基于YCbCr色彩空间的图像色偏检测方法

The method of image color cast detection based on YCb Cr color space

提出一种基于YCb Cr色彩空间的图像色偏检测方法。该方法的基本思想是利用图像统计特性,在合适的颜色空间中(本文选用YCb Cr色彩空间),使用简单的统计工具分析图像的颜色分布。在此基础上将图像的色偏分为三类:存在色偏;没有色偏;不可分类。其中不可分类是指图像考虑的部分不到图像整体的20%。该方法既充分利用了图像传感器的成像解析能力,又有效降低了算法的计算复杂性。大量自然图像测试的实验结果表明,该方法具有较好的图像色偏检测效果和较好的计算效率。

A method of image color cast detection based on YCbCr color space is put forwared, The basic idea of this method is using the image statistical characteristics, in the appropriate color space （the paper chosen YCbCr color space）, to analyze the image color distribution by using simple statistical tools. On this basis, the image is divided into three interest groups： there is a color cast, no color cast and cannot be classified, where cannot be classified means that the considering part of the image is less than 20% of the overall image. This method not only takes full advantage of the imaging resolution of image sensor, but also reduces the computational complexity of the algorithm. Experimental results on a large number of natural images show that, the method has better effect on image color cast detection and good computational efficiency.