公路隧道智能火灾检测系统设计

Design of Intelligent Fire Detection System for Highway Tunnel

随着中国基础建设的不断推进,高速公路覆盖范围不断扩大,隧道数量也越来越多,公路隧道发生事故导致的后果与影响往往较大,是极具破坏性和危险性的。利用深度学习方法,该文设计了一个基于树莓派的智能公路隧道火灾监测报警系统。该系统采用树莓派开发板和Intel Movidius神经计算棒(neural compute stick,NCS)为硬件平台,连接视频监控摄像头、烟雾传感器、声光报警器和4G通信模块组成公路隧道火灾检测硬件系统,运用训练好的基于卷积神经网络(CNN)的隧道火灾图像识别模型,对隧道交通场景进行实时检测。系统通过监控摄像头和烟雾传感器,同时检测隧道现场交通情况,能够及早准确识别火灾的发生,并实现即时现场报警与远程报警。测试结果表明,在树莓派和神经计算棒的终端平台上运行深度学习的火灾检测算法,火灾识别精度达到96%,速度到达每秒5帧。应用该系统在发生隧道火灾时对避免人员伤亡、降低财产损失具有重要意义。

With the continuous advancement of infrastructure construction in China,the coverage of the highway is expanding,and the number of tunnels is also increasing.The consequences and impact of highway tunnel accidents are often large,which are highly destructive and dangerous.By using deep learning method,an intelligent highway tunnel fire monitoring and alarm system based on raspberry pie is designed.This system uses raspberry pi and Intel Movidius neural compute stick(NCS)as the hardware platform,connects video surveillance camera,smoke sensor,sound and light alarm and 4G communication module to form the highway tunnel fire detection hardware system,and uses the trained tunnel fire image recognition model based on convolution neural network(CNN)to detect the tunnel traffic in real time.By monitoring cameras and smoke sensors,the system can detect the traffic condition of the tunnel at the same time,identify the occurrence of fire as soon as possible,and realize real-time on-site alarm and remote alarm.The test shows that the fire detection algorithm based on deep learning is run on the terminal platform of raspberry pi and NCS.The fire identification accuracy reaches 96%and the speed reaches 5 frames per second.The application of the system in the case of tunnel fire is of great significance to avoid casualties and reduce property losses.