Fake Iris Images Research Articles

LRFID-Net: A Local-Region-Based Fake-Iris Detection Network for Fake Iris Images Synthesized by a Generative Adversarial Network

Iris recognition is a biometric method using the pattern of the iris seated between the pupil and the sclera for recognizing people. It is widely applied in various fields owing to its high accuracy in recognition and high security. A spoof detection method for discriminating a spoof attack is essential in biometric recognition systems that include iris recognition. However, previous studies have mainly investigated spoofing attack detection methods based on printed or photographed images, video replaying, artificial eyes, and patterned contact lenses fabricated using iris images from information leakage. On the other hand, there have only been a few studies on spoof attack detection using iris images generated through a generative adversarial network (GAN), which is a method that has drawn considerable research interest with the recent development of deep learning, and the enhancement of spoof detection accuracy by the methods proposed in previous research is limited. To address this problem, the possibility of an attack on a conventional iris recognition system with spoofed iris images generated using cycle-consistent adversarial networks (CycleGAN), which was the motivation of this study, was investigated. In addition, a local region-based fake-iris detection network (LRFID-Net) was developed. It provides a novel method for discriminating fake iris images by segmenting the iris image into three regions based on the iris region. Experimental results using two open databases, the Warsaw LiveDet-Iris-2017 and the Notre Dame Contact Lens Detection LiveDet-Iris-2017 datasets, showed that the average classification error rate of spoof detection by the proposed method was 0.03% for the Warsaw dataset and 0.11% for the Notre Dame Contact Lens Detection dataset. The results confirmed that the proposed method outperformed the state-of-the-art methods.

A New Approach for Real-time Iris Liveness Detection Based on Eye Movement’s Features

This research introduces a new approach for iris liveness detection based on eye movement’s features. Given 50 valid eye frames from real-time video stream, this research identify them as the images that were taken whether directly from humans’ eyes or not. Our method ensures the selected images for further processing are eye ones without user’s cooperation during the sampling process. In addition, it also controls the application timeout by identifying either fake iris or low quality eye images. In my study, this Research used two features to model eye movements: (1) the width/height ratio of the eye outline and (2) the distance from the pupil center to the eye centroid. In order to model the acquired data distribution, this research used a bag of two-component Gaussian Mixing Models (GMMs). The class label of the acquired data depends on the GMMs which it belongs. The average accuracy of my proposed method on my and IriTech, Inc.’s databases was 94.704% and 100%, respectively. This result indicated that my method can make real-time iris-based systems work automatically with high security.

Local binary hexagonal extrema pattern (LBHXEP): a new feature descriptor for fake iris detection

Security agencies frequently use biometric traits for automatic recognition of a person. The human iris is the most hopeful biometric authentication that can accurately identify a person from their exclusive features. However, in recent years, different types of spoofing attacks are used to violate the security of a biometric system. Biometrics liveness detection system used to recognize persons in a fast and trustworthy way through the use of unique biological distinctiveness. Presentation of a manufactured article of a human iris in the form of photo attack and contact lens attack could hamper the projected policy of a biometric system. The quality of real and fake iris images shows different textural characteristics. In this paper, we have proposed a novel and proficient feature descriptor, i.e., local binary hexagonal extrema pattern for fake iris detection. The proposed descriptor exploits the relationship between the center pixel and its Hexa neighbor. Hexagonal shape using “six-neighbor approach” is preferable to the rectangular structure due to its higher symmetry, consistent connectivity, and efficient use of space. The proposed consideration also solves the “curse of dimensionality” problem in liveness detection. The proposed descriptor is evaluated on ATVS-FIr DB and IIIT-D CLI databases for iris liveness detection and show promising performance for liveness detection in terms green, brown, etc. of accuracy and average error rate.

Detecting fake iris in iris bio-metric system

Iris recognition is an automated method of biometric identification that uses mathematical pattern-recognition techniques on video images of the irises of an individual's eyes, whose complex random patterns are unique and can be seen from some distance. Now days, Iris is being used widely by several organizations, including governments, for identification and authentication purposes. Aadhar, India's UID project uses Iris scan along with fingerprints to uniquely identify people and allocate a Unique Identification Number. Most of the work done in the area of Iris pattern recognition systems emphasizes only on matching of the patterns with the stored templates. Security aspects of the system are still unexplored. The available security algorithms provide only some cryptographic solutions that keeps the template database in a secret cryptographic form. We successfully enhanced the detection of fake iris images and add the provision of detection of false of scanned iris images as template. This enhanced significantly the performance of the system in terms of security and reliability. We use Flash and motion detection of natural eye to detect the liveliness of real iris images before matching from stored templates.

