数字图像颜色恒常性算法及应用  

Algorithm of Image Color Constancy and Its Application

作　　者：曹生现[1] 李国英[1] 王延红[1] 郑丽婷[1] 

机构地区：[1]东北电力大学自动化工程学院,吉林吉林132012

出　　处：《化工自动化及仪表》2013年第4期467-470,共4页Control and Instruments in Chemical Industry

摘　　要：建立了BP神经网络颜色恒常性算法,对水质氯离子溶液颜色值进行了校正,以提高图像测量精度。实验结果表明:BP神经网络颜色恒常性算法实现了普通光源到标准光源的非线性映射,可以减小或消除电源及光源等因素对测量结果的影响,校正了颜色,提高了测量精度和灵敏度。The BP neural network algorithm for the color constancy was established and the color value of chloride solution in water were corrected to well improve the images＇ detection accuracy and sensitivity.The experimental results show that the BP neural network algorithm for the color constancy can realize nonlinear mapping from a normal light to a standard light and can reduce or eliminate the impact of the power,light and other factors on the measurement results,as well as correct the color.

关 键 词：颜色恒常性 图像检测 颜色校正 氯离子 

分 类 号：TH74[机械工程—光学工程]