Geometry of statistical target detection

William F. Basener; Brian Allen; Kristen Bretney

doi:10.1117/1.JRS.11.015012

21 February 2017 Geometry of statistical target detection

William F. Basener, Brian Allen, Kristen Bretney

Author Affiliations +

Journal of Applied Remote Sensing, Vol. 11, Issue 1, 015012 (February 2017). https://doi.org/10.1117/1.JRS.11.015012

Abstract

This paper presents an investigation into the underlying geometry and performance of various statistical target detection algorithms for hyperspectral imagery, presents results from algorithm testing, and investigates general trends and observable principles for understanding performance. Over the variety of detection algorithms, there is no universally best performing algorithm. In our test, often top performing algorithms on one class of targets obtain mediocre results on another class of targets. However, there are two clear trends: quadratic detectors such as ACE generally performed better than linear ones especially for subpixel targets (our top 15 scoring algorithms were quadratic detectors), and using anomaly detection to prescreen image spectra improved the performance of the quadratic detectors (8 of our top 9 scoring algorithms using anomaly prescreening). We also demonstrate that simple combinations of detection algorithms can outperform single algorithms in practice. In our derivation of detection algorithms, we provide exposition on the underlying mathematical geometry of the algorithms. That geometry is then used to investigate differences in algorithm performance. Tests are conducted using imagery and targets freely available online. The imagery was acquired over Cooke City, Montana, a small town near Yellowstone National Park, using the HyMap V/NIR/SWIR sensor with 126 spectral bands. There are three vehicle and four fabric targets located in the town and surrounding area.

Citation Download Citation

William F. Basener, Brian Allen, and Kristen Bretney "Geometry of statistical target detection," Journal of Applied Remote Sensing 11(1), 015012 (21 February 2017). https://doi.org/10.1117/1.JRS.11.015012

Received: 5 September 2016; Accepted: 26 January 2017; Published: 21 February 2017

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available