Research Of High Dynamical Range Visual Quality Metric

In recent years,with the rapid development and extensive use of digital image technology and broadband network,as well as hardware ultra-large scale integrated circuits,high-speed chip and so on,all kinds of video communication technology and video processing technology is widely used in our lives,such as video telephony,video conferencing,etc.which are based on video applications.So there are urge need for high-quality video quality assessment technology to simulate the video communication users and to mimic consumers feelings of the video subjective to assess the video quality.Video quality assessment plays a very important role in video processing and video communications area.Video quality can visually reflect the performance of various video systems and the quality of service(QoS)of the channel transmission system.Therefore,the system can adjust the channel parameters and improve the service quality through the evaluation results.With the improvement and development of video quality assessment technology,it makes the image display system in line with the human visual characteristics.For hardware equipment manufacturers and a variety of multimedia applications,they are a convenient way to assess.With the optimization of display devices and the increasing demand for photos and videos,high dynamic range images and video can be used as a high dynamic range in photographic films,video games,medical images,aerospace and other fields which can represent real scenes that have been widely used.Since the dynamic range is very limited,traditional video quality assessment technology often can not get accurate results in the assessment of high dynamic range video and far from the user subjective feelings.Therefore,this paper proposes a high dynamic range video quality evaluation method which can both meet the requirements of standard dynamic range and high dynamic range,and propose a video with similar visual characteristics quality assessment model.This paper starts from modeling the main features that affect the quality of video,and introduces the related features and human visual characteristics.Then,an existing high dynamic range video quality evaluation model is introduced,then the analysis how to make the improvement.Finally,according to the existing theoretical research,A new algorithm for evaluating multiple features the method has been improved on the basis of the existing methods.The innovation of this algorithm lies in the evaluation of multiple features and the evaluation of the characteristics of the human visual non-linearity,so that the evaluation results are more suitable for the subjective feelings of the human eye.In addition,the algorithm proposed in this paper also uses the significance graph obtained by the saliency model as the weight to improve the accuracy.The high dynamic range video quality evaluation model proposed in this paper uses the subjective and objective contrast experiments to verify the performance of the model.The results show that the model proposed in this paper has superior performance.