License Plate Identification Based On SIFT Algorithm

License plate recognition is one of the focus and difficult researches in the field ofintelligent transportation systems. In this paper, license plate recognition based on theSIFT is studied. The system includes three parts: license plate image pretreatment,license plate location and character recognition. The license plate image pretreatmentpart contains gray-scale transformation, filtering, sharpening, normalization, adding a0.8mm white border to the standard template character, and modifying the parametersof the traditional SIFT algorithm. Experiments show that modifying the parameters ofthe SIFT algorithm improves the accuracy of plate positioning. In the license platelocation part, first extract the SIFT features of the standard positioning plate and savethem, second, extract the SIFT features of license plate to be identified and save them,next, match these two features, and last, segment the license plate according to results ofthe feature point matching. In the character recognition part, first extract the SIFTfeatures of the standard template characters and save them, then extract the SIFTfeatures of the positioned license plate, at last, characters are identified by matching thefeature vectors. Experimental platform is built with OpenCV and Visual C++in thispaper. The size of21*42and resolution of96pixels/inch standard template charactersare chosen. Experiments show that the accuracy of license plate location is78%, licenseplate character recognition rate: Chinese character is100%, letter is68%, and numberis52%.