自动化学术论坛[2017第35期]：武汉科技大学柴利教授学术报告会

报告时间：11月18日（星期六）15：40

报告地点：武汉光谷金盾大酒店2楼多功能厅

报 告 人：柴利，武汉科技大学教授

报告题目：Image Restoration with Optimal SSIM Index

内容简介：As the most widely used image quality assessment measure, the structural similarity (SSIM) index has found a variety of applications in image/video processing. In this talk, we consider the optimal image restoration with SSIM criteria. Although the optimization problem is non-convex, we will show that the analytic formula for the optimal equalizer exists when the mean of the source image is zero. The formula shows that the optimal equalizer designed by SSIM criteria and the one by MSE criteria differ only from a scalar factor, which is equal to the inverse of the achievable SSIM index. The formula also establishes the relation of the SSIM performance to the property of the blurring filters and the length of the equalization filters. Various numerical examples are given to demonstrate the effectiveness of the results.

报告人简介：Prof. Li Chai received the B.S. degree in applied mathematics and the M. S. degree in control science and engineering, both from Zhejiang University, China, in 1994 and 1997 respectively, and the Ph.D. degree in electrical and electronics engineering from Hong Kong University of Science and Technology in 2002. In September 2002, he joined Hangzhou Dianzi University, China. He worked as a postdoctoral research fellow at the Monash University, Australia, from May 2004 to June 2006. He was a visiting researcher at Newcastle University in 2009, Central Queensland University in 2011, and Harvard University from July 2015 to June 2016. He received the National Science Fundation for Distinguished Young Scholars of China in 2016. Prof. Chai is the author and co-author of over 60 publications. His research interests include filter bank frames, networked control theory, industrial image processing and graph signal processing.