Font Size: a A A

Study On The Algorithm Of Vehicle License Plate Recognition Using The Features Of PCA And Grid

Posted on:2014-01-18Degree:MasterType:Thesis
Country:ChinaCandidate:Y WangFull Text:PDF
GTID:2268330401953025Subject:Navigation, guidance and control
Abstract/Summary:PDF Full Text Request
License Plate Recognition System is one of the most critical part in the Inte lligentTransportation Systems, is also the research fouse at home and abroad. It containsmany technologies, including image processing, pattern recognition and so on, whichcan be widely used in the field of automated highway toll collection, traffic monitoring,community and parking management system. The license plate character segmentationand characters recognition are studied deeply in this paper, and puts forward someimproved algorithms, and by using VC++6.0platform to develop license platerecognition system in the paper.The existing license plate recognition algorithms are analysed and improvedalgorithms are given by the paper. For license plate image tilt correction, an adaptivethreshold Canny edge detection algorithm is used. The method is able to detect thecomplete edge information, which can accurately determine the angle of the inclinedplate, to realize the calibration of the license plate. For the license plate areabinarization, an adaptive step binarization algorithm is proposed. It has an ability ofresisting noise and a strong adaptability. On this basis, a method based on verticalprojection is used to segment characters, effectively improves the success rate of thecharacter segmentation. For character recognition, aiming at the characteristics oflicense plate characters, for chinese characters, a multi-template matching algorithmbased on PCA(Principal Component Analysis) features is adopted, for alphanumericwith a single template matching algorithm based on the gird features.Finally, the license plate recognition system which developed by the paper wastested. The test result shows that the system can effectively realize the license platelocation, character segmentation and recognition, and the rate of the whole systemrecognition reached88.59%.
Keywords/Search Tags:Character segmentation, Binarization, Character recognition, PCA feature, Gird feature
PDF Full Text Request
Related items