Iris Recognition Performance Research Articles

Study the effect of eye diseases on the performance of iris segmentation and recognition using transfer deep learning methods

A new deep learning-based iris recognition system is presented in the current study in the case of eye disease. Current state of art iris segmentation is either based on traditional low accuracy algorithms or heavy-weight deep-based models. In the current study segmentation section, a new iris segmentation method based on illumination correction and a modified circular Hough transform is proposed. The current method also performs a post-processing step to minimize the false positives. Besides, a ground truth of iris images is constructed to evaluate the segmentation accuracy. Many deep learning models (GoogleNet, Inception_ResNet, XceptionNet, EfficientNet, and ResNet50) are applied through the recognition step using the transfer learning approach. In the experiment part, two eye disease-based datasets are used. 684 iris images of individuals with multiple ocular diseases from the Warsaw BioBase V1 and 1,793 iris images from the Warsaw BioBase V2 are also used. The CASIA V3 Interval Iris dataset, which contains 2,639 photographs of healthy iris, is used to train deep models once, and then the transfer learning of this normal-based eye dataset is used to retrain the same deep models using Warsaw BioBase datasets. Different scenarios for training and evaluating participants are used during experiments. The trained models are evaluated using validation accuracy, training time, TPR, FNR, PPR, FDR, and test accuracy. The best accuracies are 98.5% and 97.26%, which are recorded by the ResNet50 (2-layer of transfer learning) model trained on Warsaw BioBase V1 and V2, respectively. Results indicate that the effect of eye diseases is concentrated on the segmentation phase. For recognition, no significant impact is recognized. Some disease that affects the structure (bloody eyes, trauma, iris pigment) can affect the iris recognition step partially. Our study is compared with similar studies in the case of eye diseases. The comparison proves the efficiency and high performance of the proposed methodology against all previous models on the same iris datasets.

Improvement of iris recognition technology for biometric identification of a person

This topic is very relevant in the field of artificial intelligence as a direction of pattern recognition. In this work, the iris of the eye is considered as an image. Artificial intelligence makes this technology more accessible for use in CCTV cameras, smartphones and various areas of human activity. The article reflects the results of a study of methods and technologies of pattern recognition on the example of the human iris. The aim of the work was to study methods and technologies for human iris recognition and iris recognition of employees of a particular organization using EyeLock equipment by comparing segmentation results with Daugman standard segmentation. Comparison analysis of segmentation results with standard segmentation can be done by directly measuring the number of correctly segmented irises in both methods, or by indirectly measuring the effect of segmentation on iris recognition performance. The method using the Daugman integral-differential operator has the greatest efficiency. The performance of the neural network has been improved. To use a neural network to classify iris profiles, we selected sets of images (images per person) as training images, and the rest of the images were used as test images. Training time (in seconds): for the Daugman method 170.7, and for the parabolic method 204.7. The Daugman integro-differential operator is applied to the captured image to obtain the "maximum integral derivative of the contour" with ever-increasing radius on "successively decreasing scales" in three parameters: center coordinates and radius. Finding the maximum when the search coordinates deviate along an unwinding spiral. Methods and techniques for pattern recognition have been investigated using the human iris

A Symmetrical Encoder-Decoder Network with Transformer for Noise-Robust Iris Segmentation

Iris images captured in less-constrained environments are susceptible to the interference of noise such as specular reflections, eyelash and hair occlusions, motion and defocus blur, which makes it hard to accurately segment the valid iris region. To solve the problem, a symmetrical encoder-decoder network with Transformer is proposed for noise-robust iris segmentation. First, Swin Transformer is employed as the encoder to feed a sequence of the input image patches into hierarchical Transformer modules, so as to model the long-range dependencies of image pixels through self-attention mechanism, and enhance the interaction of contextual information. Secondly, a Transformer decoder which is symmetrical with the encoder is constructed, where the high-order context features extracted earlier are decoded in multiple layers. Besides, the skip connections are introduced to fuse the multi-scale features from the encoder with the up-sampled decoded features, which reduces the loss of spatial position information caused by down-sampling. Finally, supervised learning is carried out on each stage of the decoder, which improves the quality of extracted different-scaled features. A comparative experiment is carried out on three challenging noisy near-infrared (NIR) and visible (VIS) iris datasets, i.e., NICE.I, CASIA.v4-distance, and MICHE-I. Results show that the proposed method achieves better segmentation performance than several benchmark methods including traditional methods, convolutional neural network-based methods and existing Transformer-based methods on multiple evaluation metrics like <italic>E</italic>₁, <italic>E</italic>₂, <italic>F</italic>₁, and MIOU, and particularly demonstrates significant advantages in reducing the interference of adverse noise. The iris recognition experiment on the CASIA.v4-distance dataset also shows that the proposed method can effectively improve the performance of iris recognition, suggesting a good application potential.

