Share Email Print

Proceedings Paper

Surveillance sensor for autonomous wildfire detection
Author(s): Jan S. de Vries; Rob A. W. Kemp
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Autonomous wildfire detection systems may help to reduce hazards resulting from large wildland fires. In many situations wildfires start in the duff below trees and shrubs, which are hidden from direct view by groundbased sensors overlooking forests and wildlands. Mid- and thermal infrared measurements only detect wildfires when the fire has become a crownfire, and, by then, it usually has developed into a large wildfire. Therefore, the early discovery of wildfires using groundbased, autonomous sensors should be performed by detecting smoke clouds rather than the heat of the fire, since smoke becomes earlier visible above the trees as a result of convection than the heat of the fire. A demonstration sensor is being developed to show the feasibility of an affordable system for autonomous wildland fire detection. The system is designed to minimize false alarms by simultaneously analyzing the temporal, spatial and spectral information in the acquired imagery. The groundbased sensor will be horizon scanning and will employ linear CCD's for better contrast sensitivity in three different spectral bands.

Paper Details

Date Published: 15 November 1993
PDF: 7 pages
Proc. SPIE 1952, Surveillance Technologies and Imaging Components, (15 November 1993); doi: 10.1117/12.161400
Show Author Affiliations
Jan S. de Vries, TNO Physics and Electronics Lab. (Netherlands)
Rob A. W. Kemp, TNO Physics and Electronics Lab. (Netherlands)

Published in SPIE Proceedings Vol. 1952:
Surveillance Technologies and Imaging Components
Sankaran Gowrinathan; C. Bruce Johnson; James F. Shanley, Editor(s)

© SPIE. Terms of Use
Back to Top