基于直方图和主灰度的调光算法  

Backlight dimming algorithm based on histogram and dominant gray level

作　　者：张磊[1,2] 李亚妮[1,3] 陆吕晨[1,4] 吕国强 

机构地区：[1]合肥工业大学特种显示技术教育部重点实验室特种显示技术国家工程实验室现代显示技术省部共建国家重点实验室,安徽合肥230009 [2]合肥工业大学计算机与信息学院,安徽合肥230009 [3]合肥工业大学光电技术研究院,安徽合肥230009 [4]合肥工业大学仪器科学与光电工程学院,安徽合肥230009

出　　处：《液晶与显示》2014年第6期1090-1095,共6页Chinese Journal of Liquid Crystals and Displays

基　　金：安徽省2012科技计划资助项目(No.1206c0805007)

摘　　要：为了有效解决液晶显示中背光能耗与显示质量的矛盾,在综合分析了灰度直方图和主灰度等图像特征与图像背光关系的基础上,提出了一种基于这些特征的调光算法。首先统计输入图像像素灰度的平均值,并依据平均值将图像划分为3种类型;然后统计图像的灰度直方图和主灰度等特征值;最后采用多元回归分析方法分析图像特征与背光之间的关系从而确定背光公式系数,再依据公式计算背光调光亮度。对该算法使用30幅典型图像进行了仿真测试,实验结果表明调光效果指数DEI为0.95,平均节能率为28.22%。与其他算法相比,本文算法具有较高的节能率和较好的显示效果,较好地平衡了显示效果和节能率的矛盾。To solve the contradiction between energy consumption and display quality in the LCD display effectively,on the basis of a comprehensive analysis of the relationship between image backlight and features such as gray histogram or dominant gray level,this paper proposes the dimming algorithm with these characteristics.First of all,a classification into three classes based on the averagegray is performed for the input image.Then image characteristic values are computed such as gray histogram and dominant gray level.Finally the backlight dimming brightness is calculated according to formula,whose coefficients are determined by the multiple regression analysis to analyze the relationship between image features and backlight.For 30 typical images,the simulation result show that the dimming effect index is 0.95,the average energy saving rate is 28.22%.Compared with other dimming algorithms,this algorithm has higher power-saving rate and better display effect,moreover,it trades off the contradiction of the display effect and energy saving rate.

关 键 词：调光 背光 主灰度 多元回归分析 DEI 

分 类 号：TN141.9[电子电信—物理电子学] TN312.8