Gaze estimation for off-angle iris recognition based on the biometric eye model

The popularity of the iris biometric has grown considerably over the past two to three years. Most research has been focused on the development of new iris processing and recognition algorithms for frontal view iris images. However, a few challenging directions in iris research have been identified, including processing of a nonideal iris and iris at a distance. In this paper, we describe two nonideal iris recognition systems and analyze their performance. The word "nonideal" is used in the sense of compensating for off-angle occluded iris images. The system is designed to process nonideal iris images in two steps: 1) compensation for off-angle gaze direction and 2) processing and encoding of the rotated iris image. Two approaches are presented to account for angular variations in the iris images. In the first approach, we use Daugman's integrodifferential operator as an objective function to estimate the gaze direction. After the angle is estimated, the off-angle iris image undergoes geometric transformations involving the estimated angle and is further processed as if it were a frontal view image. The encoding technique developed for a frontal image is based on the application of the global independent component analysis. The second approach uses an angular deformation calibration model. The angular deformations are modeled, and calibration parameters are calculated. The proposed method consists of a closed-form solution, followed by an iterative optimization procedure. The images are projected on the plane closest to the base calibrated plane. Biorthogonal wavelets are used for encoding to perform iris recognition. We use a special dataset of the off-angle iris images to quantify the performance of the designed systems. A series of receiver operating characteristics demonstrate various effects on the performance of the nonideal-iris-based recognition system.

Standoff iris recognition systems have been developed to identify not only fully cooperative individuals but also non-cooperative moving subjects. Since iris images are captured in much less constrained conditions, they are likely to be non-ideal including being off-angle. However, traditional iris recognition systems are designed for only frontal iris images. Therefore, the existing databases save binary iris codes using only frontal images for iris matching. To use these existing iris codes, standoff systems need to be compatible with existing iris matching algorithms by generating the frontal version of iris codes even for off-angle images. Since gaze estimation is the first essential component to reconstruct the frontal view of off-angle iris images, we first need to estimate the eye gaze direction from a single off-angle iris image. This paper presents a deep learning-based gaze angle estimation method for standoff iris recognition frameworks. As the main contribution, our approach will allow us to estimate the gaze angle from an off-angle image using transfer learning and a regression model without segmentation and using additional hardware. The proposed CNN-based gaze estimation method is trained and tested with 108,600 off-angle images from 100 subjects. It shows high accuracy in gaze estimation with an average error of 4° in angle for off-angle images captured from every angle ranging -50° to +50° even if training images are not available for all angles.

It has been proven that hamming distance score between frontal and off-angle iris images of same eye differs in iris recognition system. The distinction of hamming distance score is caused by many factors such as image acquisition angle, occlusion, pupil dilation, and limbus effect. In this paper, we first study the effect of the angle variations between iris plane and the image acquisition systems. We present how hamming distance changes for different off-angle iris images even if they are coming from the same iris. We observe that increment in acquisition angle of compared iris images causes the increment in hamming distance. Second, we propose a new technique in off-angle iris recognition system that includes creating a gallery of different off-angle iris images (such as, 0, 10, 20, 30, 40, and 50 degrees) and comparing each probe image with these gallery images. We will show the accuracy of the gallery approach for off-angle iris recognition.

The use of iris as biometric trait has emerged as one of the most preferred method because of its uniqueness, lifetime stability and regular shape. Moreover it shows public acceptance and new user-friendly capture devices are developed and used in a broadened range of applications. Currently, iris recognition systems work well with frontal iris images from cooperative users. Nonideal iris images are still a challenge for iris recognition and can significantly affect the accuracy of iris recognition systems. In this paper, we propose a method to correct off-angle iris image. Taking into account the eye morphology and the reflectance properties of the external transparent layers, we can evaluate the distorting effect that is present in the acquired image. The correction algorithm proposed includes a first modeling phase of the human eye, a segmentation of the acquired image, and a simulation phase where the acquisition geometry is reproduced and the distortions are evaluated. Finally we obtain an image which does not contain the distorting effects due to jumps in the refractive index. We show how this correction process reduce the intra-class variations for off-angle iris images.