Accurate Iris Segmentation for Non-Cooperative Iris Recognition through Multi-task Network

Iris image suffer from noise due to non-cooperative situation. It creates challenge for segmentation. Here we propose deep learning based efficient segmentation method. It provides complete solution for such situation. To train this we have used various databases, which involved various illumination and imaging sensors. The proposed segmentation method can be utilized for all iris recognition methodology to improve performance in non-cooperative iris recognition. Key words: Iris Segmentation, multi task learning, iris localization, iris recognition

Open‐set iris recognition based on deep learning

The existing iris recognition methods offer excellent recognition performance for known classes, but they do not consider the rejection of unknown classes. It is important to reject an unknown object class for a reliable iris recognition system. This study proposes open-set iris recognition based on deep learning. In the method, by training the deep network, the extracted iris features are clustered near the feature centre of each kind of iris image. Then, the authors build an open-class features outlier network (OCFON) containing distance features, which maps the features extracted by the deep network to a new feature space and classifies them. Finally, the unknown class samples are determined by a SoftMax probability threshold. The authors conducted experiments on the open iris dataset constructed using the iris datasets CASIA-Iris-Twins and CASIA-Iris-Lamp. The experiment shows that the proposed method has good open-set iris recognition performance, can effectively distinguish iris samples of unknown classes, and has little impact on the recognition ability of known classes of iris samples.

Towards More Discriminative and Robust Iris Recognition by Learning Uncertain Factors

The uncontrollable acquisition process limits the performance of iris recognition. In the acquisition process, various inevitable factors, including eyes, devices, and environment, hinder the iris recognition system from learning a discriminative identity representation. This leads to severe performance degradation. In this paper, we explore uncertain acquisition factors and propose uncertainty embedding (UE) and uncertainty-guided curriculum learning (UGCL) to mitigate the influence of acquisition factors. UE represents an iris image using a probabilistic distribution rather than a deterministic point (binary template or feature vector) that is widely adopted in iris recognition methods. Specifically, UE learns identity and uncertainty features from the input image, and encodes them as two independent components of the distribution, mean and variance. Based on this representation, an input image can be regarded as an instantiated feature sampled from the UE, and we can also generate various virtual features through sampling. UGCL is constructed by imitating the progressive learning process of newborns. Particularly, it selects virtual features to train the model in an easy-to-hard order at different training stages according to their uncertainty. In addition, an instance-level enhancement method is developed by utilizing local and global statistics to mitigate the data uncertainty from image noise and acquisition conditions in the pixel-level space. The experimental results on six benchmark iris datasets verify the effectiveness and generalization ability of the proposed method on same-sensor and cross-sensor recognition.

Statistical Evaluation of Image Quality Measures for Improving Iris Recognition Performance

Statistical Evaluation of Image Quality Measures for Improving Iris Recognition Performance

Iris-ocular-periocular: toward more accurate biometrics for off-angle images

Iris is one of the most well-known biometrics; it is a nonintrusive and contactless authentication technique with high accuracy, enhanced security, and unique distinctiveness. However, its dependence on image quality and its frontal image acquisition requirement limit its recognition performance and hinder its potential use in standoff applications. Standoff biometric systems require a less controlled environment than traditional systems, so their captured images will likely be nonideal, including off-angle. We present convolutional neural network (CNN)-based deep learning frameworks to improve the recognition performance of iris, ocular, and periocular biometric modalities for off-angle images. Our contribution is fourfold: first, the performances of popular AlexNet, GoogLeNet, and ResNet50 architectures are presented for off-angle biometrics. Second, we study the effect of the gaze angle difference between training and testing images on iris, ocular, and periocular recognitions. Third, we investigate the network behavior for untrained gaze angles and the information fusion capability of CNN networks at multiple off-angle images. Finally, deep learning-based results are compared with a traditional iris recognition algorithm using the gallery approach. Our results with off-angle images ranging from −50 deg to 50 deg in gaze angle show that the proposed methods improve the recognition performance of iris, ocular, and periocular recognition.

