In this paper, digital image processing and fuzzy expert system are combined to study fire detection method. The flame characteristics of fire occurrence and development are studied from the perspective of image processing. Firstly, the moving target of the flame area is extracted; Secondly, four characteristic parameters are selected as evaluation indexes for the flame combustion characteristics, and fuzzy mathematics is used to determine the fuzzy membership function and weight of each evaluation index; Thirdly, the acquired knowledge is expressed in the form of rule programming, and a knowledge base of the fire detection expert system is constructed; Lastly, the analysis and calculation program is written, the test results on a large number of videos are consistent with the actual situation, verifies the effectiveness of the proposed fire detection method.
Paper
Full text
Research on Fire Detection Method Based on Fuzzy Expert System
Semantic Scholar · Computer Science · 2019
Abstract
In this paper, digital image processing and fuzzy expert system are combined to study fire detection method. The flame characteristics of fire occurrence and development are studied from the perspective of image processing. Firstly, the moving target of the flame area is extracted; Secondly, four characteristic parameters are selected as evaluation indexes for the flame combustion characteristics, and fuzzy mathematics is used to determine the fuzzy membership function and weight of each evaluation index; Thirdly, the acquired knowledge is expressed in the form of rule programming, and a knowledge base of the fire detection expert system is constructed; Lastly, the analysis and calculation program is written, the test results on a large number of videos are consistent with the actual situation, verifies the effectiveness of the proposed fire detection method.