Character recognition research based on BP neural network

Author

Zhu, Pingchang ; Xu, Kai ; Wang, Fang ; Ye, Hong ; Fang Wang ; Zhao, Huijun

Author_Institution

Coll. of Inf. Mech. & Electr. Eng., Shanghai Normal Univ., Shanghai, China

fYear

2011

fDate

15-17 July 2011

Firstpage

4814

Lastpage

4817

Abstract

This article is about the method on how to extract character recognition of vehicle brand by the combination of EMD and nonlineary PCA. The main idea is to identify the image of the object which contains character recognition technology in BP neural network that momentum factor has been added in advance. After that compared with the improved neural network performance in different parameter occasions, the test proved that this method can increase the training speed and precision of character Recognition neural network.

Keywords

backpropagation; character recognition; feature extraction; image recognition; neural nets; principal component analysis; BP neural network; EMD; character recognition research; feature extraction; momentum factor; nonlineary PCA; principal component analysis; Character recognition; Educational institutions; Electrical engineering; Instruments; Mobile computing; Presses; Principal component analysis; Character Recognition; Momentum Factor; Neural Network; Sigmoid Transfer Function;

fLanguage

English

Publisher

ieee

Conference_Titel

Mechanic Automation and Control Engineering (MACE), 2011 Second International Conference on

Conference_Location

Hohhot

Print_ISBN

978-1-4244-9436-1

Type

conf

DOI

10.1109/MACE.2011.5988090

Filename

5988090