This paper investigates the effect of segmentation errors on off-angle iris recognition. We first segment the inner and outer boundaries of off-angle iris images by fitting the best ellipses to boundaries. Then, we validate the segmentation results using Ground Truth Tool and fix the possible errors. Second, we add random errors on iris segmentation parameters to examine how Hamming distance distribution changes for different amount of segmentation error. In order to add errors, we first group the parameters into three sets as ellipse center (i.e., x, y), minor and major axis (i.e., r 1 , r 2 ) and orientation (θ) and changed the parameters by adding positive and negative random error as a noise to the ground truth segmentation parameters. Our purpose is to show how performance of iris recognition is affected if there is an error in segmentation and how important is to have a very robust segmentation algorithm for off-angle iris recognition. We test the effect of segmentation errors using our off-angle iris dataset that contains images from -50° to +50° in gaze angles. Based on our experimental results, we observed that error in ellipse center decrease the recognition performance worse than minor and major axis and orientation. The main reason is the normalization process tolerates the small amount of error in minor and major axis and bit shift in matching handles the small variations in the orientation errors. However, if the error in the ellipse orientation exceeds the limit of the bit shift method tolerance, the performance of the off-angle iris recognition dramatically drops.

The use of iris as biometric trait has emerged as one of the most preferred method because of its uniqueness, lifetime stability and regular shape. Moreover it shows public acceptance and new user-friendly capture devices are developed and used in a broadened range of applications. Currently, iris recognition systems work well with frontal iris images from cooperative users. Nonideal iris images are still a challenge for iris recognition and can significantly affect the accuracy of iris recognition systems. Moreover, accurate localization of different eye’s parts from videos or still images is a crucial step in many image processing applications that range from iris recognition in Biometrics to gaze estimation for Human Computer Interaction (HCI), impaired people aid or, even, marketing analysis for products attractiveness. Notwithstanding this, actually, most of available implementations for eye’s parts segmentation are quite invasive, imposing a set of constraints both on the environment and on the user itself limiting their applicability to high security Biometrics or to cumbersome interfaces. In the first part of this Chapter, we propose a novel approach to segment the sclera, the white part of the eye. We concentrate on this area since, thanks to the dissimilarity with other eye’s parts, its identification can be performed in a robust way against light variations, reflections and glasses lens flare. An accurate sclera segmentation is a fundamental step in iris and pupil localization, even in non-frontal noisy images. Furthermore its particular geometry can be fruitfully used for accurate eyeball rotation estimation. The proposed technique is based on a statistical approach (supported by some heuristic assumptions) to extract discriminating descriptors for sclera and non-sclera pixels. A Support Vector Machine (SVM) is then used as a final supervised classifier. Once the eyeball rotation angle respect to the camera optical axis is estimated and the limbus (the boundary between the iris and the sclera) is extracted, we propose a method to correct off-angle iris image. Taking into account the eye morphology and the reflectance properties of the external transparent layers, we can evaluate the distorting effects that are present in the acquired image. The correction algorithm proposed includes a first modeling phase of the human eye and a simulation phase where the acquisition geometry is reproduced and the distortions are evaluated. Finally we obtain an image which does not contain the distorting effects due to jumps in the refractive index. We show how this correction process reduces the intra-class variations for off-angle iris images.

Even with trained operators and cooperative subjects, it is still possible to capture off-angle iris images. Considering the recent demands for stand-off iris biometric systems and the trend towards "on-the-move-acquisition", off-angle iris recognition became a hot topic within the biometrics community. In this work, CNNs trained with the triplet loss function are applied to extract features for iris recognition. To analyze which parts of the eye are most suited for the CNN-based recognition system, experiments are carried out using image data from different parts of the eye (full eye, eye zoomed to iris, iris only, iris normalized, eye without iris). To analyze the impact of different gaze angles on the recognition performance, experiments are applied on: (1) different gaze angles separately, (2) image data with increasing differences in the gaze angles, and (3) corrected off-angle image data. The experiment results show superior performance of the CNN trained with the triplet loss on the iris images with more lateral gaze angles (≥ 30°). However, higher differences in the gaze angles between images deteriorate the network performance. Also, the results are about the same for the different parts of the eye and correcting the gaze angle did not really improve the performance of the CNN.