Local gradient pattern and deep learning-based approach for the iris recognition at-a-distance

One of the biometric techniques utilized to predict the human is based on the iris. The recognition of iris is performed by discovering an individual without human intervention utilizing the iris of human eyes. Iris offers distinct information about the person. This research presents deep learning strategy for performing iris recognition. Primarily, image is pre-processed to obtain exact region iris. Then, region of iris is extracted using Hough Transform, followed by segmentation and normalization of iris region using Daugman’s rubber sheet model. Once segmentation is performed, the features are generated with ScaT-LOOP that is the combination of Scattering Transform (ST), Tetrolet transforms (TT), Local Gradient Pattern (LGP) and Local Optimal Oriented Pattern (LOOP). Finally, steepest gradient-based Deep Belief Network (DBN) is utilized for recognizing the iris. The performance of iris recognition using the DBN classifier is computed based on accuracy, False Rejection Rate (FRR), and False Acceptance Rate (FAR). The proposed method achieves maximum accuracy of 97.96%, minimal FAR of 0.493%, and minimal FRR of 0.48% that indicates its superiority.

Iris and periocular biometrics for head mounted displays: Segmentation, recognition, and synthetic data generation

Augmented and virtual reality deployment is finding increasing use in novel applications. Some of these emerging and foreseen applications allow the users to access sensitive information and functionalities. Head Mounted Displays (HMD) are used to enable such applications and they typically include eye facing cameras to facilitate advanced user interaction. Such integrated cameras capture iris and partial periocular region during the interaction. This work investigates the possibility of using the captured ocular images from integrated cameras from HMD devices for biometric verification, taking into account the expected limited computational power of such devices. Such an approach can allow user to be verified in a manner that does not require any special and explicit user action. In addition to our comprehensive analyses, we present a light weight, yet accurate, segmentation solution for the ocular region captured from HMD devices. Further, we benchmark a number of well-established iris and periocular verification methods along with an in-depth analysis on the impact of iris sample selection and its effect on iris recognition performance for HMD devices. To the end, we also propose and validate an identity-preserving synthetic ocular image generation mechanism that can be used for large scale data generation for training purposes or attack generation purposes. We establish the realistic image quality of generated images with high fidelity and identity preserving capabilities through benchmarking them for iris and periocular verification.

Cross-spectral iris recognition using phase-based matching and homomorphic filtering

In cross-spectral iris recognition, different spectral bands are used to obtain rich information of the human iris. Previous studies on cross-spectral iris recognition are based primarily on feature-based approaches, which are prone to the changes in parameters in the feature extraction process, such as spatial position and iris image acquisition conditions. These parameters can degrade iris recognition performance. In this paper, we present a phase-based approach for cross-spectral iris recognition using phase-only correlation (POC) and band-limited phase-only correlation (BLPOC). A phase-based iris recognition system recognizes an iris using the phase information contained in the iris image; therefore, its performance is not affected by feature extraction parameters. However, the performance of a phase-based cross-spectral iris recognition is strongly influenced by specular reflection. Different illumination conditions may produce different iris images from the same subject. To overcome this challenge, we integrate a photometric normalization technique –homomorphic filtering– with phase-based cross-spectral iris recognition. The experimental results reveal that the proposed technique achieved an excellent matching performance with an equal error rate of 0.59% and a genuine acceptance rate of 95%.

Development and Analysis of a Zeta Method for Low-Cost, Camera-based Iris Recognition

Iris recognition is an alternative authentication method. Many studies have tried to improve iris recognition as a biometric-based alternative for secure authentication. Iris segmentation is an important part of iris recognition because it defines the image region that is used for subsequent processing such as feature extraction and matching, hence directly affects the overall iris recognition performance. This work focuses on the development of an authentication system using localization methods and half-polar normalization of the iris. The proposed Zeta method uses a new model of eye segmentation and normalization that can be used simultaneously on both eyes, considering different iris patterns in those two eyes. There are seven variants of the proposed and tested Zeta method: Zeta-v1, Zeta-v2, Zeta-v3, Zeta-v4, Zeta-v5, Zeta-v6, and Zeta-v7. Overall, the method achieved an average segmentation time performance of 0.0138427 seconds. The most accurate rate was by the Zeta-v1 method, with a value threshold of 100% on the wrong rejection rate and 94.9% on the correct acceptance rate.

