一种基于截取法的灰度级等距离拉伸图像增强研究

A Method of Image Enhancement Based on Gray-levels in Equal Distance with Interceptive Way

针对灰度直方图均衡化所引起的过增强和灰度级"吞并"等缺陷,提出基于截取法的灰度级等距离拉伸增强图像的对比度。通过对原始图像灰度直方图的统计,利用正态分布的"3σ"原则对灰度级进行截取,再将其有效灰度级进行等距离排列,完成排列后的直方图映射,最后利用中值滤波对图像进行平滑处理。该算法在增强图像对比度的同时,保留图像细节信息,提高图像质量,并且避免了直接均衡化所带来的灰度级断层现象。

The paper proposes a method of image enhancement based on gray-levels in equal distance with interceptive way due to the drawback of histogram equalization with the phenomenon of over-enhancement and gray-levels annexation. The paper intercepts gray-levels using the principle of ＂3σ＂, then, the gray-levels are arranged in equal distance. Finally, the output image is smoothed by median filtering. This method not only enhance image contrast, save image detail, and also overcomes the drawback of gray-level truncation in histogram equalization.