Share Email Print

Proceedings Paper

Correcting Atmospheric Effects In Multispectral Images
Author(s): Mohan M. Trivedi; Shi-Chun Zhou; Mark L. Farley
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In computer vision systems images are the primary inputs. In order to develop a robust vision system one needs to derive a correlation between the intensity values recorded in the images and the surface properties of various objects appearing in the scene. Image intensity values recorded in various spectral bands of a sensor are produced as a final result of a complex matter-energy interaction mechanism governing the illuminating source, the scene, the sensor and the intervening medium. In passive multispectral aerial images the illuminating source is the sun and the atmospheric column appears as the intervening medium between the sun and the scene for the incident beam and between the scene and the sensor for the reflected or emitted beam. In this paper an approach for compensating for the atmospheric transmitance and path radiance effects applicable for the reflective optical wavelengths is developed. It utilizes the LOWTRAN-6 package for calculating the required correction parameters. Applicability of this approach is demonstrated by correcting high resolution multispectral images.

Paper Details

Date Published: 16 September 1987
PDF: 6 pages
Proc. SPIE 0781, Infrared Image Processing and Enhancement, (16 September 1987); doi: 10.1117/12.940543
Show Author Affiliations
Mohan M. Trivedi, The University of Tennessee (United States)
Shi-Chun Zhou, The University of Tennessee (United States)
Mark L. Farley, The University of Tennessee (United States)

Published in SPIE Proceedings Vol. 0781:
Infrared Image Processing and Enhancement
Marshall R. Weathersby, Editor(s)

© SPIE. Terms of Use
Back to Top