The ability to identify individuals based on their iris is known as iris recognition. Over the past decade iris recognition has garnered much attention because of its strong performance in comparison with other mainstream biometrics such as fingerprint and face recognition. Performance of iris recognition systems is driven by application scenario requirements. Standoff distance, subject cooperation, underlying optics, and illumination are a few examples of these requirements which dictate the nature of images an iris recognition system has to process. Traditional iris recognition systems, dubbed "stop and stare", operate under highly constrained conditions. This ensures that the captured image is of sufficient quality so that the success of subsequent processing stages, segmentation, encoding, and matching are not compromised. When acquisition constraints are relaxed, such as for surveillance or iris on the move, the fidelity of subsequent processing steps lessens.;In this dissertation we propose a multi-faceted framework for mitigating the difficulties associated with non-ideal iris. We develop and investigate a comprehensive iris image quality metric that is predictive of iris matching performance. The metric is composed of photometric measures such as defocus, motion blur, and illumination, but also contains domain specific measures such as occlusion, and gaze angle. These measures are then combined through a fusion rule based on Dempster-Shafer theory. Related to iris segmentation, which is arguably one of the most important tasks in iris recognition, we develop metrics which are used to evaluate the precision of the pupil and iris boundaries. Furthermore, we illustrate three methods which take advantage of the proposed segmentation metrics for rectifying incorrect segmentation boundaries. Finally, we look at the issue of iris image interoperability and demonstrate that techniques from the field of hardware fingerprinting can be utilized to improve iris matching performance when images captured from distinct sensors are involved.

We describe and analyze the performance of a non-ideal iris recognition system. The system is designed to process non-ideal iris images in two steps: (i) estimation of the gaze direction and (ii) processing and encoding of the rotated iris image. We use two objective functions to estimate the gaze direction: Hamming distance and Daugman's integro-differential operator and determine an estimated angle by picking the value that optimizes the selected objective function. After the angle is estimated, the off-angle iris image undergoes geometric transformations involving the estimated angle and is further processed as if it were a frontal view image. The encoding technique developed in this work is based on application of the global independent component analysis (ICA) to masked iris images. We use two datasets: CASIA dataset and a special dataset of off-angle iris images collected at WVU to verify the performance of the encoding technique and angle estimator, respectively. A series of receiver operating characteristics (ROCs) demonstrates various effects on the performance of the non-ideal iris based recognition system implementing the global ICA encoding.

The human iris is traditionally imaged in Near Infrared (NIR) wavelengths (700nm-900nm) for iris recognition. The absorption co-efficient of color inducing pigment in iris, called Melanin, decreases after 700nm thus minimizing its effect when iris is imaged at wavelengths greater than 700nm. This thesis provides an overview and explores the efficacy of iris recognition at different wavelength bands ranging from visible spectrum (450nm-700nm) to NIR (700nm-900nm) and Short Wave Infrared (900nm-1600nm). Different matching methods are investigated at different wavelength bands to facilitate cross-spectral iris recognition.;The iris recognition analysis in visible wavelengths provides a baseline performance when iris is captured using common digital cameras. A novel blob-based matching algorithm is proposed to match RGB (visible spectrum) iris images. This technique generates a match score based on the similarity between blob like structures in the iris images. The matching performance of the blob based matching method is compared against that of classical 'Iris Code' matching method, SIFT-based matching method and simple correlation matching, and results indicate that the blob-based matching method performs reasonably well. Additional experiments on the datasets show that the iris images can be matched with higher confidence for light colored irides than dark colored irides in the visible spectrum.;As part of the analysis in the NIR spectrum, iris images captured in visible spectrum are matched against those captured in the NIR spectrum. Experimental results on the WVU multispectral dataset show promise in achieving a good recognition performance when the images are captured using the same sensor under the same illumination conditions and at the same resolution. A new proprietary 'FaceIris' dataset is used to investigate the ability to match iris images from a high resolution face image in visible spectrum against an iris image acquired in NIR spectrum. Matching in 'FaceIris' dataset presents a scenario where the two images to be matched are obtained by different sensors at different wavelengths, at different ambient illumination and at different resolution. Cross-spectral matching on the 'FaceIris' dataset presented a challenge to achieve good performance. Also, the effect of the choice of the radial and angular parameters of the normalized iris image on matching performance is presented. The experiments on WVU multispectral dataset resulted in good separation between genuine and impostor score distributions for cross-spectral matching which indicates that iris images in obtained in visible spectrum can be successfully matched against NIR iris images using 'IrisCode' method.;Iris is also analyzed in the Short

