Vehicles’ License Plate Recognition System Based On Digital Image Pressing

As a key technology for the road transport systems of the21st century, ITS’s(IntenigentTransportation System) research, development, and practical application effotrs are beingmade in many countires around the world. As an important content of ITS, LPR ’s (LicensePlate recognition) system can be widely used in urban traiffc management, expressway vehiclemonitor, electronic tolling, parking lots management, vehicle tracking and communityintelligence management etc. License Plate recognition technique, which is based on theory ofcomputer image processing and pattern recognition, is the key part of LPR system. So it issignificant to study the technique not only in theory field but in practical application.In general, LPRS can mainly divide into three modules t including: license plateslocatingf character segmentations and character recognitions. Techniques of all key modulesrelated to LPRS are deeply studied and analyzed in this Paper. Finally, we settle down a seiresof algorithms used in LPRS based on analyzing the typical license plate recognition algorithmsin recent years. In the plate positioning, first of all, with a series of digital image processingalgorithms to be the edge of the picture image, and then processed using mathematicalmorphology, forming multiple characteristics in line with license plates connected region.Nextj each feature according to the shape of plates, to get connected region analysis andexclusion, and thus suspected to be a rough plate area. In the character segmentation stage, thepaper mainly associated with the projection of information and piror knowledge of thecharacter segmentation of the feature information extraction conditions on the suspectedvehicle registration for the decision accurate positioning region, in the precise positioning atthe same time to achieve segmentation of the characters. The license plate location andcharacter segmentation method of combining the advantage of license plates can be filteredout near the region that color, texture characteirstics are not consistent with the license platetexture interference, such as the gird, borders and so on. For a single character recognitionproblem, this paper presents a formulation according to the structural features of charactercoding system, using codes to quickly and accurately identify the license plate characters.The testes of the LPR using the proposed algoirthms have been conducted. Theexperimental results show that the proposed approaches are reasonable, and they are feasibleand effective in application.