A Study and Implementation of Real-Time License Plate Detection and Recognition System

• Main Authors: Yung-Hung Lin, 林永弘
• Other Authors: Jung-his Lee
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/95830925495195386617

碩士 === 元智大學 === 電機工程學系 === 101 === In this thesis, a real-time system built on Android mobile platform for license plate detection and recognition has been developed. The system is comprised of three major components, which are license plate localization, character segmentation and character recognition. For license plate localization, AdaBoost first is used for training with input features obtained by Haar-like features. Considering real-time constraint, cascade classifiers are developed for the purpose of efficient filtering. To locate each character, edge and contour features are employed. To recognize license plate efficiently, an open source of Tesseract-ocr provided by Google is used. In addition, JavaCV is also used for basic image processing, which can be directly used based on function call on Android mobile platform. To overcome the problem of limited capturing distance when using mobile phones, the functions of semi-automatic zooming and auto-focus have been included in our developed system. Conducting our system in real-time can achieve precision rate above 90% and thus shows its feasibility for real world environments.