Iris Liveness detection using SIFT, SURF and SVM with Quality Metrics for Biometric Authentication

Iris recognition technology has become increasingly popular in applications involving biometric security because of its precision and uniqueness. These systems are vulnerable to fraudulent attacks, in which counterfeit iris samples are used to spoof the verification process. To address this issue, various methods for detecting iris authenticity have been developed to differentiate legitimate and counterfeit iris images. This paper presents a methodology for identifying iris Liveness using a Scale-Invariant Feature Transform (SIFT), Speeded Up Robust Features (SURF), and Support Vector Machine (SVM) with quality Metrics. Our approach includes pre-processing iris images, extracting distinguishing characteristics using SIFT and SURF techniques, and SVM for classification. The results of experiments conducted on a standard Clarkson dataset 2017 and CASIA-Iris-Thousand, demonstrate the effectiveness of our approach, achieving remarkable precision in distinguishing between authentic and counterfeit iris images. This study contributes to the development of iris liveness detection technology and demonstrates the potential for enhancing the security of biometric verification systems.