Offline Chinese Signature Verification Based on AlexNet

Cui Wencheng; Guo Xiaopeng; Shao Hong; Zou Limin

Advanced Hybrid Information Processing. First International Conference, ADHIP 2017, Harbin, China, July 17–18, 2017, Proceedings

Research Article

Offline Chinese Signature Verification Based on AlexNet

Download

255 downloads

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1007/978-3-319-73317-3_5,
    author={Cui Wencheng and Guo Xiaopeng and Shao Hong and Zou Limin},
    title={Offline Chinese Signature Verification Based on AlexNet},
    proceedings={Advanced Hybrid Information Processing. First International Conference, ADHIP 2017, Harbin, China, July 17--18, 2017, Proceedings},
    proceedings_a={ADHIP},
    year={2018},
    month={2},
    keywords={AlexNet Convolution neural network Offline signature verification Writer-dependent},
    doi={10.1007/978-3-319-73317-3_5}
}

Cui Wencheng
Guo Xiaopeng
Shao Hong
Zou Limin
Year: 2018
Offline Chinese Signature Verification Based on AlexNet
ADHIP
Springer
DOI: 10.1007/978-3-319-73317-3_5

Cui Wencheng¹, Guo Xiaopeng¹^,*, Shao Hong¹, Zou Limin¹

1: Shenyang University of Technology

*Contact email: wayneguo279@gmail.com

Abstract

In order to break the limitation of traditional pattern recognition in offline Chinese signature verification, the method of applying machine learning is put forward. First, the offline Chinese signature data set is pre-processed, include removing noises, binarization and normalization. Then the architecture and implementation methods of AlexNet are proposed. The experimental results show the average accuracy of classification has been up to 99.77%, and verification rate is 87.5%.

Keywords: AlexNet Convolution neural network Offline signature verification Writer-dependent

Published: 2018-02-09
Appears in: SpringerLink

: http://dx.doi.org/10.1007/978-3-319-73317-3_5