针对直方图均衡化直接对彩色图像处理会导致图像色彩失真的情况，在对传统的直方图均衡化方法进行改进的基础上，提出了一种新的带色彩恢复的均衡化算法。该算法对于彩色图像R,G,B各个分量子图的灰度直方图，首先根据其灰度中值和分割直方图等面积原则进行两次分割，同时对分割后的各子灰度直方图分别进行均衡化处理；然后通过计算R，G，B各分量子图的灰度级总数占原彩色图像灰度级总数的比例，将均衡化处理后的R，G，B各分量子图进行合并。实验表明，该算法对彩色图像处理具有较好的增强处理效果。

A new method is proposed in this paper to improve the traditional histogram equalization, mainly to solve the problem of color distortion of the color images caused by the direct usage of histogram equalization. This algorithm firstly divides the histograms of three channels of R, G, and B twice based on the mean of the grey levels and the principle of equal area. Then it equalizes the sub-histogram respectively. Finally, the processed images of each channel are merged by calculating the proportion of the three channels in the original image. The experiments indicate that the proposed method is quite effective in processing color images.