Enhanced Iris Recognition System Using Daugman Algorithm and Multi-Level Otsu’s Thresholding
Darrel Tristan U. Virtusio, Fil Janssen D. Tapaganao, Rafael G. Maramba · 2022 International Conference on Electrical, Computer and Energy Technologies (ICECET) · 2022
Iris recognition is regarded as one of the ideal biometrics due to its advantages in terms of accuracy, speed, and efficiency. However, there are still some instances when iris identification is limited, such as when the acquired iris image is of poor quality. This study aims to propose an enhanced iris recognition system using multi-level Otsu’s thresholding to improve the quality of the iris image. The program captures iris samples using Raspberry Pi camera NoIR, saves the images into the database, and outputs whether a match has occurred. Two databases were created for the comparison of results. The first database contains the iris samples applied with multi-level Otsu’s thresholding, while the second database contains the raw image of iris samples. Data gathering includes 30 different iris samples, and 15 samples were registered. For each sample, 30 trials were done to test the accuracy. The accuracy for iris images with multi-level Otsu’s thresholding is 96.60%, and the accuracy for iris images without image enhancement is 94.23%. It is concluded that applying multi-level Otsu’s thresholding can enhance the performance of the iris recognition system.