Dissertation Defense: Deep Learning-based Image Restoration and Synthesis in Biomedical and Natural Image Processing

Date: 2023/05/18 - 2023/05/18

Dissertation Title: Deep Learning-based Image Restoration and Synthesis in Biomedical and Natural Image Processing

Speaker: Da He, Ph.D. candidate at UM-SJTU Joint Institute

Time: May 18th from 12:00 a.m., 2023 (Beijing Time)

Location: Room 403, Longbin Building

Abstract

Various image degradation factors and limitations may happen to numerous imaging systems and thereby influencing various image-based applications. Therefore, a semi-deep learning method was firstly investigated to restore the out-of-focus fluorescence microscopy images, while the end-to-end deep learning method was studied to improve the imaging quality of photoacoustic microscopy. After discussing well-known degradation tasks, a new image degradation about adherent mist and raindrops was proposed, defined, and handled for natural images in the daily life. In addition, to alleviate the quantity limitation of magnetic resonance imaging data, a conditional image synthesis pipeline was explored to benefit high-level clinical applications.

Biography

Da He received his B.S. degree in Optoelectronic Information Science and Engineering from Nankai University. In 2018, he joined the University of Michigan-Shanghai Jiao Tong University Joint Institute, Shanghai Jiao Tong University, Shanghai, China, as a graduate student. He is interested in biomedical image processing and computer vision.