Research in Character Recognition Based on Generalized Regression Neural Network

Author

Wang, Shenghui ; Yang, Fei

Author_Institution

Inf. Technol. Center, China Guangdong Nucl. Power Holding Co. Ltd., Shenzhen, China

fYear

2012

fDate

7-9 July 2012

Firstpage

318

Lastpage

320

Abstract

Template matching and machine learning are the common methods in character recognition. Generalized regression neural network(GRNN) belongs to machine learning, which has the advantages of simple design, fast convergence speed, requiring less training samples, with global approximation and the best approximation property, robustness and strong nonlinear processing ability than traditional classifiers. In this paper we use generalized regression neural network in character recognition and do some research in license plate recognition system.

Keywords

approximation theory; character recognition; image matching; learning (artificial intelligence); neural nets; object recognition; regression analysis; GRNN; character recognition; convergence speed; generalized regression neural network; global approximation; license plate recognition system; machine learning; nonlinear processing ability; template matching; Biological neural networks; Character recognition; Feature extraction; Licenses; Neurons; Training; GRNN; character recogintion; license plate recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Computing, Measurement, Control and Sensor Network (CMCSN), 2012 International Conference on

Conference_Location

Taiyuan

Print_ISBN

978-1-4673-2033-7

Type

conf

DOI

10.1109/CMCSN.2012.77

Filename

6245876