No-reference image quality assessment based on svm for video conferencing system*

Abstract

Traditional image quality assessment method is mainly subjective assessment method, that is to say, the human eye is used to observe and judge the quality of the image. This paper proposes a no-reference image quality assessment method and introduces the support vector machine (SVM) into image quality assessment of the video conferencing system. A SVM classifier is employed to distinguish the image sequences of poor quality from the good ones. The experimental results show that the proposed method has good performance. © Springer-Verlag Berlin Heidelberg 2012.