Towards Complete and Accurate Iris Segmentation Using Deep Multi-Task Attention Network for Non-Cooperative Iris Recognition

Iris images captured in non-cooperative environments often suffer from adverse noise, which challenges many existing iris segmentation methods. To address this problem, this paper proposes a high-efficiency deep learning based iris segmentation approach, named IrisParseNet . Different from many previous CNN-based iris segmentation methods, which only focus on predicting accurate iris masks by following popular semantic segmentation frameworks, the proposed approach is a complete iris segmentation solution, i.e., iris mask and parameterized inner and outer iris boundaries are jointly achieved by actively modeling them into a unified multi-task network. Moreover, an elaborately designed attention module is incorporated into it to improve the segmentation performance. To train and evaluate the proposed approach, we manually label three representative and challenging iris databases, i.e., CASIA.v4-distance, UBIRIS.v2, and MICHE-I, which involve multiple illumination (NIR, VIS) and imaging sensors (long-range and mobile iris cameras), along with various types of noises. Additionally, several unified evaluation protocols are built for fair comparisons. Extensive experiments are conducted on these newly annotated databases, and results show that the proposed approach achieves state-of-the-art performance on various benchmarks. Further, as a general drop-in replacement, the proposed iris segmentation method can be used for any iris recognition methodology, and would significantly improve the performance of non-cooperative iris recognition.

A 3D Iris Scanner From a Single Image Using Convolutional Neural Networks

A 3D model of the human iris provides an additional degree of freedom in iris recognition, which could help identify people in larger databases, even when only a piece of the iris is available. Previously, we reported developing a 3D iris scanner that uses 2D images of the iris from multiple perspectives to reconstruct a 3D model of the iris. This paper focuses on the development of a 3D iris scanner from a single image by means of a Convolutional Neural Network (CNN). The method is based on a depth-estimation CNN for the 3D iris model. A dataset of 26,520 real iris images from 120 subjects, and a dataset of 72,000 synthetic iris images with their aligned depthmaps were created. With these datasets, we trained and compared the depth estimation capabilities of available CNN architectures. We analyzed the performance of our method to estimate the iris depth in multiple ways: using real step pyramid printed 3D models, comparing the results to those of a test set of synthetic images, comparing the results to those of the OCT scans from both eyes of one subject, and generating the 3D rubber sheet from the 3D iris model proving the correspondence with the resulting 2D rubber sheet and binary codes. On a preliminary test the proposed 3D rubber sheet model increased iris recognition performance by 48% with respect to the standard 2D iris code. Other contributions include assessing the scanning resolution, reducing the acquisition and processing time to produce the 3D iris model, and reducing the complexity of the image acquisition system.

Enhancing human iris recognition performance in unconstrained environment using ensemble of convolutional and residual deep neural network models

Despite the prominent advancements in iris recognition, unconstrained image acquisition through heterogeneous sensors has been a major obstacle in applying it for large-scale applications. In recent years, deep convolutional networks have achieved remarkable performance in the field of computer vision and have been employed in iris applications. In this study, three distinct models based on the ensemble of convolutional and residual blocks are proposed to enrich heterogeneous (cross-sensor) iris recognition. In order to analyze their quantitative performances, extensive experiments are carried out on two publicly available iris databases, ND-iris-0405 dataset and ND-CrossSensor-Iris-2013 dataset. Further, the final model has been scrutinized based on the least error rate and then fused using score-level fusion with two preeminent feature extraction methods, i.e., scale-invariant feature transform and binarized statistical information features. The resultant model is examined for cross-sensor iris recognition and reported the top two error rates as 1.01% and 1.12%. It infers that the proposed approach constitutes vital discerning iris features and can recognize that the micro-patterns exist inside the iris region. Furthermore, a comparative study is carried out with the state of the art, where the proposed approach obtains significantly improved performance.

Fusion Iris and Periocular Recognitions in Non-Cooperative Environment

The performance of iris recognition in non-cooperative environment can be negatively impacted when the resolution of the iris images is low which results in failure to determine the eye center, limbic and pupillary boundary of the iris segmentation. Hence, a combination with periocular features is suggested to increase the authenticity of the recognition system. However, the texture feature of periocular can be easily affected by a background complication while the colour feature of periocular is still limited to spatial information and quantization effects. This happens due to different distances between the sensor and the subject during the iris acquisition stage as well as image size and orientation. The proposed method of periocular feature extraction consists of a combination of rotation invariant uniform local binary pattern to select the texture features and a method of color moment to select the color features. Besides, a hue-saturation-value channel is selected to avoid loss of discriminative information in the eye image. The proposed method which consists of combination between texture and colour features provides the highest accuracy for the periocular recognition with more than 71.5% for the UBIRIS.v2 dataset and 85.7% for the UBIPr dataset. For the fusion recognitions, the proposed method achieved the highest accuracy with more than 85.9% for the UBIRIS.v2 dataset and 89.7% for the UBIPr dataset.

