Stereo Image Quality Assessment And Human Visual System

With the rapid development of stereoscopic technologies,the applications of 3D technology are emerging fast.The rapid development of Internet and Multimedia technologies also impelled this trend.However,researches on stereoscopic image quality assessment can' t catch up with applications An acceptable metric for stereoscopic image quality assessment will undoubtedly promote the development of stereoscopic technologies and applications.Hence,it is very important to develop an effective quality assessment model for stereoscopic images that is accordance with the human visual system(HVS).To solve the above problem,in this paper a great effort in researching some stereo vision related characteristics of human visual system has been made.Based on these characteristics,two effective stereo image quality assessment metric have been proposed.In method one,a full-reference metric for quality assessment of stereoscopic images is proposed based on the binocular difference channel and binocular summation channel.For a stereo pair,the binocular summation map and binocular difference map are computed first by adding and subtracting the left image and right image.The quality of binocular summation map can be regarded as the image quality while the quality of binocular difference map can be deemed as the stereo sensing quality.Then the binocular summation is decoupled into two parts,namely additive impairments and detail losses.The quality of binocular summation is obtained as the adaptive combination of the quality of detail losses and additive impairments.The quality of binocular summation is computed by using the Contrast Sensitivity Function(CSF)and weighted multi-scale(MS-SSIM).Finally,the quality of binocular summation and binocular difference is integrated into an overall quality index.In method two,we we propose a framework to evaluate the quality of stereoscopic images that are afflicted by possible symmetric distortions or asymmetric distortions.In the framework,binocular combination models and visual saliency are thought to be the key factors for the stereoscopic image quality assessment metric.Like the concept of cyclopean image,a saliency map for cyclopean image called ”cyclopean saliency”is proposed.In this frame work,cy-clopean image and cyclopean saliency is computed at first,then the cyclopean image is weighted by the cyclopean saliency.The quality of stereo pair is obtained by evaluating the weighted cyclopean image with a 2D quality assessment metric.Both of the proposed metric can precisely predict the distorted quality of stereo pair.Noticing that the proposed metric 2 is applicable for both symmetric and asymmetric distortions and performs well,we can conclude that the proposed metric 2 is a strong applicable tool for stereoscopic image quality prediction.