快速自适应图像颜色恒常性增强方法

摘要

To deal with the large computation amount when the red,green,blue (RGB) color space is needed to be transformed into hue,saturation,value (HSV) color space for the existing methods,a fast and adaptive image color constancy enhancement algorithm was proposed on the basis of analyzing the existing color constancy algorithms and their implementations,which followed the principle of maintaining constant hue and properly changing saturation.The principle of saturation enhancement in RGB color space was analyzed,and the color constancy enhancement method without color space transformation was proposed.The parameters of method were adaptively calculated,and the final enhanced image was obtained.The visual and quantitative evaluation results of experiments using actual images show that compared with the conventional method,the proposed algorithm can not only maintain brightness,definition and hue,but also reasonably change the saturation component,so as to maintain good color constancy and details,and achieve better visual effects.Since there is no need for color space transformation,the proposed method is faster.%针对现有颜色恒常性增强方法大多在进行色彩空间转换时计算量较大的问题,在分析颜色恒常性理论的基础上,遵循保持色调不变而合理改变饱和度的原则,提出一种快速自适应图像颜色恒常性增强方法.对RGB空间中进行饱和度增强的原理进行分析,提出无需HSV色彩空间变换,直接在RGB空间中进行增强的颜色恒常性增强方法,对方法中的参数进行自适应计算,得到增强图像.利用实际拍摄的图像进行实验,视觉和定量评价结果表明,与现有增强方法相比,该算法在保持亮度、清晰度和色调等信息基本不变的同时,能合理地改变饱和度分量,保持良好的颜色恒常性和丰富的细节信息,取得较好的可视效果;由于无需色彩空间变换,该方法能够取得更快的计算速度.