Entropy with Local Binary Patterns for Efficient Iris Liveness Detection

Iris anti-spoofing is one of the most important topics, in which the development is increasing rapidly. This paper introduces an efficient system for detecting iris attacks. The system avoids the segmentation and the normalization stages employed traditionally in fake detection systems. Wavelet packets (WPs) are used to decompose the original image into wavelet approximation and detail channels. Entropy values are extracted from the wavelet channels, and also from the local binary pattern (LBP) images of the channels. These features are used for discriminating between real and fake iris images. Support vector machines are used for the classification purpose. The aim is to contribute for improved classification accuracy with less computational complexity and reduced processing time. Entropy of the WP channels gives 99.9237% classification accuracy, and the entropy of the LBP images yields 99.781%, using ATVS-FIr-DB. Fusion of these features yields 100% classification accuracy. Entropy of the wavelet channels is sufficient to obtain 100% accuracy using CASIA-Iris-Syn database, without fusion. All images in both databases are used, without the need to discard images with unsuccessful segmentation. Segmented images from both databases are used for comparison. Results show that more discriminative features can be obtained using the proposed algorithm. System complexity and processing time are reduced noticeably, and the system is robust to different types of fakes.

Iris Image Classification Based on Hierarchical Visual Codebook

Iris recognition as a reliable method for personal identification has been well-studied with the objective to assign the class label of each iris image to a unique subject. In contrast, iris image classification aims to classify an iris image to an application specific category, e.g., iris liveness detection (classification of genuine and fake iris images), race classification (e.g., classification of iris images of Asian and non-Asian subjects), coarse-to-fine iris identification (classification of all iris images in the central database into multiple categories). This paper proposes a general framework for iris image classification based on texture analysis. A novel texture pattern representation method called Hierarchical Visual Codebook (HVC) is proposed to encode the texture primitives of iris images. The proposed HVC method is an integration of two existing Bag-of-Words models, namely Vocabulary Tree (VT), and Locality-constrained Linear Coding (LLC). The HVC adopts a coarse-to-fine visual coding strategy and takes advantages of both VT and LLC for accurate and sparse representation of iris texture. Extensive experimental results demonstrate that the proposed iris image classification method achieves state-of-the-art performance for iris liveness detection, race classification, and coarse-to-fine iris identification. A comprehensive fake iris image database simulating four types of iris spoof attacks is developed as the benchmark for research of iris liveness detection.

Fake iris detection method using Purkinje images based on gaze position

Until now, most research on iris recognition has been focused on recognition algorithms and iris camera systems. There has been little research into fake iris detection, although recently its importance has been greatly emphasized. Fake iris detection refers to the process of detecting and defeating fake iris images. In this work, we propose a new method of defeating fake iris attacks using Purkinje images based on gaze position. Our research presents the following four improvements over previous works. First, we calculate the theoretical positions and distances between the Purkinje images based on the 3-D human eye model. Second, by using these positions and distances (which changed according to gaze positions), we design a more robust way of detecting fake irises. Third, since it is not necessary to align the center of the user's eyeball with the optical axis of the camera, the proposed method can be used in practical iris systems. Fourth, by activating the illumination infrared-light-emitting diode (IR-LED), the distance-measuring IR-LED and the Purkinje IR-LED alternatively, we obtain accurate positions for the Purkinje images according to the user's gaze position. Experimental results show that the false rejection rate (FRR) is 0.2% and the false acceptance rate (FAR) is 0.2%.

Multifeature-based fake iris detection method

The purpose of fake iris detection is to discriminate between real and fake iris images and to defeat fake (forged) iris images. A robust fake iris detection method should be able to detect various types of fake iris images obtained from a fake printed iris, an artificial eye, or a fake contact lens, correctly and nonintrusively. To solve the problem, we propose a new fake iris detection method. We measure distinctive physiological multifeatures [the first and second features refer to the reflectance ratios of the iris to the sclera (RRIS) at 750 and 850 nm, respectively, and the third feature refers to the thickness of the corneoscleral limbus], and classify those features extracted from live irises and fake irises using a support vector machine (SVM). Using the proposed method, we can discriminate various types of fake iris images without inconveniencing users by shining visible light. To measure the performance of the method, three types of fake irises are made: a printed iris, an artificial eye, and a fake contact lens. Our experimental results show that it is possible to detect those fake iris images with high accuracy.