期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2014
卷号：14
期号：12
页码：9-15
出版社：International Journal of Computer Science and Network Security
摘要：In this paper, we have conducted an analysis of the influence of image contents on facial muscle activity for image quality assessment of JPEG coded image, using multiple regression analysis. In previous studies, we have successfully established a relationship between facial muscle activity and the subjective assessment score of image quality, by using Facial Electromyography (fEMG). Measurements were taken through fEMG and compared to subjective assessments conducted through questionnaires. We measured six relevant facial muscles in our experiments. With this paper, we are advancing the model a step further, by introducing a new set of variables, which focus on the contents represented in the images, and the different features represented, compared to the previous general look at all images as equal input sets. For this purpose, we have used the entropy information in order to differentiate image contents and improve the image quality prediction model
