Traffic Video Identifies The Text Detection And Recognition Technology Research

It is important in theory and applications to study the detection and segmentation of object identifying character image in video sequence, which is used in content segmentation of image sequence, high efficient representation, transportation, storage and interactive control, image content searching, recognition and tracking, security barrier of communication network and information etc. In this thesis, is studied the segmentation and recognition of mobile identifying character image in video sequence. In Intelligent Transportation System(ITS), the recognition of the license plate by computer has very important significance in the society,such as no-stop toll gates in highway, vehicle survey, monitor and management system.License Plate Recognition(LPR) system is an example of application using computer visual and pattern recognition in ITS. In this thesis, is introduced some concepts, research background and present situation of research and application of LPR system. On the detection and extraction of license plate region and character recognition is applied the emphasis of this thesis.In the detection and extraction of license plate region, at first, a color filter is designed according to the background color of LP. If this method failed, then a method based on gray image segmentation is applied. PC simulating results are also given.Character recognition is belonged to classic pattern recognition. Traditional template matching recognition could be realized easily. After analyzing the space of the LP character, correlation recognition scheme is discussed. For some subgroup of characters, it is probably required to distinguish them further.