V2I8P42

Presentation Attack Detection using Iris Biometric by Enhanced Category Boosting Method

Dharmendra Kaushik^1*, Dr. Alok Kumar Singh Kushwaha², Dr. Vinay Kumar³

Abstract

Biometric authentication systems have become increasingly prevalent in various domains, from security access control to personal device authentication. The unique patterns and characteristics of the human iris make it an ideal candidate for biometric authentication, offering high accuracy and resistance to forgery. However, despite its effectiveness, the integrity of iris recognition systems is threatened by the emergence of counterfeit iris images. These fabricated images can deceive the authentication process, leading to potential security breaches and unauthorized access. As such, there is a pressing need to develop robust mechanisms for detecting fake iris images and enhancing the security of iris recognition systems. In response to this challenge, researchers have explored various approaches, including advanced machine learning algorithms, to improve the speed and efficiency of iris recognition. By leveraging these algorithms, such as two-Dimensional-Convolutional Neural Network, K-Nearest Neighbours, Logistic Regression, Decision Trees, Random Forest, Gradient Boosting, Extreme Gradient Boosting, and Enhanced Category Boosting Method (ECBM) researchers aim to expedite the identification process without compromising accuracy. This paper discusses the utilization of these advanced machine learning algorithms in iris recognition. Among these algorithms the ECBM method finds quicker and more efficient way which is practically and effectively useful for iris recognition technology for various applications.

Keywords:

Iris Biometric, Presentation Attack, Anti-Spoofing, Enhanced Classification, Iris Recognition.