A CAPTCHA Recognition Algorithm Based on Holistic Verification

Author

Shu-guang, Huang ; Liang, Zhang ; Peng-po, Wang ; Hong-wei, Han

Author_Institution

Dept. of Comput. Applic., Electron. Eng. Inst., Hefei, China

fYear

2011

fDate

21-23 Oct. 2011

Firstpage

525

Lastpage

528

Abstract

CAPTCHA is a new kind of network security mechanism. Studying the recognition of CAPTCHA can help to discover its hidden defects and thus make it more secure. For closely-connected CAPTCHAs that can hardly be recognized by methods of state of art, this paper proposed a new recognition algorithm based on holistic verification. During the process of this algorithm, Recurrent Neural Network (RNN) was first used to recognize unknown CAPTCHAs. Then, recognition results were verified by SVM rejection, synthetic data generation and Extreme Learning Machine (ELM). Experiments results show that this algorithm can not only recognize closely-connected CAPTCHAs but also effectively boost the recognition rate of RNN.

Keywords

image recognition; recurrent neural nets; support vector machines; CAPTCHA recognition algorithm; ELM; RNN; Recurrent Neural Network; SVM; extreme learning machine; holistic verification; network security mechanism; Character recognition; Feature extraction; Humans; Image recognition; Image segmentation; Security; Training; CAPTCHA recognition; Extreme learning machine; Network security; Synthetic data generation;

fLanguage

English

Publisher

ieee

Conference_Titel

Instrumentation, Measurement, Computer, Communication and Control, 2011 First International Conference on

Conference_Location

Beijing

Print_ISBN

978-0-7695-4519-6

Type

conf

DOI

10.1109/IMCCC.2011.136

Filename

6154161