Eyelid Occlusion Research Articles

Two-Stage and Two-Discriminator generative adversarial network for the inpainting of irregularly incomplete iris images

Due to the influence of the light source environment during image acquisition or the subject not fully opening their eyes, there are phenomena such as light spot interference, eyelash or eyelid occlusion in the iris area. This will cause the loss of effective iris information, ultimately affecting the success rate of recognition. To address the aforementioned issues, this paper introduces a Two-stage and Two-discriminator Generative Adversarial Network (TT-GAN) to restore irregularly incomplete iris images. The specific process is as follows: Firstly, TT-GAN employs gated convolution in place of ordinary convolution to more effectively extract image information; Secondly, TT-GAN incorporates a PC-Attention attention mechanism within the generator network to enhance the network's restoration capabilities; lastly, to ensure consistency in global and local semantic information of the repaired images, dual SN-PatchGAN discriminators are utilized. The experimental results based on three different spectral iris datasets, IITD, CASIA-Iris-Interval, and UBIRIS.V2, show that the repaired iris images significantly improve objective evaluation indicators compared to the incomplete iris images. Among them, TAR increased by 93.42 %, 68.41 %, and 59.72 %, respectively; EER increased by 27.17 %, 30.18 %, and 23.78 %, respectively; PSNR increased by 17.8712 dB, 18.818 dB, and 24.6053 dB respectively; SSIM increased by 0.0325, 0.0529, and 0.1337 respectively.

The Human Eye Pupil Detection System Using BAT Optimized Deep Learning Architecture

The pupil recognition method is helpful in many real-time systems, including ophthalmology testing devices, wheelchair assistance, and so on. The pupil detection system is a very difficult process in a wide range of datasets due to problems caused by varying pupil size, occlusion of eyelids, and eyelashes. Deep Convolutional Neural Networks (DCNN) are being used in pupil recognition systems and have shown promising results in terms of accuracy. To improve accuracy and cope with larger datasets, this research work proposes BOC (BAT Optimized CNN)-IrisNet, which consists of optimizing input weights and hidden layers of DCNN using the evolutionary BAT algorithm to efficiently find the human eye pupil region. The proposed method is based on very deep architecture and many tricks from recently developed popular CNNs. Experiment results show that the BOC-IrisNet proposal can efficiently model iris microstructures and provides a stable discriminating iris representation that is lightweight, easy to implement, and of cutting-edge accuracy. Finally, the region-based black box method for determining pupil center coordinates was introduced. The proposed architecture was tested using various IRIS databases, including the CASIA (Chinese academy of the scientific research institute of automation) Iris V4 dataset, which has 99.5% sensitivity and 99.75% accuracy, and the IIT (Indian Institute of Technology) Delhi dataset, which has 99.35% specificity and MMU (Multimedia University) 99.45% accuracy, which is higher than the existing architectures.

Nystagmus patterns classification framework based on deep learning and optical flow

Benign paroxysmal positional vertigo (BPPV) is the most common vestibular peripheral vertigo disease characterized by brief recurrent vertigo with positional nystagmus. Clinically, it is common to recognize the patterns of nystagmus by analyzing infrared nystagmus videos of patients. However, the existing approaches cannot effectively recognize different patterns of nystagmus, especially the torsional nystagmus. To improve the performance of recognizing different nystagmus patterns, this paper contributes an automatic recognizing method of BPPV nystagmus patterns based on deep learning and optical flow to assist doctors in analyzing the types of BPPV. Firstly, we present an adaptive method for eliminating invalid frames that caused by eyelid occlusion or blinking in nystagmus videos and an adaptive method for segmenting the iris and pupil area from video frames quickly and efficiently. Then, we use a deep learning-based optical flow method to extract nystagmus information. Finally, we propose a nystagmus video classification network (NVCN) to categorize the patterns of nystagmus. We use ConvNeXt to extract eye movement features and then use LSTM to extract temporal features. Experiments conducted on the clinically collected datasets of infrared nystagmus videos show that the NVCN model achieves an accuracy of 94.91% and an F1 score of 93.70% on nystagmus patterns classification task as well as an accuracy of 97.75% and an F1 score of 97.48% on torsional nystagmus recognition task. The experimental results prove that the framework we propose can effectively recognize different patterns of nystagmus.

