Iris Images Research Articles

Two image quality assessment methods based on evidential modeling and uncertainty: application to automatic iris identification systems

The performance of an Automatic iris Identification System is impacted by both the poor quality of iris images and the uncertainty of information. Assessing image quality and rejecting poor-quality images can substantially improve the performances of the current biometric systems. The main idea behind our proposed Image Quality Assessment approaches is to take advantage, firstly, of the texture of iris images and, secondly, of the uncertainty of these information. This is achieved by defining a set of Contextual Quality Indicators extracted from the image texture and transforming them into Quality Assessment Criteria in the evidential framework, taking into account the information uncertainty degree. The Contextual Quality Indicators are defined based on a priori analysis of the context of the application. We use ‘iris’ as the context of application. Generally, only the normalized iris image is saved, i.e. the acquired iris image is not always available. So, the main advantage of our approaches over other related methods is that it can act in the normalization level of the processing chain to reject poor-quality images. So that, the subsequent Automatic iris Identification System can process only good-quality images, which result in better recognition rate performance. The functioning of our evidential approaches is illustrated using image samples from CASIA 1.0 database. The performance of over the proposed image quality assessment approaches is compared with the standard iris identification system without an image quality assessment step. A statistical test, based on 95% confidence interval, is used to assess if there is a statistically significant difference between the performances of the proposed approaches. The CASIA 1.0 has been used to make the comparison. The comparison results highlight the effectiveness of the proposed approaches for iris domain of applications. The source code of our paper is available at https://github.com/Sonda09/IIQA

Cycle Generative Adversarial Network Based on Gradient Normalization for Infrared Image Generation

Image generation technology is currently one of the popular directions in computer vision research, especially regarding infrared imaging, bearing critical applications in the military field. Existing algorithms for generating infrared images from visible images are usually weak in perceiving the salient regions of images and cannot effectively highlight the ability to generate texture details in infrared images, resulting in less texture details and poorer generated image quality. In this study, a cycle generative adversarial network method based on gradient normalization was proposed to address the current problems of poor infrared image generation, lack of texture detail and unstable models. First, to address the problem of limited feature extraction capability of the UNet generator network that makes the generated IR images blurred and of low quality, the use of the residual network with better feature extraction capability in the generator was employed to make the generated infrared images highly defined. Secondly, in order to solve issues concerning severe lack of detailed information in the generated infrared images, channel attention and spatial attention mechanisms were introduced into the ResNet with the attention mechanism used to weight the generated infrared image features in order to enhance feature perception of the prominent regions of the image, helping to generate image details. Finally, to tackle the problem where the current training models of adversarial generator networks are insufficiently stable, which leads to easy collapse of the model, a gradient normalization module was introduced in the discriminator network to stabilize the model and render it less prone to collapse during the training process. The experimental results on several datasets showed that the proposed method obtained satisfactory data in terms of objective evaluation metrics. Compared with the cycle generative adversarial network method, the proposed method in this work exhibited significant improvement in data validity on multiple datasets.

Possibilistic Reasoning based No-Reference Iris Image Quality Assessment

Low quality of data is one of the most critical issues for consumers of data distributed by autonomous sources. The outcomes of non-quality of data, or of poor quality, on decision-making are considerable and have disastrous effects. In this paper, we deal with this problem in the context of biometric applications, specifically those based on the iris modality. In fact, the iris image can be affected by different types of distortion which greatly affect its quality and generate different imperfection forms. Thus, the deterioration of the quality of these images increases the false rejection rate and therefore decreases significantly the performances of iris identification systems. In order to reduce the impact of bad quality of iris image, we add an Image Quality Assessment (IQA) phase to the identification system to reject poor quality images and thus improves system performances. In this paper, a new method based on the possibilistic modeling of imperfect information extracted from the iris images is proposed to assess the iris image quality. The aim of the proposed approach, noted Poss-IIQA, is to handles the different imperfection forms resulting from any type of distortions. As a result, the suggested method key advantage over the baseline system and other earlier approaches, which focuses more on distortions than on imperfections.

