Research On The Key Technology Of Smoky Vehicle Monitoring System

At present,the regulation measures of motor vehicle exhaust pollution mainly include periodic environmental inspection and road sampling inspection.It is of great practical significance to monitor the black smoke of motor vehicles on urban roads.In order to detect the black smoke of moving vehicles without affecting the traffic,this paper presents a vehicle black smoke monitoring system based on surveillance video and laser telemetry.The system mainly consists of two key technologies,namely the vehicle black smoke identification technology based on surveillance video and the smoke plume impermeability measurement technology based on laser telemetry.The main work of this paper is as follows:An eight-channel smoke plume impenetrability measurement method is proposed.Based on Lambert-Beer law,the method uses the national standard 560 nm green laser as the light source and uses an eight-point fiber beam splitter to obtain eight parallel beams,which enhances the stability and reliability of the system.The system was calibrated by using the standard filter,and the experimental data were fitted by using the least square fitting method.The accuracy and stability of the system are tested,and the system error is not more than 2%,which verifies the feasibility of the method.A vehicle black smoke recognition model based on two stream convolutional neural network is proposed.Vi Be algorithm is used to detect moving vehicle targets in surveillance video,and hough line detection is used to obtain candidate black smoke areas,avoiding the influence of complex road environment and vehicle body on black smoke detection.The optical flow algorithm is used to calculate the optical flow image of the candidate black smoke region,and the dynamic and static characteristics of the black smoke are fully utilized.The images of the candidate black smoke region and the corresponding optical flow images are input into the two-stream convolutional neural network for black smoke recognition,which improves the accuracy of black smoke recognition.The proposed method has a high black smoke recognition rate of93.7%,which provides an effective scheme for monitoring the black smoke of moving vehicles.