Filterable sample consensus based on angle variance for pupil segmentation

Pupil segmentation is a first and important topic of iris recognition, identity recognition and eye movement information extraction for mental analysis. However, due to the negative effects of eyelash occlusion, eyelid occlusion and off-gaze deflection, making a precise pupil segmentation is a difficult task. Therefore, we propose a precise and robust algorithm for pupil segmentation, namely Angle Variance based Filterable Sample Consensus (AVBFSC), which is composed of an outlier filter and a boundary locator. The outlier filter can eliminate negative effects mentioned above, and also a best pupil segmentation is performed with our boundary locator, which learns a circular mathematical equation by selecting sub-samples from pupil edge pixels randomly. Experiment results in comparison with state-of-the-art methods on CASIA-Iris-V4-Interval dataset, indicate that our algorithm achieved best performance, that is, Accuracy of 98.99%, False Acceptance Rate (FAR) of 2.09% and Genuine Acceptance Rate (GAR) of 98.54%. In addition, it also has robust results under the condition of specific non-ideal scenes from CASIA-Iris-V4-Lamp dataset and Indian Institute of Technology Delhi (IITD) dataset including dark light conditions, eyelash occlusion, eyelid occlusion and off-gaze deflection.

Texture synthesis and modified filter bank in contourlets for improved iris recognition

Iris recognition in the presence of eyelash occlusions is a challenging task over the years since it has begun. The active area captured under non-ideal imaging conditions usually suffers from low contrast, poor brightness, blur due to camera or subject’s relative motion and particularly eyelash and eyelid occlusions. Accurate segmentation methods avoid occlusions to some extent but not completely always. In the proposed method, pixel-wise texture synthesis is done on occluded regions which improves the correct recognition rate (CRR). The contourlet transform which is a multiresolution tool, decomposes an image into different scales and directions with the help of pyramidal and directional filter bank (PDFB). A new FIR filter named as SSK filter is proposed by the authors for the PDFB in contourlets to extract apt features of iris such that the CRR is further improved. The performance of the proposed method is checked against CASIA-Iris-Interval (V3), IITD, CASIA-V1 and UBIRIS-V1 iris databases, and from the results obtained, it is proved that the proposed method is very much worthwhile for improved iris recognition even in the presence of eyelash and eyelid occlusions.

Extracting Unique Personal Identification Number from Iris

The aim of this research is to propose an enhanced model that significantly reduces the noise in the iris templates and thus present an efficient and reliable iris recognition system. Accuracy of Iris recognition system significantly depends on reducing noise at every stage of recognition process especially at eye image acquisition and iris segmentation stages. The proposed Iris recognition model combines Canny edge detector and circular Hough transform which yields more rapid and accurate localization of Iris inner and outer boundaries. The eyelid and eyelashes occlusion is also dealt with by applying non maxima radial suppression technique. The segmented Iris area is then normalized and converted from annular shape to fixed-sized rectangular block. In this stage all iris templates converted from circular to rectangular shape will have standard dimensions thus enabling Iris recognition system to perform comparison between them. Fast Fourier Transform function is then applied to normalized Iris image. Input image is viewed as a signal that is converted from time to frequency domain in which the number of frequencies is equal to number of image pixels. The resulting matrix is consisted of real and imaginary part represented by complex numbers. The matrix is encoded according to the pre established encoding pattern that assigns certain sequence of binary numbers to each possible combination of complex numbers thus producing Iris code. In the matching phase the iris codes are compared and if the determined degree of dissimilarity between the two is below the pre-established threshold value than the codes belong to the same user. The proposed Iris recognition method, therefore, successfully treats eyelids and eyelashes noise issues and present the efficient set of algorithms whose performance is superior compared to other existing methods.

