The Analysis And Identification Method Of Objective Dynamic Characteristics Based On Image Understanding

Intelligent Transportation Systems is the world's cutting-edge field of transportation research. It combines electronic information technology, communications technology, automatic control theory, computer technology and traditional theory of traffic engineering multi-disciplinary theory, and applied to a modern traffic management system, in order to achieve transport services and management of intelligent, establish a large extent, the full role of real-time, accurate and efficient integrated management system. Therefore, ITS has become an information processing system in an important subject.Intelligent Transportation Systems in the video detection system mainly uses computer vision, image recognition and other technologies for the management of transportation systems to provide more intuitive and convenient analytical tools. It applies to conventional road vehicles, highway vehicle management, car park management, and has broad application prospects.This text to improve the accuracy and precision of video detectors detection is proposed feature-based detection methods. Among them, road traffic the road can not be replaced as the background, its recognition is very important. In this paper, the color information of their analysis and the use of edge detection and transformation methods such as extraction Hough lane line to identify the road surface, to distinguish between off-road information.Transport vehicle as an important tool for intelligent traffic system in the process of video detection, vehicle detection is very important. In this paper, characteristics of vehicles, from the edge features, corner features, symmetry, etc. to do an in-depth study of the shadow of proposed based on characteristics of vehicle chassis and body adhesion test segmentation method, and use the windows feature extraction and location of vehicles.Person as one of the main traffic, video traffic detection system for pedestrians is also very important. This paper analyzes the characteristics of pedestrians using the shape features, Haar features and characteristics in accordance with its improved triangular pedestrian recognition.