Since 1990, many mathematicians studied the image processing based on partial differential equations and variational methods. Using TV(total variation) and some optimization techniques, there were a lot of improvement in image processing areas. Until deep learning methods coming out, those were the state of art methods in these areas. But once using deep learning techniques, it turns out that in almost every areas in image processing fields, deep learning is the one of the best method. I will explain and compare the some results for Wafer defect detection using the traditional image enhancement methods and deep learning methods. Also I will explain a little bit about the networks of deep learning which we are using for detecting the defect.

This presentation is part of Minisymposium “MS24 - Data-driven approaches in imaging science (3 parts)”
organized by: Jae Kyu Choi (Institute of Natural Sciences, Shanghai Jiao Tong University) , Chenglong Bao (Yau Mathematical Sciences Center, Tsinghua University) .

Authors:

Myung Joo Kang (Department of Mathematical Sciences, Seoul National University)

Keywords: