Optimal design of IR sensor's spectral bands for material classification

The spectral matched filtering is based upon the principle of forming a superposition of basis functions such as a set of sensor's spectral bands to perform target classification. We report a band-engineering routine, which is critical to find the optimal design setting for each sensor band to...

Full description

Saved in:
Bibliographic Details
Main Authors: Woo-Yong Jang, Zahyun Ku, Park, James, Urbas, Augustine, Noyola, Michael
Format: Conference Proceeding
Language:English
Subjects:
Amplitude modulation
Feature extraction
Matched filters
material classification
Minerals
Reflectivity
sensor band
Shape
spectral matched filtering
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The spectral matched filtering is based upon the principle of forming a superposition of basis functions such as a set of sensor's spectral bands to perform target classification. We report a band-engineering routine, which is critical to find the optimal design setting for each sensor band to improve the estimation of target's signature such as reflectance of material. The quality of engineered sensor's spectral bands is validated as they apply to the various material classification problems.
ISSN:2379-2027
DOI:10.1109/NAECON.2016.7856814