Contextual Measures for Iris Recognition

The iris patterns of the human contain a large amount of randomly distributed and irregularly shaped microstructures. These microstructures make the human iris informative biometric traits. To learn identity representation from them, this paper regards each iris region as a potential microstructure and proposes contextual measures (CM) to model the correlations between them. CM adopts two parallel branches to learn global and local contexts in iris image. The first one is the globally contextual measure branch. It measures the global context involving the relationships between all regions for feature aggregation and is robust to local occlusions. Besides, we improve its spatial perception considering the positional randomness of the microstructures. The other one is the locally contextual measure branch. This branch considers the role of local details in the phenotypic distinctiveness of iris patterns and learns a series of relationship atoms to capture contextual information from a local perspective. In addition, we develop the perturbation bottleneck to make sure that the two branches learn divergent contexts. It introduces perturbation to limit the information flow from input images to identity features, forcing CM to learn discriminative contextual information for iris recognition. Experimental results suggest that global and local contexts are two different clues critical for accurate iris recognition. The superior performance on four benchmark iris datasets demonstrates the effectiveness of the proposed approach in within-database and cross-database scenarios.

ANALYSIS AND EVALUATION OF BIOMETRIC AUTHENTICATION MEANS BASED ON THE IMAGE OF THE FACE AND IRIS OF THE STAFF OF CRITICAL INFRASTRUCTURE FACILITIES

Dedicated to the analysis and evaluation of biometric authentication systems for personnel of critical infrastructure facilities. It is shown that tools based on the image of the face and the iris of the eye have broad prospects, which is explained by the proven solutions in the field of face image analysis, the availability and distribution of video recording tools that allow the iris to be recorded simultaneously with the face with satisfactory quality. It was determined that one of the ways to improve the effectiveness of such tools is to increase the accuracy of face recognition and resistance to obstacles that cover part of the face. It is also shown that an additional direction of improving the effectiveness of biometrics can be recognition of the current psycho-emotional state of personnel of critical infrastructure facilities. The need to evaluate the effectiveness of face and emotion recognition tools based on face and iris images is determined. Based on the analysis of the literature, two groups of efficiency criteria were formed: basic and additional. Criteria characterizing the effectiveness of the recognition process are included in the main group, and criteria corresponding to the technical implementation features and service capabilities of recognition tools are included in the additional group. An evaluation of modern means of face and emotion recognition based on the image of the face and iris of the eye was carried out, and their non-compliance with a number of criteria was determined. It is proposed to correlate the ways of further research with the solution of the task of ensuring the fulfillment of the criteria related to the possibility of recognizing emotions and a person based on part of the image of the face, facial expressions and part of the image of the iris of the eye, with the technical implementation of expert solutions. The possibility of performing the specified task due to the use of modern neural network technologies is shown.

An Optimised Defensive Technique to Recognize Adversarial Iris Images Using Curvelet Transform

Deep Learning is one of the most popular computer science techniques, with applications in natural language processing, image processing, pattern identification, and various other fields. Despite the success of these deep learning algorithms in multiple scenarios, such as spam detection, malware detection, object detection and tracking, face recognition, and automatic driving, these algorithms and their associated training data are rather vulnerable to numerous security threats. These threats ultimately result in significant performance degradation. Moreover, the supervised based learning models are affected by manipulated data known as adversarial examples, which are images with a particular level of noise that is invisible to humans. Adversarial inputs are introduced to purposefully confuse a neural network, restricting its use in sensitive application areas such as biometrics applications. In this paper, an optimized defending approach is proposed to recognize the adversarial iris examples efficiently. The Curvelet Transform Denoising method is used in this defense strategy, which examines every sub-band of the adversarial images and reproduces the image that has been changed by the attacker. The salient iris features are retrieved from the reconstructed iris image by using a pre-trained Convolutional Neural Network model (VGG 16) followed by Multiclass classification. The classification is performed by using Support Vector Machine (SVM) which uses Particle Swarm Optimization method (PSO-SVM). The proposed system is tested when classifying the adversarial iris images affected by various adversarial attacks such as FGSM, iGSM, and Deepfool methods. An experimental result on benchmark iris dataset, namely IITD, produces excellent outcomes with the highest accuracy of 95.8% on average.

