The Design Of Fire Smoke Monitoring System Based On Monitoring Platform

In recent years, the frequent occurrence of fires posed a serious threat to thepeople’s lives and property, while that of the traditional fire detectors have somelimitations in practical applications, the fire can not be fast and accurate judgment ofthe existing fire occurrence, prevention and remedial measures are urgently need to beimproved.This paper makes a deep research on Smoke Flame Detection Algorithm basedon monitoring platform, combines the computer digital image processing technologyand monitoring platform, without changing the existing monitoring platform, andfinally designs a smoke alarm system which both has the accuracy and the real-timeperformance. Results of experiments indicate that the system can produce rapid andaccurate identification of smoke, and alarm automatically.This paper mainly completed the following work:1. Makes a comparison between the traditional fire detectors and fire detectorbase on image processing, mainly discusses the traditional fire detectors limitatio nsand advantages of fire detector base on image processing.2. Reviews all the available theory, useful methods and tools about imageprocessing used in the smoke detection algorithms.3. Introduces the current common algorithms of motion region extraction.4. Introduces various smoke features used in smoke detection algorithm,including color features, morphological characteristics, frequency characteristics,texture characteristics, initial position invariant feature, direction of motion featureand area increase.5. Provides the overall framework of fire smoke monitoring system, and tests theperformance of the system.