Citation

Kolarić D, Skala K, Dubravić A. Periodicum biologorum 2007; 110(2): 205-211.

Copyright

(Copyright © 2007)

DOI

unavailable

PMID

unavailable

Abstract

Background and Purpose: Forest fires are a universal problem that both confronts and confounds many countries. Such fires not only destroy large amount of natural resources, but also destroys wildlife and their natural habitat, wreaks general havoc on ecosystems and creates environmental pollution. For all that, fire fighting is one of today's most important matters for natural and environmental resource protection and preservation. Material and Methods: The growing concern regarding environmental devastation of this kind is the underlying reason for the development of modern fire-detection system. Forest-fire automatic detection is a complex problem that involves substantial amount of various sensorial information and data. Furthermore, the reliability of automatic detection systems is still a significant issue in the domain. This paper presents a scheme of an integrated system for early detection of forest fires. The system uses a telecommunication network to link several components for cooperative detection. The perception system is based on a network of sensorial stations and central stations. The sensorial stations collect data including infrared and visual images and meteorological information. The central stations exchange data to perform distributed analysis. An implementation of the system has been carried out at the Ruđer Bošković Institute, Centre for Informatics and Computing. The paper includes some experiments carried out in the Natural Park of island Mljet and Molat (Croatia) (1).

RESULTS: This paper proposes a new distributed intelligent system for reliable forest-fire detection based on the integration of information from several distributed sensors and sources of information.

Language: en