Fusion Iris and Periocular Recognitions in Non-Cooperative Environment

The performance of iris recognition in non-cooperative environment can be negatively impacted when the resolution of the iris images is low which results in failure to determine the eye center, limbic and pupillary boundary of the iris segmentation. Hence, a combination with periocular features is suggested to increase the authenticity of the recognition system. However, the texture feature of periocular can be easily affected by a background complication while the colour feature of periocular is still limited to spatial information and quantization effects. This happens due to different distances between the sensor and the subject during the iris acquisition stage as well as image size and orientation. The proposed method of periocular feature extraction consists of a combination of rotation invariant uniform local binary pattern to select the texture features and a method of color moment to select the color features. Besides, a hue-saturation-value channel is selected to avoid loss of discriminative information in the eye image. The proposed method which consists of combination between texture and colour features provides the highest accuracy for the periocular recognition with more than 71.5% for the UBIRIS.v2 dataset and 85.7% for the UBIPr dataset. For the fusion recognitions, the proposed method achieved the highest accuracy with more than 85.9% for the UBIRIS.v2 dataset and 89.7% for the UBIPr dataset.

Assessing the Impact of Corneal Refraction and Iris Tissue Non-Planarity on Iris Recognition

Assumptions regarding eye morphology are implicit in iris recognition algorithms. The cornea is assumed to have little to no effect on the view of the iris texture when the eye gaze is non-frontal, the iris is assumed flat, and the eye is assumed to be imaged via an orthographic projection. If these assumptions hold, affine transformations may be used to rectify non-frontally posed images to a frontal view and the rubber sheet model may be accurately used to normalize the iris annulus to a rectangular image. This paper examines how iris recognition performance degrades when the first two assumptions are violated. Using a computer renderable eye model, a data set is created varying the presence of the cornea and a parameterized non-planarity of the iris shape across a large range of eye gaze angles. Matching scores are created using a commercial matcher. When comparing the relative impact of each assumption violation, it is observed that iris non-planarity presents a more significant problem than corneal refractive distortion with regard to iris recognition accuracy in non-frontal images.

Iris Recognition After Death

This paper presents a comprehensive study of post-mortem human iris recognition carried out for 1200 near-infrared and 1787 visible-light samples collected from 37 deceased individuals kept in mortuary conditions. We used four independent iris recognition methods (three commercial and one academic) to analyze genuine and impostor comparison scores and check the dynamics of iris quality decay over a period of up to 814 h after death. This study shows that post-mortem iris recognition may be close-to-perfect approximately 5–7 h after death and occasionally is still viable even 21 days after death. These conclusions contradict the statements present in the past literature that the iris is unusable as a biometrics shortly after death, and show that the dynamics of post-mortem changes to the iris that are important for biometric identification are more moderate than previously hypothesized. This paper contains a thorough medical commentary that helps to understand which post-mortem metamorphoses of the eye may impact the performance of automatic iris recognition. An important finding is that false-match probability is higher when live iris images are compared with post-mortem samples than when only live samples are used in comparisons. This paper conforms to reproducible research and the database used in this study is made publicly available to facilitate research on post-mortem iris recognition. To the best of our knowledge, this paper offers the most comprehensive evaluation of post-mortem iris recognition and the largest database of post-mortem iris images.

Advances in Cross-Spectral Iris Recognition Using Integrated Gradientface-Based Normalization

Cross-spectral iris recognition represents the ability of the system to identify the iris images acquired in different electromagnetic spectrums. An iris captured in the near-infrared spectrum (NIR) is matched with an iris obtained in the visual light spectrum (VIS) to boost the recognition performance. In cross-spectral iris recognition, the illumination factor between NIR and VIS images significantly degrades the recognition performance. Therefore, the existing method only achieved recognition performance with an equal error rate (EER) larger than 5%, and it is a challenging issue for cross-spectral performance to have EER below 5%. In this paper, we improve iris recognition performance by concatenating the Gradientfaces-based normalization technique (GRF) to a standard (conventional) iris recognition method to alleviate the illumination effect. In addition, we integrate the GRF with a Gabor filter, a difference of Gaussian (DoG) filter, and texture descriptors, namely a binary statistical image feature (BSIF) and a local binary pattern (LBP). The experimental results show that the GRF can boost the cross-spectral iris recognition performance with an EER equals to 1.69%. In addition, the best cross-spectral iris recognition performance is achieved when the GRF is integrated with the Gabor filter and the BSIF.