摘要
Retinex理论是颜色恒常知觉的计算理论,可以用于图像清晰度严重失真状况下的图像增强。在研究Retinex算法的基础上,对低照度彩色图像失真中色彩恢复存在的问题进行研究,构建了一个恢复效果较好的色彩恢复函数——余弦色彩恢复函数。给出了处理后图像的自动补偿/增益方法以及补偿/增益中参数实际选取的经验值。对处理后的图像进行了质量评价分析,表明该恢复函数在处理低照度图像时具有较明显的改善效果。
Retinex theory is the color constancy theory of computation which can be used in image enhancement of the severe distortion of the image sharpness. According to the problems of low-light color image distortion color restoration,it is based on the study of Retinex algorithm builds a recovery better color recovery function, cosine color recovery function in this article. It gives the automatic compensation ,gain method of the processed images and the actual selected value of the parameters in the compensation, gain. The quality evaluation and analysis of the processed images shows that the proposed color recovery function in dealing with low-light images has more obvious imorovement effect.
出处
《电视技术》
北大核心
2013年第15期4-6,10,共4页
Video Engineering
