Research On Legible Image Of Dynamic Vehicle Capture

Video surveillance in the areas of traffic management and traffic information has been collected and used as key information. Because traffic video information has been collected and used as an important part of traffic surveillance and management, great importance has been attached to video surveillance technology,hich is based on video image processing, analyzing, and understanding, to improve intelligence of traffic surveillance and management. The research for the traffic image detection and process methods, with important theoretical meaning and wide practical value for traffic management and traffic control methods, has become a focus of attention.In the field of capturing the image of motive vehicles, the basis of the License Plate Recognition is the legible vehicle picture, where considerable attention has been arisen and much study has been developed. This paper analyses the relationship between image sharpness and the system hardware, and provides some ameliorative arithmetic and makes them a integrated.The paper presents a model based on YCbCr space, which can get over flaw of the traditional gray method. When updating the background, self-adapting method is used, which can fit the changing of the light and the things moved in for long time. When object is extracted, region growing method and objective feature identify method are adopted. The procession of selecting seeds is cut and the overall time is reduced. Then the timing to snap the object is confirmed according to image grads and extremes method. And the distinct image in field of vision with motive objective is acquired.By testing indoor and simulating experiments outdoor, when system parameter are setting properly tests showed that the methods were fast and accurate and suitable for real time applications.