Video/Image Quality Evaluation Algorithm Research And Application

Nowadays people like to receiving information in the form of video orimage with the popularity of smart phones and tablet which led to growingdata traffic. The speed of expanding Wired or wireless bandwidth is muchslower than the growth rate of the demand from customers. The bandwidthproportion occupied by the flow of video and image data makes it necessaryto optimize the content bandwidth. And how the network can give bettervideo subjective feelings in a smaller bandwidth by designing videoencoding parameters is urgent study.This paper focuses on two aspects of the video and image qualityassessment: subjective and objective. On subjective side, we design testplan which covers different platforms and propose recommending bitratebased on different cases. On objective side, first we apply distortion basedmethod to mobile application according to real circumstance. Also, we doresearch on natural scene statistics based feature and find a novel way toform the feature which leads to good result.Video and image quality evaluation can optimize the network contentencoding parameters, and ease the contradiction between the growingdemand and limited bandwidth network video.