Development of Personal Identification Number Authorization Algorithm Using Real- Time Eye Tracking & Dynamic Keypad Generation

The digital financial transaction is on a continuous rise and going to be the order of the day. These transactions rely on the entry of the Personal Identification Number (PIN) by the user. The PIN is a common user authentication method for many applications such as ATM's, unlocking personal devices. Cyber-crimes are committed by shoulder surfing or thermal tracking. PIN entry is found vulnerable to password attacks such as shoulder surfing or thermal tracking. Intruders try to gain passwords or personal identification numbers (PIN) by glancing over the user's shoulder and observing the pattern of PIN entry (shoulder surfing). Thermal tracking is another method followed by cyber thieves, using the heat traces to decode the entered PIN by the user. This paper proposes a novel approach to authorization of PIN. To demonstrate the same, a web application is developed with trio-based authentications using machine learning techniques. Initially, the application detects and recognizes the user's face. A dynamic keypad is displayed which prompts the user to provide input PIN via eye blink. User's eye is detected and monitored by the application in order to capture the PIN and verifies the same with the existing PIN in the database. On a successful PIN verification process, the application allows the user to proceed with the transaction. The results show the proposed novel approach is better than existing approaches.