Personal identification based on biometrics technology is a trend in the future. Traditional approaches, for example, keys, ID cards, username and password, are neither satisfactory nor reliable enough in many security fields, biometrics authorizations based on face, iris, fingerprint have become a hot research filed. In those methods, iris recognition is regarded as a high accuracy verification technology, so that many countries have the same idea of adopting iris recognition to improve the safety of their key departments. The human iris can also be considered a valid biometrics for personal identification [Richard P W, 1996]. Biometrics recognition based on iris patterns is a hotspot as face recognition and fingerprint recognition recently years. The iris is the colored ring on the human eye between the pupil and the white sclera. Lots of physical biometric can be found in the colored ring of tissue that surrounds the pupil, such as corona, crypts, filaments, flecks, pits, radial furrows and striations. The iris features can be encoded by mathematical representation so that the patterns can be compared easily. In real-time iris recognition application system, iris localization is a very important step for iris recognition. The iris regions segmentation accuracy and localization real-time performance will affect the whole recognition system’s correct rate and effectiveness for large-scale database. Because iris region is a small object and has low grey value, it is very difficult to capture high contrast iris image clearly. In order to improve iris image contrast, usually some illuminations such as near infrared light source are used to increase intensity; however these illuminations may result in some faculas in iris image and affect iris segmentation and iris features. Here, we will discuss iris recognition system’s algorithm, all steps of iris recognition system will be introduced in details. Finally, we will show the experimental results based on iris database.

