Hyperspectral Image Denoising and Inpainting based on Low-Rank Representation
Find out more about subscribing to add all events.
Hyperspectral Image Denoising and Inpainting based on Low-Rank Representation
Fri, 02/11/2018 - 10:00
Location:
Speaker:
Dan Yao
Affiliation:
Heriot-Watt University - ISSS
Synopsis:
This talk presents a low-rank representation based framework for solving inverse problems in hyperspectral imaging (denoising and inpainting). We propose one new HSI denoising algorithm and one new HSI inpanting algorithm under this framework, which fully exploit the compact HSI representation linked with low-rank and self-similarity characteristics. The denoising algorithm extends its application in hyperspectral mineral mapping as a pre-processing step, while the inpainting algorithm puts a sharp focus on the visible near infrared (VNIR) wavebands data restoration for China's first prototype space station Tiangong-1 (TG01) Target Vehicle.
Biography:
Dan Yao is currently a first year PhD student in Signal and Image processing. Prior to joining Heriot-Watt University, she studied Electronics and Communication Engineering as a master student in Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, where she obtained her M.Eng degree this June. Her PhD research focuses on developing new computational methods for low illumination multidimensional imaging and sensing.