Iris Presentation Attack Detection Based on Photometric Stereo Features

Abstract

We propose a new iris presentation attack detection method using three-dimensional features of an observed iris region estimated by photometric stereo. Our implementation uses a pair of iris images acquired by a common commercial iris sensor (LG 4000). No hardware modifications of any kind are required. Our approach should be applicable to any iris sensor that can illuminate the eye from two different directions. Each iris image in the pair is captured under near-infrared illumination at a different angle relative to the eye. Photometric stereo is used to estimate surface normal vectors in the non-occluded portions of the iris region. The variability of the normal vectors is used as the presentation attack detection score. This score is larger for a texture that is irregularly opaque and printed on a convex contact lens, and is smaller for an authentic iris texture. Thus the problem is formulated as binary classification into (a) an eye wearing textured contact lens and (b) the texture of an actual iris surface (possibly seen through a clear contact lens). Experiments were carried out on a database of approx. 2,900 iris image pairs acquired from approx. 100 subjects. Our method was able to correctly classify over 95% of samples when tested on contact lens brands unseen in training, and over 98% of samples when the contact lens brand was seen during training. The source codes of the method are made available to other researchers.

A 3D Iris Scanner From Multiple 2D Visible Light Images

“…Additionally, our proposed method could be useful for detecting contact lenses, and therefore is another possible area of research in iris recognition [53]- [55]. The 3D model of a colored contact lens will have a spherical shape, since the dominant texture is printed in the surface of the contact lens.…”

Open Source Iris Recognition Hardware and Software with Presentation Attack Detection

“…In another work by Gragnaniello et al [13], the authors combine domainspecific knowledge into the design of their CNN, namely the network architecture and loss function. The PAD methods we adopt in this paper are the most recent, open-source methods evaluated by Fang et al [12] as best-performing in their experiments: OSPAD-2D [26], OSPAD-3D [6], and OSPAD-fusion [12]. OSPAD-2D is a multi-scale BSIFbased PAD method on textural features, OSPAD-3D is a photometric-stereo-based PAD method on shape features, and OSPAD-fusion is a combination of them.…”

Iris presentation attack detection: Where are we now?

“…Based on methods in [24] and [7], [12] proposed an algorithm that fuses the 2D textural features and 3D photometric stereo features (OSPAD-fusion). The authors identified that OSPAD-3D often fails to detect attack presentations of highly opaque contact lens, as they produce very little shadow, and OSPAD-2D often achieves a high APCER and low BPCER on unknown samples, so the samples marked as "attack" by OSPAD-2D are usually correctly classified.…”

“…To the best of our knowledge, no new open source methods have been released since that paper. Three modern publications in this paper, [24,7,12], along with three older ones are included in the comparison using the same protocol. The authors found that the PAD method based on photometric stereo features [7] generalize better to attacks of contact lens of unknown textures, while the BSIF texture-based PAD method [24] performs better in closed-set scenarios.…”

“…Three modern publications in this paper, [24,7,12], along with three older ones are included in the comparison using the same protocol. The authors found that the PAD method based on photometric stereo features [7] generalize better to attacks of contact lens of unknown textures, while the BSIF texture-based PAD method [24] performs better in closed-set scenarios. Experiments also show that the fusion method [12] outperforms the other two methods in both known and unknown settings.…”