Lecture on Introduction to hyperspectral data with focus on Spectral Unmixing
Date and Time
- Date: 11 Dec 2017
- Time: 11:00 AM to 01:00 PM
- All times are (UTC+05:30) Chennai
- Add Event to Calendar
- A-101, A-Block
- NIRMA University
- Ahmedabad , Gujarat
- India 382481
- Building: Institute of Technolgy
- Click here for Map
- Contact Event Host
- Co-sponsored by NIRMA University
Introduction to hyperspectral data with focus on Spectral Unmixing
Hyperspectral remote sensing analysis is an exciting avenue of research. It bring to the table a huge volume of data and there by a variety of techniques for exploitation. Hyperspectral unmixing is an important task for remotely sensed hyperspectral data exploitation. It amounts at finding the spectrally pure constituents in the scene (called endmembers in hyperspectral imaging terminology) and their fractional abundances on a sub-pixel level. Spectral unmixing allows for a detailed analysis of hyperspectral images with sub-pixel precision. Research in spectral unmixing has evolved significantly in the last few years, from the first efforts focused on linear spectral unmixing (assuming linear interactions between the endmembers) in which techniques assumed the presence of pure spectral endmembers in the data, to a current scenario in which most unmixing techniques assume that pure spectral signatures may not be present in the image scene due to spatial resolution and other phenomena. In this talk, we begin with introducing the hyperspectral data, start with the basics of unmixing and progress through the variety of methods of unmixing to conclude with some of the state of the art techniques.
Sumit Chakravarty currently works as an Assistant Professor of Electrical Engineering with Kennesaw State University. He has completed his doctoral studies from University of Maryland, Baltimore County and his Master of Science from Texas A&M University, Kingsville, both in Electrical Engineering. His PhD dissertation is on analysis of Hyperspectral Signatures and Data Analysis. He was a Post doc researcher with Section on Biomedical Image Analysis at UPENN. He has utlized his expertise in remote sensing working as a scientist in industry wherein he also worked on sensor modalities like Multispectral, Lidar and SAR. This includes working with a cross disciplinary team comprising of specialists from industry, government bodies like NASA, Goddard and academia like University of Maryland.
His other industrial experience in engineering and research include working in various roles such as Instrumentation Engineer for Hyundai, Research Intern at Siemens CAD and Apex Eclipse Communications, Scientist for SGT Inc and Lead Scientist for Honeywell Research (Automatic Control Solutions-Advanced Technology Labs).
He has multiple peer-reviewed journal publications, conference publications, a book chapter and three granted patents (and one under review) besides the current work under progress. Additional information: http://facultyweb.kennesaw.edu/schakra2/
Address:Kennesaw State University, , GA, United States, 30144