Design And Implementation Of Traffic Video Analysis System Based On Yolo Algorithm

With the development of economy and the acceleration of urbanization,using motor vehicles to travel has become the first choice for many people.The increasing number of motor vehicles has brought about the problems of road congestion and frequent accidents.Public security departments use information technology to manage congested road sections,invest a lot of money to install a large number of video monitoring equipment on the road,and establish video monitoring center.However,the monitoring method is mainly manual monitoring,which is labor-consuming and inefficient.At the same time,in recent years,in the field of convolutional neural network research,many high-quality target detection algorithms have come out.Yolo algorithm is a fast target detection algorithm proposed in 2015,which is faster and more accurate than other target detection algorithms.Therefore,in this paper,a traffic video analysis system is designed and implemented by using the Yolo algorithm.Yolo V3 is the latest generation model proposed by the author of Yolo algorithm.Under the same recognition accuracy,the detection speed is significantly faster than other detection models,which can achieve the real-time detection effect.In this system,the Darknet framework is used as the framework for training.Referring to the structure and training method of Yolo V3 network model,we make a data set by the video captured in the field to train the target detection model for traffic scenes,which can quickly and accurately identify and count vehicles and pedestrians.Taking the model as the core algorithm,a complete traffic video monitoring system is designed and implemented.The system can automatically analyze the real-time video in the camera,count the vehicles and the number of people,cooperate with the video source configuration,data viewing,automatic alarm and other functions of the background system,assist the traffic police in law enforcement,improve the efficiency of law enforcement and reduce the labor cost.We record the work of each stage in system development,including system requirement analysis,system design,system implementation and test.After the system test,it is confirmed that the system is stable in operation,perfect in function and without obvious defects.After showing the system to customers and providing customers with experience,the system gets good feedback from customers,achieves customer expectations and meets the design requirements.