一种基于小波分析和人眼视觉特性的图像增强方法

在分析视觉系统特性的基础上，提出了一种基于小波分析的图像对比度增强算法。为降低对噪声增强程度，算法对不同尺度上的系数进行不同程度的对比度增强。在此过程中，算法考虑了视觉特性以得到更好的效果。为保证得到足够的对比度增强效果，算法首先对分解得到的尺度系数进行增强；为避免视觉明显的噪声放大，只对处于边缘区域的系数进行对比度拉伸；为了能在各尺度上较为准确地找出边缘区域，算法采用局部阈值技术并利用了各分解尺度上边缘区域的相关性；为降低视觉明显的噪声放大，不同区域的对比度增强程度决定于区域的亮度。算法的实验表明，上述特点使本文算法在充分增强对比度的同时，极大地降低了对噪声的过增强，具有较好的视觉效果。

An Approach for Image Enhancement Based on Wavelet Analysis and Properties of Human Visual System

The principal properties of human visual system are analysed, then a wavelet-based algorithm for image contrast enhancement is proposed. The decomposed coefficients at different scales are enhanced in different degrees so as to decrease the degree of enhancement of noise in image. In the procedure, the algorithm considers the properties of HVS to achieve better performance. Both scale and wavelet coefficients are enhanced to ensure achieving enough effect of enhancement, and only the contrast of the coefficients in edge region is stretched to avoid visual obvious amplification of noise. In order to find the edge regions at each scale more correctly, the algorithm has advantage of the technique of local threshold and the correlation of edge regions at neighboring scales. The average intensity of a region is determined by the degree of enhancement in the region so as to make the amplification of noise unobvious visually. Simulations demonstrate that these characters of the apporach achieve better enhancement effect than the traditional approaches.