• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.3
• /
• pp.379-388
• /
• 2020

As the use of credit cards increases, security threats increase. In particular, despite being vulnerable to related crimes, such as fraudulent use of credit cards and theft of names, there are virtually no security procedures to authenticate the validity of user while paying with the credit card. In order to overcome these limitations of current credit card payments, we add a process of signing payment using a stylus pen with built-in acceleration sensor in the existing transaction method, and classifying and comparing the image of the signature and signature information measured by the sensor through the convolutional neural network. we propose a method to improve security in financial transactions by performing the user authentication process through the possession of the stylus pen and the characteristic values of the signature.