TY - GEN
T1 - Verification Code Recognition Based on Active and Deep Learning
AU - Xu, Dongliang
AU - Wang, Bailing
AU - Du, Xiao Jiang
AU - Zhu, Xiaoyan
AU - Guan, Zhitao
AU - Yu, Xiaoyan
AU - Liu, Jingyu
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/4/8
Y1 - 2019/4/8
N2 - A verification code is an automated test method used to distinguish between humans and computers. Humans can easily identify verification codes, whereas machines cannot. With the development of convolutional neural networks, automatically recognizing a verification code is now possible for machines. However, the advantages of convolutional neural networks depend on the data used by the training classifier, particularly the size of the training set. Therefore, identifying a verification code using a convolutional neural network is difficult when training data are insufficient. This study proposes an active and deep learning strategy to obtain new training data on a special verification code set without manual intervention. A feature learning model for a scene with less training data is presented in this work, and the verification code is identified by the designed convolutional neural network. Experiments show that the method can considerably improve the recognition accuracy of a neural network when the amount of initial training data is small.
AB - A verification code is an automated test method used to distinguish between humans and computers. Humans can easily identify verification codes, whereas machines cannot. With the development of convolutional neural networks, automatically recognizing a verification code is now possible for machines. However, the advantages of convolutional neural networks depend on the data used by the training classifier, particularly the size of the training set. Therefore, identifying a verification code using a convolutional neural network is difficult when training data are insufficient. This study proposes an active and deep learning strategy to obtain new training data on a special verification code set without manual intervention. A feature learning model for a scene with less training data is presented in this work, and the verification code is identified by the designed convolutional neural network. Experiments show that the method can considerably improve the recognition accuracy of a neural network when the amount of initial training data is small.
KW - Verification code recognition
KW - convolutional neural network
KW - feature learning
UR - http://www.scopus.com/inward/record.url?scp=85064986513&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85064986513&partnerID=8YFLogxK
U2 - 10.1109/ICCNC.2019.8685560
DO - 10.1109/ICCNC.2019.8685560
M3 - Conference contribution
AN - SCOPUS:85064986513
T3 - 2019 International Conference on Computing, Networking and Communications, ICNC 2019
SP - 453
EP - 456
BT - 2019 International Conference on Computing, Networking and Communications, ICNC 2019
T2 - 2019 International Conference on Computing, Networking and Communications, ICNC 2019
Y2 - 18 February 2019 through 21 February 2019
ER -