Title :
A novel DSP based real-time character classification and recognition algorithm for car plate detection and recognition
Author :
Lim, B.L. ; Yeo, Wenzheng ; Tan, K.Y. ; Teo, C.Y.
Author_Institution :
Dept. of Electron. & Comput. Eng., Ngee Ann Polytech., Singapore
Abstract :
In this paper, a novel character classification and recognition algorithm is developed and experimented on. Our classification and recognition technique is based on the properties of characters´ features, including their intersection, line ending and continuity features. The application of our novel algorithm to car licence number recognition has been tested using the real-time IVP-150 TMS320C50 digital signal processor based video image processing system. In our implementation, a remote camera will first capture the car image, the captured image will then proceed for image segmentation and pre-processing. Subsequently, the licence number will be located and recognised. In general, excellent performance has been obtained and is demonstrated with our experimental results
Keywords :
automobiles; character recognition; digital signal processing chips; feature extraction; image processing equipment; image segmentation; video signal processing; DSP based real-time character classification; DSP based real-time character recognition; IVP-150 TMS320C50 digital signal processor; captured image; car image; car plate detection; car plate recognition; character continuity features; character intersection; character line ending; image segmentation; remote camera; video image processing system; Character recognition; Classification algorithms; Digital signal processing; Digital signal processors; Image processing; Image recognition; Licenses; Real time systems; Signal processing algorithms; System testing;
Conference_Titel :
Signal Processing Proceedings, 1998. ICSP '98. 1998 Fourth International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4325-5
DOI :
10.1109/ICOSP.1998.770850
