基于镜头特征的动态调光算法

A Backlight Dimming Algorithm Based on Shot Characteristics

视频是由若干个镜头联合构成的,不同的镜头在亮度和主灰度方面存在差异。为了有效提升对于不同类型视频的节能率,提出了一种基于镜头特征的动态调光算法。该算法运用亮度和主灰度等参数进行镜头分割,然后依据每个镜头的亮度对比度进行分类,针对不同类型镜头设置不同的调光参数进行背光调节。对该算法使用不同类型视频进行了仿真,结果表明镜头的平均检出率达到67.21%,平均精确度达到81.81%。开发了工程样机对不同类型视频进行了节能率测试,整机平均节能率达到24.89%。

The video is composed of many shots, which may have different brightness and main gray level. To effectively increase energy saving rate of different video, a bacldight dimming is proposed. The shot segmentation is implemented by using brightness and algorithm based on shot characteristics main gray level. Then classification is performed for each shot based on its brightness and contrast. Different dimming parameters of backlight control are applied for different type of shots. The simulation results for different kinds of video show that the average detection rate is 67.21% and the average precision is 81.81%. An engineering prototype was developed to test different kinds of video. The result shows that the overall average energy saving rate is 24.89%.