A Novel Idea for Objective Image and Video Quality Measurement

Mr. Zhou Wang
Dept. of Electrical and Computer Engineering
The University of Texas at Austin

Friday, October 5th, 3:00 PM, ENS 637

zwang@mail.utexas.edu

Abstract

Objective image and video quality measures play important roles in almost all kinds of image and video processing applications, such as image and video coding, analysis, restoration, enhancement, halftoning and registration. Traditional methods have used peak signal-to-noise ratio (PSNR) and error sensitivity-based human visual system models. Our new philosophy in designing image quality metrics is:

The main function of the human eyes is to extract structural information from the viewing field, and the human visual system is highly adapted for this purpose. Therefore, a measurement of structural distortion would be a good approximation of perceived image distortion.

A simple and effective implementation of this idea indicates that it can provide image quality metrics significantly better than PSNR. Demo images and free software are available at

http://anchovy.ece.utexas.edu/~zwang/research/quality_index/demo.html

We also developed a new video quality measurement algorithm. Experimenting on the video quality experts group (VQEG) test data set shows that the new quality metric has better correlation with perceived video quality than all the VQEG proponents. We expect that this research will have impacts on computational vision science as well as image processing and computer vision applications.

Biography

Zhou Wang is currently a Ph.D. candidate and Research Assistant at the Laboratory for Image and Video Engineering (LIVE) in the Department of Electrical and Computer Engineering at the University of Texas at Austin. His thesis topic is about rate scalable image and video communications considering human visual system features. He has a broad research interest including image and video processing, coding, communication and quality assessment, computer vision, wavelets, fractals, fuzzy technologies, and artificial neural networks.

A list of Telecommunications and Signal Processing Seminars is available at from the ECE department Web pages under "Seminars". The Web address for the Telecommunications and Signal Processing Seminars is http://signal.ece.utexas.edu/seminars