摘要
为了改善直方图均衡对低灰度层密集分布图像的处理效果,通过分析图像直方图均衡理论对同一灰度像素在处理后映射到同一灰度的过程中存在的理论缺陷,提出了一种高效的、自适应的算法.该算法首先对图像进行常规的直方图均衡处理,然后对处理后所形成的图像灰度在一个较大的灰度范围内进行映射,保证变换后的图像有较大的灰度动态范围,最后对结果进行适当的亮度修正,保证变换图像整体有较高的对比度,便于观察图像的细节.通过实验证明,改进后的算法可以明显地改善直方图均衡处理低灰度层密集分布图像的效果,同时具有较高的算法效率,适合在实时图像处理系统中采用.
There are disadvantages in the image histogram equalization algorithm when it maps the pixels in the same gray level to the same new gray level, which results in unsatisfactory performance of the processed image, especially when the picture has many pixels distributed at low gray levels. To resolve this problem, an improved self-adaptive image histogram equalization algorithm was presented. After the image histogram equalization processing, the new method extends the gray levels to ensure the pixels distribute at large-scale gray levels and corrects the brightness of the resulting image to make the details be seen easily. Theoretical analysis and experiments showed that the algorithm is effective for histogram equalization to low gray level distributing image and that it is relatively better for real-time image processing system.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2007年第4期630-633,共4页
Journal of Zhejiang University：Engineering Science
关键词
直方图均衡
图像增强
自适应算法
实时图像处理
image histogram equalization
image enhancement
self-adaptive algorithm
real time image processing
作者简介
张懿（1980-），男，浙江杭州人，硕士生，从事图像显示的研究工作．E—mail：ezhang5@hotmail．com
通讯联系人：刘旭，男，教授，博导．E-mail：liuxu@zju．edu．cn
