| Paper: | IMDSP-L5.6 | ||
| Session: | Image and Multidimensional Signal Processing: Applications III | ||
| Time: | Thursday, May 20, 17:10 - 17:30 | ||
| Presentation: | Lecture | ||
| Topic: | Image and Multidimensional Signal Processing: Geophysical Signal, Image Processing | ||
| Title: | HYPERSPECTRAL SIGNAL MODELS AND IMPLICATIONS TO MATERIAL DETECTION ALGORITHMS | ||
| Authors: | Dimitris Manolakis; MIT Lincoln Laboratory | ||
| Abstract: | The purpose of this paper is to present a concise overview ofhyperspectral signal models and the target detection algorithmsresulting from their adoption. We focus on detection algorithmsderived using established statistical techniques and whoseperformance is predictable under reasonable assumptions abouthyperspectral imaging data. We show that the family ofelliptically contoured distributions (ECDs), in general, and thet-ECD, in particular, provide a more accurate model forhyperspectral backgrounds, compared to the widely usedmultivariate normal distribution. Since many detection algorithmsderived for normal distributions apply to ECDs as well, the ECDmodels provide a better framework for modeling and analyzinghyperspectral imaging data. | ||
| Back | |||
Home -||-
Organizing Committee -||-
Technical Committee -||-
Technical Program -||-
Plenaries
Paper Submission -||-
Special Sessions -||-
ITT -||-
Paper Review -||-
Exhibits -||-
Tutorials
Information -||-
Registration -||-
Travel Insurance -||-
Housing -||-
Workshops