Research On Vehicle Violation Behavior Detection Based On Video Image Processing

In recent years,due to the continuous increase of urban population,the demand for motor vehicles is on the rise.As a result,it has brought a series of traffic problems,among which the impact caused by traffic accidents is particularly serious.Traffic accidents are mostly the result of illegal traffic.However,the traditional traffic technology is not satisfactory in solving this problem.Therefore,the intelligent traffic system is concerned by domestic and foreign governments and scientific and technical personnel.The vehicle violation detection system is used as an intelligent traffic system.An important technology that has the value of research.This thesis studies the detection method of vehicle violations based on video image processing.The key research violation is illegal parking.It mainly includes three parts:moving target detection,vehicle identification,and parking violation judgment.In the moving target detection part,due to a series of problems in the original Vibe algorithm,such as false targets caused by ghosting problems,incomplete detected moving targets,and shading problems caused by light,an improved Vibe algorithm is proposed.The algorithm utilizes five frames of difference to perform background initialization,introduces a Gaussian model for moving object detection,and eliminates shadows in the YCb Cr color space.It can be seen from the experimental results that the algorithm used in this thesis can effectively solve the problems of the original Vibe algorithm.In the vehicle identification part,this thesis utilizes multi-features cascading classifier to identify the vehicle.By extracting the Haar-like features and HOG features of the vehicle and using the Adaboost classifier to train,according to the experimental data,the algorithm can be used to identify the vehicle.The accuracy rate achieved the desired effect.In the part of illegal parking of vehicles,this thesis uses pixel gray value changes to determine whether the violation.By setting up the prohibited area,the gray value of the background pixel is calculated within the prohibited area,and whether or not to stop is determined by judging the change of the gray value.This thesis proposes a vehicle image violation detection method based on video image processing.Through the experimental results,it can be shown that the algorithm used in this thesis has a high accuracy in illegal parking detection and can meet the real-time detection requirements of the traffic management department.It has strong feasibility and practicality.