Research On Target Motion Detection And Tracking Technology

With the development of the information society, the computer is widely used to all fileds of life. And computer vision as a comprehensive discipline, has attracted a cross-disciplinary scholars devoted to research on it. Computer vision is to use a computer and related equipment of biological vision of a kind of simulation, it is through the main task of gathering pictures or video processed to obtain corresponding scene 3-d information, like humans and many other creature has done for every day.Target motion detection and tracking technology is one of a wide range of areas on computer vision research. It depends on the practicability of this technology, whether in the civil or high military field have high research use value. Video intelligent monitoring system is the most moving object detection and can reflect the core value of tracking technology platform.This article from target motion detection and sports target tracking the two aspects the technology; First, it introduced several methods which are the most commonly used in moving object detecting:optical flow method, frame difference method, background subtraction. These methods are the most common basic methods in nowdays, because of its usage scope limitation, cause a single use one way cought not get accurate results. According to detect moving target next to classification of different background, respectively from the static background target detection and dynamic background target detection two aspects to discuss it.Sports target tracking is built based on the target motion detection the research contents, but also more in-depth research foundation follow-up, such as:abnormal behavior analysis, traffic flow statistics, etc. This paper introduces the research moving target tracking technology, analyzes the key points based on the Shift since the target tracking algorithm based on kalman filter and target tracking algorithm.At the end of this article, intelligent transportation system as a platform to run as the research object designed vehicle traffic statistical system. The system used the target detection based on background model method. And while the vehicle tracking using the method based on kalman filtering principles. The experimental results show that the system has good adapt to the characteristics of environmental changes, effective vehicle detection and tracking, and accurate statistics in the flow of traffic virtual line.