Automated Detection of Spoof Iris Images using Local Texture Analysis
Chun-Wei Tan, Ajay Kumar · 2013
Iris recognition has been regarded as one of the most promising technologies to provide reliable human identification [1]. For example, iris biometrics has been employed in the Aadhar project to provide reliable identification for millions of citizens [2]. Therefore, the integrity of such large scale iris recognition system must also be ensured to avoid the potential threats such as spoofing attacks, e.g. [3], which can pose vulnerability to the iris recognition systems. Ref. [4] details the feasibility of using high quality printed iris images to compromise the iris recognition technologies deployed today. In this project, our objective has been to develop new algorithms that can be used to for detecting fake iris images using the advanced image processing and machine learning techniques.