Segmentation Algorithm Based On The Morphology Of The License Plate Location And Character

Vehicle Plate Recognition technology, as a key component of intelligent transport system, is playing a more and more role in modern society. The system can recognize several kinds of vehicle plate through digital image process, mathematics morphology and pattern recognition etc. It realizes scientific management and ensures traffic order. It has a comprehensive application prospect.The VPR system is composed with vehicle plate capture, vehicle plate recognition module, database management and data transfer through net etc. The vehicle plate recognition module is composed with vehicle plate location, character segmentation and character recognition.The algorithm of vehicle plate location and character segmentation is studied systemically in this paper.In vehicle plate location part, an algorithm applied to license plate location based on Renyi's entropy and binary area morphology is introduced. This algorithm binaries the plate image by one-dimensional Renyi's entropy first, then the area of the connected component is computed. At last, area morphology combined with its elemental calculation is used to realize the license plate location. Experiment results show that such method is not only simple and fast, but also robust.In character segmentation part, an advance algorithm is introduced. We propose a new license plate image preprocessing approach for character segmentation based on Renyi's entropy and improved mathematical morphology edge detection. The algorithm binaries the license plate by two-dimensional Renyi's entropy first, then detects the edge by morphology erosion. After getting ride of the outer bounding frame, license plate characters are segmented by projection. Experiment results show the algorithm avoids disadvantages of traditional algorithm, and wipes off the noise. It is an efficient algorithm that has a fast computation speed and a distinct segment result.