The Character Recognition of Vehicle's License Plate Based on BP Neural Networks

Abstract:

This paper based on license plate image preprocessing ,license plate localization, and character segment ,using BP neural network algorithm to identify the license plate characters. Through k-l algorithm of characters on the feature extraction and recognition of license plate character respectively then taking the extraction of license plate character features into the character classifier to the training. When the end of training, extracting the net-work weights and offset matrix, and storing in the computer. To take the identified character images input to the MATLAB, and with the preservation weights and offset matrix operations, obtain the final results of recognition.

• Periodical: Applied Mechanics and Materials (Volumes 513-517)
• Pages: 3805-3808
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.513-517.3805
• Online since: February 2014
• Authors: Wen Bo Liu*, Tao Wang
• Keywords: Character Recognition, Feature Extraction, K-L Algorithm, Neural Network (NN)

References

[1] Lu Xiaobo, Zhang Guanghua, Precise positioning method based on the binary image of the license plate [J], Southeast University, November (2005).

[2] Bai Hongliang, Lou Zhengliang, Zou Mingfu. Morphology-based license plate recognition system with complex background [J]. Highway and Transportation Research and . (2004).

[3] Zongquan Yu, Han Yu. The Neural Network Control". Xi, an University of Electronic Science and Technology Publishing House, (2009).

[4] Zhongzhi Shi. The Neural Networks,. Higher Education Press, (2009).

[5] Shuntian Lou, Yang Shi. Systems analysis and design - based on MATLAB neural network". Xi, an University of Electronic Science and Technology Publishing House, (2000).

[6] Yan Shi, Liqun Han, Xiaoqin Lian. Neural network analysis of design methods and examples,. Beijing University of Posts and Telecommunications Press, (2009).