IRIS RECOGNITION BASED ON THE MODIFIED CHAN-VESE ACTIVE CONTOUR

Over recent years, iris recognition has been an explosive growth of interest in human identification due to its high accuracy. Iris recognition is a biometric system that uses iris to verify and identify human identity. Iris has pattern that is rich with textures and can be compared among humans. There are many methods can be used in iris recognition. The methods based on the integro-differential operator and Hough transform are the most widely used in iris recognition. Unfortunately, both methods require more time to execute and has less accurate recognition accuracy due to the eyelid occlusion. In order to solve these problems, the Chan-Vese active contour is modified to reduce the execution time and to increase the recognition accuracy of iris recognition. Then, this method is compared with the integro-differential operator method. The iris images from CASIA-v4 database are used for the experiments. According to the results, the proposed method recorded 0.91 s for execution time which was 61.28 % faster than the integro-differential operator method. The proposed method also achieved 0.9831 for area under curve (AUC) which was 2.66 % higher recognition accuracy than the integro-differential operator method. To conclude, the modified Chan-Vese active contour was able to improve the performance of iris recognition compared to the integro-differential operator method.

Gaze Estimation From Color Image Based on the Eye Model With Known Head Pose

This paper proposes a novel method of gaze estimation based on an eye model with known head pose. The most crucial factors in the eye-model-based approach to gaze estimation are the 3-D positions of the eyeball and iris centers. In the proposed method, an RGB-D camera, Kinect sensor, is used to obtain the head pose as well as the eye region of the color image. The 3-D position of the eyeball center is determined in the calibration phase by gazing at the center of the color image camera. Then, to estimate the 3-D position of the iris center, the 3-D contour of the iris is projected onto the color image with the known head pose obtained from color and depth cues of an RGB-D camera. Thus, the ellipse of the iris in the image can be described using only two parameters: the yaw and pitch angles of the eyeball in the iris coordinate system, rather than the conventional five parameters of an ellipse. The proposed method can fit an iris that is not complete due to eyelid occlusion. The average errors of vertical and horizontal angles of the gaze estimation for seven subjects are 5.9° and 4.4°, respectively. The processing speed is as high as 330 ms per frame. However, for lower resolution and poor illumination images, as tested on the public database EYEDIAP, the performance of the proposed eye-model-based method is inferior to that of the-state-of-the-art appearance-based method.

Iris Image Enhancement for the Recognition of Non-ideal Iris Images

Iris recognition for biometric personnel identification has gained much interest owing to the increasing concern with security today. The image quality plays a major role in the performance of iris recognition systems. When capturing an iris image under uncontrolled conditions and dealing with non-cooperative people, the chance of getting non-ideal images is very high owing to poor focus, off-angle, noise, motion blur, occlusion of eyelashes and eyelids, and wearing glasses. In order to improve the accuracy of iris recognition while dealing with non-ideal iris images, we propose a novel algorithm that improves the quality of degraded iris images. First, the iris image is localized properly to obtain accurate iris boundary detection, and then the iris image is normalized to obtain a fixed size. Second, the valid region (iris region) is extracted from the segmented iris image to obtain only the iris region. Third, to get a well-distributed texture image, bilinear interpolation is used on the segmented valid iris gray image. Using contrast-limited adaptive histogram equalization (CLAHE) enhances the low contrast of the resulting interpolated image. The results of CLAHE are further improved by stretching the maximum and minimum values to 0–255 by using histogram-stretching technique. The gray texture information is extracted by 1D Gabor filters while the Hamming distance technique is chosen as a metric for recognition. The NICE-II training dataset taken from UBRIS.v2 was used for the experiment. Results of the proposed method outperformed other methods in terms of equal error rate (EER).

An Effective Iris Recognition System

An iris recognition system uses the iris to distinguish the identity of a person using the rich iris texture feature. To effectively remove noise and precisely segment the stable iris region is a crucial stage prior to recognition. Most noises on iris images are caused by occlusion of eyelids or eyelashes in certain areas. In this paper, we propose an iris recognition system which precisely locates and segments iris regions. We extract the iris feature from a relatively reliable portion of the iris region using a DoG filter. Experimental results show that the proposed iris recognition system has satisfactory results in terms of time efficiency and recognition rate. DOI : http://dx.doi.org/10.11591/telkomnika.v12i5.4951

Eyelid Detection Method Based on a Fuzzy Multi-Objective Optimization

Iris recognition is one of the most robust human identification methods. In order to carry out accurate iris recognition, many factors of image quality should be born in mind. The eyelid occlusion is a quality factor that may significantly affect the accuracy. In this paper we introduce a new fuzzy multi-objective optimization approach based on the eyelid detection method. This method obtains the eyelid contour which represents the best solution of Pareto-optimal set taking into account five optimized objectives. This proposal is composed of three main stages, namely, gathering eyelid contour information, filtering eyelid contour and tracing eyelid contour. The results of the proposal are evaluated in a verification mode and thus a few performance measures are generated in order to compare them with other works of the state of the art. Thereby, the proposed method outperforms other approaches and it is very useful for implementing real applications as well.