A Novel Fusion System Based on Iris and Ear Biometrics for E-exams

With the rapid spread of the coronavirus epidemic all over the world, educational and other institutions are heading towards digitization. In the era of digitization, identifying educational e-platform users using ear and iris based multimodal biometric systems constitutes an urgent and interesting research topic to preserve enterprise security, particularly with wearing a face mask as a precaution against the new coronavirus epidemic. This study proposes a multimodal system based on ear and iris biometrics at the feature fusion level to identify students in electronic examinations (E-exams) during the COVID-19 pandemic. The proposed system comprises four steps. The first step is image preprocessing, which includes enhancing, segmenting, and extracting the regions of interest. The second step is feature extraction, where the Haralick texture and shape methods are used to extract the features of ear images, whereas Tamura texture and color histogram methods are used to extract the features of iris images. The third step is feature fusion, where the extracted features of the ear and iris images are combined into one sequential fused vector. The fourth step is the matching, which is executed using the City Block Distance (CTB) for student identification. The findings of the study indicate that the system’s recognition accuracy is 97%, with a 2% False Acceptance Rate (FAR), a 4% False Rejection Rate (FRR), a 94% Correct Recognition Rate (CRR), and a 96% Genuine Acceptance Rate (GAR). In addition, the proposed recognition system achieved higher accuracy than other related systems.

A Hybrid Classification Approach for Iris Recognition System for Security of Industrial Applications

The biometric authentication system is demanded to identify a particular person from the set of persons. Even though many biometric authentication methods are available such as fingerprint, palm, face, and iris, the iris-based recognition system is effective due to its simplified process. This article proposes an iris recognition system using a hybrid classification approach for security applications. The proposed method includes three modules: preprocessing, augmentation, and classifier. The preprocessing module converts the color iris images into grey scale images and also resizes the image into 256 × 256. The preprocessed iris images are now data augmented to construct the larger dataset. The data augmented images are classified into either genuine or imposter images using a hybrid classification approach. The hybrid classification approach functions in two modes as training and testing. In this article, the Convolutional Neural Networks (CNN) is integrated with the Adaptive Neuro-Fuzzy Inference System (ANFIS) classifier to enhance the recognition rate of the iris recognition system. The performance analysis of the proposed approach is shown in terms of sensitivity, accuracy, recognition rate, specificity, false-positive rate, and false-negative rate. The experimental results of the proposed iris recognition system stated in this article significantly outweigh other design methods.

Infrared nanospectroscopy depth-dependent study of modern materials: morpho-chemical analysis of polyurethane/fibroin binary meshes.

Infrared scattering-type scanning near-field optical microscopy (IR s-SNOM) and imaging is here exploited together with attenuated total reflection (ATR) IR imaging and scanning electron microscopy (SEM) to depict the chemical composition of fibers in hybrid electrospun meshes. The focus is on a recently developed bio-hybrid material for vascular tissue engineering applications, named Silkothane®, obtained in the form of nanofibrous matrices from the processing of a silk fibroin-polyurethane (SFPU) blend via electrospinning. Morphology and chemistry of single fibers, at both surface and subsurface level, have been successfully characterized with nanoscale resolution, taking advantage of the IR s-SNOM capability to portray the nanoscale depth profile of this modern material working at diverse harmonics of the signal. The applied methodology allowed to describe the superficial characteristics of the mesh up to a depth of about 100 nm, showing that SF and PU do not tend to co-aggregate to form hybrid fibers, at least at the length scale of hundreds of nanometers, and that subdomains other than the fibrillar ones can be present. More generally, in the present contribution, the depth profiling capabilities of IR s-SNOM, so far theoretically predicted and experimentally proven only on model systems, have been corroborated on a real material in its natural conditions with respect to production, opening the room for the exploitation of IR s-SNOM as valuable technique to support the production and the engineering of nanostructured materials by the precise understanding of their chemistry at the interface with the environment.

