Development of a Non-contact Two-Tier Biometric Security System for the DSWD 4Ps using Iris recognition and Speech Recognition

Abstract

There has been a steep rise of contactless payment during COVID-19. The rapid improvements of miniaturized sensors and biometric recognition systems for face identification, fingerprint, iris, and voice are conducive and fit during this rise of COVID-19. Thus, non-contact interactions are the most effective way to fight against the spread of the virus and any other diseases. One of the most used is iris scanners and speech recognition. The study promotes contactless payments to address the accompanying issues in cash aid distribution particularly in the DSWD 4Ps, where it has a two-tier biometric security system which is iris recognition and speech recognition. This can provide the same type of service and securities as a normal ATM while removing the worry of getting different kinds of viruses and diseases. Testing the iris recognition system, a False acceptance ratio of 13% and 3% of False Rejection rates were achieved. While for the testing of speech recognition (security questions), a False Acceptance Ratio of 0% and False Rejection Ratio of 12.12% were achieved. Lastly, testing of speech recognition (navigation)a False Acceptance Ratio of 0% and False Rejection Ratio of 3.62% were achieved. Giving the system an 84% accuracy for the iris recognition, 87.88% for the security questions, and 96.36% for the navigation.