Efficient and Robust Pupil Size and Blink Estimation From Near-Field Video Sequences for Human–Machine Interaction

Monitoring pupil and blink dynamics has applications in cognitive load measurement during human-machine interaction. However, accurate, efficient, and robust pupil size and blink estimation pose significant challenges to the efficacy of real-time applications due to the variability of eye images, hence to date, require manual intervention for fine tuning of parameters. In this paper, a novel self-tuning threshold method, which is applicable to any infrared-illuminated eye images without a tuning parameter, is proposed for segmenting the pupil from the background images recorded by a low cost webcam placed near the eye. A convex hull and a dual-ellipse fitting method are also proposed to select pupil boundary points and to detect the eyelid occlusion state. Experimental results on a realistic video dataset show that the measurement accuracy using the proposed methods is higher than that of widely used manually tuned parameter methods or fixed parameter methods. Importantly, it demonstrates convenience and robustness for an accurate and fast estimate of eye activity in the presence of variations due to different users, task types, load, and environments. Cognitive load measurement in human-machine interaction can benefit from this computationally efficient implementation without requiring a threshold calibration beforehand. Thus, one can envisage a mini IR camera embedded in a lightweight glasses frame, like Google Glass, for convenient applications of real-time adaptive aiding and task management in the future.

Iris Boundary Localization Based on Improve Hough Transform

ris location is one of the key steps of iris recognition system. Non-ideal iris image has some problems, such as eyelid and eyelash occlusion, low contrast of iris and sclera, uneven illumination, and so on. Because of that, its difficult to identify the boundary, especially the exterior boundary. Therefore, this paper proposes a method based on the improved Hough Transform. First, use the minimum method to find the datum point in the pupil, after that identify the valid area of the interior boundary base on that point. Apply the improved Hough Transform to that valid area to identify the interior boundary of the iris image. Then regard the center of the interior circle as our new datum point, use the same method to identify the exterior boundary. Experiment results show that our algorithm has higher accuracy than traditional method on the non-ideal iris image segmentation.

A new approach to the design of hybrid finer directional wavelet filter bank for iris feature extraction and classification using k-out-of-n:A post-classifier

This paper presents a novel iris feature extraction scheme using two-dimensional non-separable, non-redundant, multi-scale hybrid finer directional wavelet filter bank and classification using fused post-classifier under non-ideal environmental conditions. The proposed approach overcomes limited directionality of wavelet transform. The designed filter bank converts the wavelet basis functions to a set of directional basis elements by employing the combination of designed bi-orthogonal wavelet filter bank and modified checkerboard-shaped filter bank (designed using triplet of half-band filters). The false rejection rate is reduced with the help of k-out-of-n:A rule. The proposed technique is capable to handle possible artifacts especially inaccurate iris localization, occlusion of eyelids, pupil, eyelashes, reflection on iris, non-linear deformation, head-tilt, etc. The performance of the proposed scheme is validated using CASIA-Iris V2.0, MMU1, and UBIRIS databases. Finally, the performance of proposed approach is compared with other existing iris recognition algorithms to demonstrate its potential.

Amblyopia in Childhood Eyelid Ptosis

To report the prevalence and causes of amblyopia among children with ptosis diagnosed in a well-defined population over a 40-year period. Retrospective, population-based cohort study. We retrospectively reviewed the charts of 107 patients younger than 19 years for the prevalence and causes of amblyopia who were diagnosed with childhood ptosis and were residents of Olmsted County, Minnesota, from January 1, 1965, through December 31, 2004. Amblyopia was diagnosed in 16 (14.9%) of the 107 patients with childhood ptosis. Fourteen (14.6%) of 96 patients diagnosed with a congenital form of ptosis demonstrated amblyopia. Twelve (14.8%) of the 81 patients diagnosed with simple congenital ptosis had amblyopia, 7 (8.6%) cases of which solely were the result of eyelid occlusion of the visual axis. The causes of amblyopia in the remaining 5 patients were significant refractive error in 3 patients and strabismus in 2 patients. Amblyopia occurred in 1 in 7 children diagnosed with ptosis in this population-based cohort. In approximately half of those with amblyopia, or less than 10% of all patients, the disease solely was the result of eyelid occlusion of the visual axis.

