Abstract
Conference Title: 2013 International Conference on IT Convergence and Security (ICITCS) Conference Start Date: 2013, Dec. 16 Conference End Date: 2013, Dec. 18 Conference Location: Macao Fires are one of the main causes of death amongst the world. Although there are various fire detection systems most of them did not proof its effectiveness in detecting fires due to inefficiency or restrictions. For example, systems based on smoke sensors in detecting fires cannot be used in open areas because they are restricted to the existence of a ceiling or a wall. Besides that, having many flammable objects in open large areas, such as trees and fuel of car may interact with oxygen in the air which make fire growth and expansion very fast. According to the prior stated facts in this paper we present the project we worked on with the aim of implementing a more efficient and trustworthy fire detecting system. The project was done by using computer vision and image processing techniques to detect fire flames based on studying the fire properties besides an alarm notification system. [PUBLICATION ABSTRACT]