Microscene evaluation using the Bhattacharyya distance

William F. Basener; Marty Flynn

doi:10.1117/12.2327004

23 October 2018 Microscene evaluation using the Bhattacharyya distance

William F. Basener, Marty Flynn

Proceedings Volume 10780, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications VII; 107800R (2018) https://doi.org/10.1117/12.2327004
Event: SPIE Asia-Pacific Remote Sensing, 2018, Honolulu, Hawaii, United States

Abstract

A microscene is a hyperspectral image collected using a hyperspectral sensor mounted above a tray, typically in a laboratory setting. Materials can be placed in the tray and illumination controlled to either analyze the materials used or to simulate overhead (aerial or satellite) imagery. Choosing the materials allows simulation of overhead imagery in controlled experiments, for example mixtures and abundances of chemicals, materials as they undergo physical and chemical processes such as oxidation and weathering, and vegetation at different stages in environmental processes. Microscene imagery enables experiments in controlled circumstances not easily producible in overhead imagery. Moreover, the cost of collecting microscene imagery is a small fraction of overhead collection. Microscene imagery is an emerging technology, and in this paper we address an evaluation microscene imagery to determine how well it simulates overhead imagery, comparing microscene imagery of vegetation to overhead AVIRIS and HYDICE imagery over vegetation. We use statistical measures to compare microscene imagery to overhead imagery, including comparing material spectra, means, eigenvalues, the Mahalanobis distance between image means, and for the first time the Bhattacharyya distance between image covariances. The Bhattacharyya is a statistical measure of the distance between two statistical distributions, related to the Mahalanobis distance.

Citation Download Citation

William F. Basener and Marty Flynn "Microscene evaluation using the Bhattacharyya distance", Proc. SPIE 10780, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications VII, 107800R (23 October 2018); https://doi.org/10.1117/12.2327004

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