【作者】 张仁斌； 赵凯； 李钢； 吴燎原； 杨嘉斌；

【Author】 Zhang Renbin Zhao Kai Li Gang Wu Liaoyuan Yang Jiabin (School of Computer and Information,Hefei University of Technology,Hefei 230009,China)

【机构】 合肥工业大学 计算机与信息学院；

【摘要】 雾天图像存在对比度低,分辨率低的缺点,因此在对雾天图像进行分析之前,通常都需要对雾天图像进行图像增强处理。一些经典的图像增强算法,如直方图均衡化算法,通过拉伸原图像的灰度范围能够明显增强图像的对比度,达到图像增强的效果,但是算法存在局限性,在处理彩色图像时,由于对几个颜色通道的灰度值使用相同的灰度拉伸,会造成图像颜色的严重失真。为了克服传统图像增强算法的局限性,本为设计了一种基于MSR算法的雾天图像增强算法。在处理低对比度,低分辨率的雾天图像时,图像增强效果明显优于传统图像增强方法,并且大大降低图像颜色的失真程度。更多还原

【Abstract】 Fog image has the shortcomings of low contrast and low-resolution.Image enhancement is usually required before the analysis to a fog image.Histogram equalization algorithm is one classical algorithm for image enhancement,which enhances the original image by stretching the range of the gray area to enhance the image contrast.However,when this algorithm is used to process the color image,it will cause a serious distortion of the image because of the same stretching method to each color channels.In order to overcome the limitations of the traditional image enhancement algorithms,this thesis designs a fog image enhancement algorithm based on MSR algorithm.In dealing with the low contrast and low-resolution fog image,the algorithm can achieve a better result than the traditional image enhancement methods.Besides,it is able to reduce the distortion of the image color.更多还原