Estimation of Sea Level Difference: Velocities of Sea Surface Markers Derived from IR Images in Comparison with Satellite Altimetry Data

Estimation of Sea Level Difference: Velocities of Sea Surface Markers Derived from IR Images in Comparison with Satellite Altimetry Data

Chalcogenide Glass IR Artificial Compound Eyes Based on Femtosecond Laser Microfabrication (Adv. Mater. Technol. 2/2023)

Artificial Compound Eyes In article number 2200741, Feng Chen and co-workers present chalcogenide glass IR artificial compound eyes inspired by the fly in the nature and fabricated by femtosecond laser could capture the IR image with the advantages of large field of view and low aberration and be used in complex environments, exhibiting broad prospects in IR imaging, robot vision, IR 3D motion tracking and so on.

Investigation of Depth Camera Potentials for Variable-Rate Sprayers

Highlights A commercial depth camera with a custom-designed graphical user interface was evaluated to detect tree canopy. Measurement variations under different indoor conditions were negligible for practical applications. Measurement errors ranged from 2.8% to 15.8%, which were acceptable for outdoor applications. Variation of crabapple canopy detection rate was less than 6% from sunrise to sunset. Abstract. To reduce crop protection product use and environmental impacts while maintaining application efficacy and convenience for applicators, an automatic variable rate sprayer coupled with a canopy detection sensor is required. A commercial depth camera was tested as a means of detecting the canopy of ornamental and tree crops for the sprayer. A custom-designed graphical user interface was developed to control the depth camera and save RGB and IR images and depth data to a local computer. Indoor evaluations showed that measurements could be influenced by the temperature and illumination; however, the influence was minimal, with a relative error of less than 1% and a maximum difference of 14 mm between the average measurements. The depth camera was able to detect a 31% to 72% area of a 20-mm wide target, and the rates went up 72% to 89% when the target width increased to 40 mm. The depth camera showed acceptable performance in detecting canopy contour changes and had measurement errors of 2.8% to 15.3% while detecting the distances to outdoor crabapple and oak trees. In addition, the depth camera detected tree canopy in various outdoor conditions from sunrise to sunset with reasonable accuracy (less than 10% of relative errors). In terms of measurement stability, the depth camera detected crabapple canopy with less than 6% variations under various illuminations between sunrise and sunset. The results suggested that the performance of the depth camera was adequate for detecting canopy under outdoor conditions for future variable-rate spray applications in ornamental and tree crop production. In addition, the study outlined the performance of the depth camera, which provided a guideline for future applications. Keywords: Machine Vision, Precision Agriculture, Specialty Crop, Stereo Vision, Variable Rate Application.

Machine learning-based iris liveness identification using fragmental energy of cosine transformed iris images

Iris biometric identification provides a contactless authentication preventing the spread of COVID-19 like diseases. These systems are made vulnerable and unsafe because of the spoofing attacks attempted with the help of contact lenses, video replays and print attacks. The paper proposes the iris liveness detection method to mitigate spoofing attacks, taking fragmental coefficients of cosine transformed iris image to be used as features. Seven variants of feature formation are considered in experimental validations of the proposed method, and the features are used to train eight assorted machine learning classifiers and ensembles for iris liveness identification. Recall, F-measure, precision and accuracy are used to evaluate performances of the projected iris liveness identification variants. The experimentation carried out on four standard datasets have shown better iris liveness identification by the fragmental coefficients of cosine transformed iris image with size 4 * 4 using random forest algorithm having 99.18% accuracy immediately followed by an ensemble of classifiers.

Deep Learning-Based Iris Segmentation Algorithm for Effective Iris Recognition System

In this study, a 19-layer convolutional neural network model is developed for accurate iris segmentation and is trained and validated using five publicly available iris image datasets. An integrodifferential operator is used to create labeled images for CASIA v1.0, CASIA v2.0, and PolyU Iris image datasets. The performance of the proposed model is evaluated based on accuracy, sensitivity, selectivity, precision, and F-score. The accuracy obtained for CASIA v1.0, CASIA v2.0, CASIA Iris Interval, IITD, and PolyU Iris are 0.82, 0.97, 0.9923, 0.9942, and 0.98, respectively. The result shows that the proposed model can accurately predict iris and non-iris regions and thus can be an effective tool for iris segmentation.