Eyelid Noise Detection Based on Gray-Scale Morphology and Judgement of Noise Occlusion

In iris recognition, eyelid occlusions will reduce the accuracy of iris recognition, an algorithm of eyelid occlusions detection based on Gray-scale Morphology and the judgement of the noise occlusion is proposed in this paper. By judging the projection relation between eyelid and eyelashes, the algorithm in this paper could determine which part need to be detected, instead of detecting all of the eyelid and eyelashes occlusion blindly. So the efficiency of the algorithm is largely improved and in addition, the noise detection method based on gray-scale morphology could perfect fit out the eyelid edge.

Occurrence of eye movement disorders in motor neuron disease

The diagnosis of amyotrophic lateral sclerosis (ALS) relies on symptoms and signs related to upper and lower motor neuron injury. Preservation of normal ocular motor movements is an important criterion for making this diagnosis as oculomotility pathways are classically spared in ALS. However, some authors report eye disturbances resulting from nuclear and supranuclear ophthalmoplegia in autopsy-proven ALS. Here, we report a case in which eye movement disorders were an early sign associated with a bulbar-onset ALS. The association of progressive ophthalmoplegia, dysexecutive syndrome and automatico-voluntary dissociation of eyelid occlusion suggested a 'progressive supranuclear palsy variant' of ALS caused by a disturbance in the descending frontal projections, even though morphological imaging was normal. Motor neuron disease with eye movement disorders must not be considered as a distinct clinical entity and must not exclude a diagnosis of ALS.

A Robust Iris Identification System Based on Wavelet Packet Decomposition and Local Comparisons of the Extracted Signatures

This paper presents a complete iris identification system including three main stages: iris segmentation, signature extraction, and signature comparison. An accurate and robust pupil and iris segmentation process, taking into account eyelid occlusions, is first detailed and evaluated. Then, an original wavelet-packet-based signature extraction method and a novel identification approach, based on the fusion of local distance measures, are proposed. Performance measurements validating the proposed iris signature and demonstrating the benefit of our local-based signature comparison are provided. Moreover, an exhaustive evaluation of robustness, with regards to the acquisition conditions, attests the high performances and the reliability of our system. Tests have been conducted on two different databases, the well-known CASIA database (V3) and our ISEP database. Finally, a comparison of the performances of our system with the published ones is given and discussed.

A novel approach to image quality assessment in iris recognition systems

With increasing needs in security systems, iris recognition is an important technique as one of the most reliable solutions for biometrics-based identification systems. However, not all of the iris images acquired from the device are in-focus and sharp enough for recognition. Thus, the poor quality of iris images has serious influence on the accuracy of iris recognition. Sometimes these images are not good enough due to a variety of factors: defocus blur, motion blur, eyelid occlusion and eyelash occlusion. This paper presents an approach for quality assessment of iris images, which can select the high quality iris images from the image sequences to be used in iris recognition systems. First, the gradient information of the iris regions (64 × 64) adjoining the pupil on the right and left sides is calculated to distinguish the blurred images from the in-focus images. Next, the valid iris regions are employed to discriminate between the occluded images and useful images. We present underlying theory as well as experimental results from both the CASIA iris database and the database provided for the iris challenge evaluation (ICE). The results show that this evaluation approach can actually reflect the real quality of iris images and significantly improve the overall performance of the iris recognition systems.

Fast and accurate personal identification based on iris biometric

In this paper, we propose an efficient and accurate technique to process iris biometric. Our approach handles degraded quality of eye images, occlusion of eyelids and eyelashes. Our approach considers non-circular pupil and iris boundary. With our approach it is possible to store an iris template with 304 bits only unlike 1024 bits used in majority of the existing approaches. Further, we assign different weights to different iris regions enabling us to compare two iris templates with a lesser number of features and thus achieving better accuracy. Experimental results on several iris databases establish the efficacy of our approach.