A comparison between a neural network and a SVM and Zernike moments based blob recognition modules

Author

Fedorovici, Lucian-Ovidiu ; Dragan, Florin

Author_Institution

Dept. of Automatics & Appl. Inf., Univ. of Timisoara, Timisoara, Romania

fYear

2011

fDate

19-21 May 2011

Firstpage

253

Lastpage

258

Abstract

This paper proposes a new algorithm to recognize the printed characters as part of the blob recognition modules of Optical Character Recognition systems. The algorithm uses a SVM classifier and Zernike moments for feature extraction. A comparison with another algorithm based on a five layer convolutional neural network is done. An analysis of the accuracy and the time needed to process one character leads to useful conclusions on the advantages and disadvantages of each algorithm.

Keywords

Zernike polynomials; feature extraction; neural nets; optical character recognition; pattern classification; support vector machines; SVM classifier algorithm; Zernike moment; blob recognition module; feature extraction; five layer convolutional neural network; optical character recognition system; printed character; Artificial neural networks; Character recognition; Classification algorithms; Feature extraction; Optical character recognition software; Pixel; Support vector machines; OCR engine; SVM classifier and Zernike moments; convolutional neural network; performance comparison;

fLanguage

English

Publisher

ieee

Conference_Titel

Applied Computational Intelligence and Informatics (SACI), 2011 6th IEEE International Symposium on

Conference_Location

Timisoara

Print_ISBN

978-1-4244-9108-7

Type

conf

DOI

10.1109/SACI.2011.5873009

Filename

5873009