Share Email Print

Proceedings Paper

Detectability and simulation of smoke and dust clouds in images
Author(s): Donald W. Hoock; John C. Giever
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Smoke and dust clouds are often more easily detected than the sources producing the clouds. Long range detection depends on the cloud's size and contrast against its background. However, a cloud can sometimes be detected, even in low average contrast conditions, if it significantly masks the apparent clutter of its background or if it provides sufficiently different clutter itself. This paper investigates a basic cloud detection method that includes both cloud contrast and background clutter. It is assumed that the sensor minimum resolvable contrast (or thermal temperature difference) as a function of spatial frequencies is known. This is used to define a detection threshold. We consider contrast in mean intensity, changes in the width of the background intensity histogram, changes in edge strength, and comparison of spatial frequency content as methods to detect the cloud relative to its background. Methods are examined using measured and simulated images.

Paper Details

Date Published: 1 September 1990
PDF: 12 pages
Proc. SPIE 1312, Propagation Engineering: Third in a Series, (1 September 1990); doi: 10.1117/12.21888
Show Author Affiliations
Donald W. Hoock, U.S. Army Atmospheric Sciences Lab. (United States)
John C. Giever, New Mexico State Univ. (United States)

Published in SPIE Proceedings Vol. 1312:
Propagation Engineering: Third in a Series
Luc R. Bissonnette; Walter B. Miller, Editor(s)

© SPIE. Terms of Use
Back to Top