Study Of The Detection And The Tracking On Moving Objects In The Automatic Stake Test System

The detection and the tracking on moving objects is one of the main areas of the digital image processing. It has been highly concerned in recent years. It is used widely in Intelligent Video Surveillance, Moving Analysis, Navigation and so on.The main study of this paper is the detection and the tracking on moving objects in the Automatic Stake Test System. We can search, detect and track the car in the test field by the Real-time analysis of the video images. We can also analyse and estimate the characteristic of the car's moving to gained the moving track. We can find whether the car is out of the rule or not to make sure whether examinee is qualified or not by estimating Real-time moving direction and the signal of the infrared transducer. The study of the detection and the tracking on moving objects mainly include three aspects below :The pretreatment of the image. We summarize the main techniques of the image denoising. We use the autokinetic median filter to eliminate the noise of the image on the characteristic of the Automatic Stake Test System. We can preserve the edge and the detail of the image at the same time.The detection on moving objects. We find a new background updating algorithm—the Surendra Background Updating Algorithm on subtraction between foreground image and background image. We can find the area of the moving objects at the same time. We use the method of the color space of HSV and the grads of the characteristic for detecting and eliminating the shadow of the image. We can obtain accurately the information of the position and the characteristic of the moving object by the Blob Analysis of the Mathematical Morphology.The tracking on moving objects. We find the centroid of the moving object to track the object and draw the track of the object, on the own characteristic of the Automatic Stake Test System and the information of the position and the characteristic of the moving object.The experiment proves that we can detect, track and identify moving objects well by the way of this article. It can raise the equity and veracity of the Automatic Stake Test System by using the digital image processing.