用小波变换将图像分解为低频分量和高频分量,然后仅对低频分量做直方图均衡处理,再由均衡后的低频分量与各高频分量进行小波重构。实验结果表明,该算法对于亮度较高的灰度图像有较好的增强效果。
The traditional histogram equalization processing algorithm lost image details and enhanced noise signal while it enhanced image,which led to the descent of information entropy.A high brightness image enhancement algorithm based on wavelet frequency division and histogram equalization is put forward in combination with the characteristics of multi-scale and multi-resolution of wavelet transform and the predominance of histogram equalization.At first,the image is divided into the low frequency parts and the high frequency parts,and histogram equalization processing is only applied to the low frequency parts.Then,the wavelet is reconstituted by the low frequency parts which have been equalized with the original high frequency parts.The experimental results show that this algorithm can enhance image of high brightness effectively.
出处
《现代电子技术》
2010年第16期149-150,153,共3页
Modern Electronics Technique
基金
无源毫米波成像探测机理及关键技术研究(D200711003)
关键词
图像增强
小波变换
小波分频
直方图均衡
image enhancement
wavelet transform
wavelet frequency division
histogram equalization
