Analysis And Design Of Intelligent Building Fire Alarm System

Becomes more and more popular with the development of science and technology innovation intelligence,intelligent about basic facilities of People's Daily life and in the rapid development of the intelligent building fire alarm system is one of the embodiment.To improve the detection efficiency,sensitivity and reliability of fire alarm system,to achieve early detection and alarm,fire has a certain practical value.This thesis mainly studies the principle of the composition of automatic fire alarm system in intelligent building,this paper introduces the system about the basic concept,basic structure,basic performance.Analysis of data and information recognition and digital image processing techniques to identify research,on the basis of the research is mainly focused on the TuXiangXing information identification analysis method of the scene of the fire.Filtering through the image pre-processing,segmentation,optimization,eliminate noise,to obtain optimized image samples,through to the flame area changes size,shape and edge characteristic information extraction and detection,such as a series of fire recognition experiment,the simulation technology,test and verify the reliability of fire information recognition algorithm is effective.This paper introduces the BP neural network in the scene of the fire image detection algorithm,combined with the characteristics of the algorithm and the related mathematical model of the function,to build the experiment,the design of the experiment are given specific input and output unit and specific topological structure of neural network.To a large number of fire image samples and interference image sample has carried on the related control experiment.By the result of experiment shows that the fire alarm system based on BP neural network algorithm has more obvious advantages compared with the conventional fire alarm system,greatly reduced the rate of false positives of fire,improves the precision of fire fire alarm,the future can be widely used in modern intelligent building complex.