In recent years, enhancing the security and convenience of banking transactions has become a primary focus in financial technology. This paper presents the design and development of a Facial Biometric-Based ATM Transaction System aimed at reducing dependency on physical ATM cards and enhancing user authentication. The proposed system initiates the transaction by requesting the user to enter their mobile number, followed by bank selection in case the number is linked to multiple accounts. Once the bank is selected, the user is prompted to enter a 4-digit PIN. Upon successful authentication, the system proceeds to facial recognition for identity verification. Only after a positive match, the user is granted access to the main ATM menu, where standard banking operations can be performed. This multi-layered approach—combining mobile number, PIN, and facial biometric verification—significantly reduces the risks associated with card theft, skimming, and unauthorized access. The system is built using computer vision technologies integrated with secure backend validation, ensuring a balance between usability and security. The solution aims to provide a robust, cardless ATM experience, aligning with modern digital banking trends.
Paper
The full text of this publication is not hosted on 44B due to licensing.
Read it at OpenAlex