Vehicles Video Image Processing Based On Detection And Tracking Algorithm

With economic development and social progress, the demand for all kinds oftransportation is also rapid growth, the existing road network management capabilities andmanagement facilities can not meet the growing demand for traffic and transportationproblems have become increasingly serious traffic congestion when long vehicles changelanes random case occurs, causing a lot of unnecessary accidents. To solve these trafficproblems, Intelligent Transportation System has been referred to the important position.Moving vehicle detection and tracking system as an important part of ITS, as the currentresearch focus. Vehicle tracking technology is one of the key technologies for videosurveillance vehicles, how to achieve study video image processing algorithm based on thedetection of vehicles, and vehicle tracking has become an important research issues at homeand abroad.This article will be divided into vehicle detection vehicle tracking systems, vehicletracking two major technology modules, and each module is mainly technical methods wereintroduced. In the vehicle detection module, through the detection of several commonly usedmethods were analyzed and compared to determine the detection method using the vehicledetection method based on frame difference, focusing on the background of the algorithm toextract and update methods selected threshold key technologies. In the vehicle trackingmodule, the focus of several vehicle tracking algorithm were analyzed and compared todetermine the use of feature-based tracking algorithm for tracking markers. Finally get totrack the target vehicle trajectory tracking markers on the basis that can detect whether atraffic violation by its trajectory. After determining violations, vehicle pictures through thecamera memory to determine its license plate images based license plate recognition matlablicense plate recognition information to obtain vehicle information.Analysis and examples using MATLAB software algorithm selected for verification, todetermine whether the chosen algorithm can achieve the desired effect.