Despite numerous biometric challenges promoting iris and ocular recognition from less controlled images, iris localization and segmentation are still challenging topics. Examples include the Multiple Biometrics Grand Challenge (MBGC), Face and Ocular Challenge Series (FOCS), Noisy Iris Challenge Evaluation (NICE) and Iris Challenge Evaluation (ICE). For controlled environments and as long as iris segmentation algorithms can be tuned for the employed set, the segmentation problem can be managed. However, independent studies have shown, that algorithms may fail considerably if preconditions are not met. For datasets like FOCS segmentation performance ranging from 51% accuracy (using an integro-differential operator), to 90% (for active contours) is reported. Violations of preconditions or assumptions of controlled conditions such as particular hardware challenge the generalisation of successful segmentation techniques. Especially non-circularity of boundaries, non-frontal acquisition, or blurred images can impose difficulties. On the other hand, very flexible parameterless solutions allowing high degrees of freedom (e.g. off-axis acquisition) are sensitive to noise or reflections. They sometimes deliver inferior results to classical techniques for images where certain preconditions hold. Examples of challenging iris images include weak boundaries or reflections in visible-range iris images, making it harder for iris local- ization and segmentation algorithms to properly segment acquired images. Further undesirable conditions are on-the-move motion blur, out-of-focus images, narrowed eyes to a slit, or images with weak contrast. Besides accuracy, also speed and usability play important factors for efficient iris segmentation. For quick rejection of frames in video streams containing no iris image or unlikely leading to a successful verification, quality measures as predictors of recognition or segmentation accuracy can be used. This is useful to save precious processing time, as processing errors at early stages usually also result in classification errors. This chapter reviews recent approaches of iris segmentation techniques and discusses approaches towards robust segmentation, highlighting the following topics: • Segmentation performance assessment: Performance assessments are usually done with regards to recognition accuracy (thus evaluating an entire processing chain rather than individual algorithms). However, systematic seg- mentation errors can have a minor impact on performance if the same algorithm is used to segment gallery and probe images. This scenario is not realistic in cross-sensor applications. However, ground-truth only evaluations are restrictive, as they do not account for the tolerance of small segmentation in- accuracies by the employed feature extraction and comparison algorithm (e.g. via pooling operations). Different forms of segmentation accuracy evaluation are presented. There are separate segmentation-only challenges (e.g. NICE) and also ground-truth sets available, to which this chapter con- tributes a manual segmentation of the multispectral iris database UTIRIS. • Near infrared (NIR) vs. visible range (VIS) iris segmentation and the tuning problem: Algorithms developed for specific image datasets or character- istics are not agnostic towards the type of imagery used. NIR images, for example, exhibit a pronounced inner pupillary boundary, while VIS images tend to have better outer limbic contrast. This is usually exploited by seg- mentation algorithms to achieve better accuracy on particular datasets. While more generic techniques offering the same processing chain for VIS and NIR images have been suggested, tuning can have a strong impact on perfor- mance. This chapter presents modifications to the open source USIT - Univer- sity of Salzburg Iris Toolkit for better performance, if prior knowledge on iris data characteristics (NIR vs. VIS, iris-to-pupil ratio, size of iris, etc.) is available. Source code of the modifications is released. • Iris fusion at segmentation-level: Instead of developing completely new seg- mentation algorithms, the idea behind segmentation-level fusion is to make efficient use of different existing optimised approaches estimating their ac- curacy. The best individual segmentation technique can be used as the final segmentation result. Techniques can even be combined at a lower level leading to new segmentation results. For example, over- and under-segmentation methods can be fused to balance errors towards more robust iris segmentation. A learning-based approach predicting ground-truth segmentation performance is compared to non-quality based combinations in a multispectral setup assessing the impact of NIR and VIS spectras on performance. New techniques are analysed with regards to a challenging multispectral NIR and VIS iris database involving the same subjects. This allows a direct comparison of robustness of segmentation results (localisation of pupillary and limbic boundaries) across the two spectra and illustrates the high versatility of the proposed methods.

In Deep Learning, recent works show that neural networks have a high potential in the field of biometric security. The advantage of using this type of architecture, in addition to being robust, is that the network learns the characteristic vectors by creating intelligent filters in an automatic way, grace to the layers of convolution. In this paper, we propose an algorithm “FMnet” for iris recognition by using Fully Convolutional Network (FCN) and Multi-scale Convolutional Neural Network (MCNN). By taking into considerations the property of Convolutional Neural Networks to learn and work at different resolutions, our proposed iris recognition method overcomes the existing issues in the classical methods which only use handcrafted features extraction, by performing features extraction and classification together. Our proposed algorithm shows better classification results as compared to the other state-of-the-art iris recognition approaches.

Iris Recognition for Personal Identification System

As a reliable approach to human identification, iris recognition has received increasing attention in recent years. The most distinguishing feature of an iris image comes from the fine spatial changes of the image structure. So iris pattern representation must characterize the local intensity variations in iris signals. However, the measurements from minutiae are easily affected by noise, such as occlusions by eyelids and eyelashes, iris localization error, nonlinear iris deformations, etc. This greatly limits the accuracy of iris recognition systems. In this paper, an elastic iris blob matching algorithm is proposed to overcome the limitations of local feature based classifiers (LFC). In addition, in order to recognize various iris images efficiently a novel cascading scheme is proposed to combine the LFC and an iris blob matcher. When the LFC is uncertain of its decision, poor quality iris images are usually involved in intra-class comparison. Then the iris blob matcher is resorted to determine the input iris' identity because it is capable of recognizing noisy images. Extensive experimental results demonstrate that the cascaded classifiers significantly improve the system's accuracy with negligible extra computational cost.