The design and implementation of fire smoke detection system based on FPGA

摘要

This paper uses modular method to design smoke detection system. On the basis of the requirement analysis, the paper designs and implements all the modules, including image capture module, SDRAM data buffer module, image display module and smoke detection module. Among them the smoke detection module is the core of the system, which decides the performance of the system. Using a synergistic manner through software and hardware, the smoke detection module is realized. First of all, using the algorithm of frame difference, erosion and dilation, it realizes suspected smoke area extraction with hardware description language. This method ensures to process real-time sequential image. And then we detect the suspected smoke areas with smoke dynamic features including smoke area rising feature and smoke regional stability feature. Finally, the paper builds smoke detect module using SOPC Builder, which reads image data captured by CMOS image sensor monitors into SRAM. Based on smoke detect hardware module this paper realizes smoke background fuzzy characteristics based on the 2d discrete wavelet transform in Nios II IDE environment. After a period of debugging and running, the system realizes real-time smoke detection.