报告题目：Classification of Non-tumorous Facial Pigmentation Disorders Using SMOTE and Deep Learning

报告人：Zhiping Lin教授（新加坡南洋理工大学）

报告时间：2021年12月16日上午10:00—12:00

报告地点：腾讯会议652-363-681

报告人简介：

Zhiping Lin received the B.Eng. degree in control engineering from the South China Institute of Technology, Guangzhou, China, in 1982, and the Ph.D. degree in information engineering from the University of Cambridge, Cambridge, U.K., in 1987. He worked with the University of Calgary, Calgary, AB, Canada, Shantou University, Shantou, China, and DSO National Laboratories, Singapore, before joining the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, in 1999. His research interests are in statistical and biomedical signal/image processing, and machine learning. Dr. Lin was the Editor-in-Chief of Multidimensional Systems and Signal Processing from 2011 to 2015, and has been in its editorial board since 1993. He was an Associate Editor of IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS—PART II: EXPRESS BRIEFS and the Subject Editor for the Journal of the Franklin Institute. He is the co-author of the 2007 Young Author Best Paper Award from the IEEE SIGNAL PROCESSING SOCIETY. He was a Distinguished Lecturer of the IEEE Circuits and Systems Society (CAS) from 2007 to 2008, and served as the Chair of IEEE CAS Singapore Chapter from 2007 to 2008, and in 2019.

报告摘要：

The diagnosis of non-tumorous facial pigmentation disorders is crucial since facial pigmentations can serve as a health indicator for other more serious diseases. The computer-based classification of non-tumorous facial pigmentation disorders using images / photographs allows automated diagnosis of such disorders. However, the classification performance of existing methods is still not satisfactory due to the limited real-world images available for research. In this talk, we present our research on using deep learning and synthetic minority over-sampling technique (SMOTE). Specifically, both transfer learning and generative adversarial network (GAN) are adopted which provide better classification accuracy than conventional machine learning methods. Moreover, with the application of improved SMOTE, more data is provided to train GAN models. A significant increase in the classification accuracy has been achieved by the proposed method compared to the state-of-the-art methods.