Scholarlink

In modern days payments and transactions are concern about safer way. It takes place due to hackers on OTP. Our proposes a best method for credit card transactions system which integrates with the face detection. One of the major issue about the online transactions are the issues of the hackers and to provide safer transactions. The existing credit card fraud system uses two layered security model to detect fraud or to identify the genuine transaction. But this model is also compromising as hackers are hacking OTP these days, we propose a solution which includes one more security layer to improve the existing ones. This model uses third layer security by authenticating the user through his/her face. The customer’s face is captured during the transaction and the system will detect his/her authorization using face detection technique. Our model uses LBP algorithm for facial feature extraction and uses RESNET to classify the genuine/fraudulent customer. This facial recognition system is simulated/tested and has given the accuracy rate of 92%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Three Layered Security Model to Prevent Credit Card Fraud Using LBPH and CNN-Resnet Architecture

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Chandrakala Kuruba,
  • S. K. Sharmila,
  • V. Mounika,
  • D. Aswini,
  • G. Poojitha

摘要

In modern days payments and transactions are concern about safer way. It takes place due to hackers on OTP. Our proposes a best method for credit card transactions system which integrates with the face detection. One of the major issue about the online transactions are the issues of the hackers and to provide safer transactions. The existing credit card fraud system uses two layered security model to detect fraud or to identify the genuine transaction. But this model is also compromising as hackers are hacking OTP these days, we propose a solution which includes one more security layer to improve the existing ones. This model uses third layer security by authenticating the user through his/her face. The customer’s face is captured during the transaction and the system will detect his/her authorization using face detection technique. Our model uses LBP algorithm for facial feature extraction and uses RESNET to classify the genuine/fraudulent customer. This facial recognition system is simulated/tested and has given the accuracy rate of 92%.