License Plate Recognition System Based On Support Vector Machine

Intelligent Transport System(ITS) is an information technology which improves traffic condition by high technology.License plate recognition system that is one important part of Intelligent Transport System,is an attractive subject.Recognition of license Plate characters is the last step in the license plate recognition system,and the result of characters recognition shows the success or failure of the whole license Plate recognition system.The upgrade of recognition rate and accuracy for license plate recognition system,is becoming increasingly important for modern transportation technology.License plate recognition system mainly includes image pre-processing,license plate location,character segmentation and character recognition. Image pre-processing is the precondition of character recognition.In this paper,such approaches as image filtering,gray-scale transformation,Sobel edge detection,binarization,mathematical morphology studied image acquisition are used so that the post-processed license plate image can be protected from many interference factors.After the image pre-processing that reduces the complexity background of the license plate image,improves the accuracy of license plate recognition and saves the computation time and storage space.License plate location is an important step in this system.On the basis of analyzing Multi-classification SVM,aiming at the question of a small quantity of image classification,we present a SVM classification method.Plate regions that conform to the license plate characteristics are preliminary located by the contour rectangle area with the aspect ratio acquired by image pre-processing.Then the more precise plate region is located by SVM machine learning.Finally,the plate characters are classified by the BP neural network.Character recognition through the collection of samples of license plate characters are normalized to use wide-gridding character,then use the SVM classifier to recognize the characters.We present a new Multi-SVM classification method based on MLP and unilateral binary decision tree.Through training of MLP network,Lagrange multiplier vector and threshold value b in decision function,constant in kernel function,restricted value in ?-SVM classification can be gotten.In the end,character images are classified step by step using unilateral binary decision tree.When classifying,one procedure of adjusting parameter is adopted to shorten the time of classification and advance the precision of classification.Finally,the analysis of experimental data shows that the new recognition approach proposed in this paper can improve the recognition rate and accuracy of license plate recognition system tosome extent.