Video Quality Assessment Based on Edge Structural Similarity

Measurement of visual quality is of fundamental importance to many image and video processing applications. Traditionally, quality assessment metrics predict visual quality following the paradigm of modeling the known properties of the Human Visual System (HVS). The Structural Similarity (SSIM) index is a new approach which assumes that the HVS is highly adapted for extracting structural information from a scene and has been proved outperforms PSNR and many state-of-the-art HVS-based metrics. In this paper, we present a new metric for video quality assessment which selects local sampling regions according to their edginess and uses the SSIM index as local distortion measurement. We validate the performance of our metric by testing it on the VQEG Phase I dataset, and showed that our metric has a good correlation with perceived visual quality.