Iris Liveness Detection Using Fragmental Energy of Haar Transformed Iris Images Using Ensemble of Machine Learning Classifiers

Contactless verification is possible with iris biometric identification, which helps prevent infections like COVID-19 from spreading. Biometric systems have grown unsteady and dangerous as a result of spoofing assaults employing contact lenses, replayed the video, and print attacks. The work demonstrates an iris liveness detection approach by utilizing fragmental coefficients of Haar transformed Iris images as signatures to prevent spoofing attacks for the very first time in the identification of iris liveness. Seven assorted feature creation ways are studied in the presented solutions, and these created features are explored for the training of eight distinct machine learning classifiers and ensembles. The predicted iris liveness identification variants are evaluated using recall, F-measure, precision, accuracy, APCER, BPCER, and ACER. Three standard datasets were used in the investigation. The main contribution of our study is achieving a good accuracy of 99.18% with a smaller feature vector. The fragmental coefficients of Haar transformed iris image of size 8 * 8 utilizing random forest algorithm showed superior iris liveness detection with reduced featured vector size (64 features). Random forest gave 99.18% accuracy. Additionally, conduct an extensive experiment on cross datasets for detailed analysis. The results of our experiments show that the iris biometric template is decreased in size to make the proposed framework suitable for algorithmic verification in real-time environments and settings.

Design and development of fusion-based expert system for multimodal biometric recognition with facemask authentication

A multimodal biometric system can be processed on more than one biometric feature and make the authentication process more impactful in comparison with the traditional identification system. Here in this paper, with the help of discrete cosine transformation (DCT), Hough transformation technique, and discrete wavelet transformation (DWT) we have designed a decision-level fusion-based authentication. In our proposed authentication model, the system will take the face, ear, and iris images individually and doing DCT, DWT, and Hough transformation, finally apply decision level fusion for the biometric features extraction. Our proposed system has more authentication capability due to iris authentication features, which cannot be easily tampered with. This experimental process clearly shows that the proposed multimodal systems have more efficiency, reliability, and are more robust than other existing biometric multimodal systems and are able to authenticate with or without wearing a surgical face mask, useful for hazardous and contaminated environments.

A SWIN TRANSFORMER IN THE CLASSIFICATION OF IR IMAGES OF PHOTOVOLTAIC MODULE

The faults occurring in the photo voltaic module as to be detected in order to increase its efficiency. The infrared images, electroluminescent images, and photo luminescent images of the photo voltaic modules have been used to detect and classify the faults. The infrared data set is used for classification and it is highly imbalanced. To make it balanced, generative adversarial networks are used to generate images for each fault category. If the fault classification is done using a convolution neural network, the feature maps are generated by convolution operation with filters on images. The convolution neural network model is to be trained for 1000 epochs and the time required for training is greater than 24 hours. The computational cost of a convolution neural network (CNN) is reduced in transformers through the attention mechanism. The Swin (shifted window) transformers which are used for classification as to be trained for 40 epochs and the maximum training time is less than 6 hours. The computation time, categorical classification accuracy, and top-5-accuracy obtained using Swin classifier are compared with the existing methods. It is found that the computational cost is very much reduced by the Swin transformer and top-5-accuracy of 99.04% is obtained by it while classifying 11 faults of IR images.

Statistics of Galactic-scale Quasar Pairs at Cosmic Noon

The statistics of galactic-scale quasar pairs can elucidate our understanding of the dynamical evolution of supermassive black hole (SMBH) pairs, the duty cycles of quasar activity in mergers, or even the nature of dark matter, but they have been challenging to measure at cosmic noon, the prime epoch of massive galaxy and SMBH formation. Here we measure a double quasar fraction of ∼6.2 ± 0.5 × 10−4 integrated over ∼0.″3–3″ separations (projected physical separations of ∼3–30 kpc at z ∼ 2) in luminous (L bol > 1045.8 erg s−1) unobscured quasars at 1.5 < z < 3.5 using Gaia EDR3-resolved pairs around SDSS DR16 quasars. The measurement was based on a sample of 60 Gaia-resolved double quasars (out of 487 Gaia pairs dominated by quasar+star superpositions) at these separations, corrected for pair completeness in Gaia, which we quantify as functions of pair separation, magnitude of the primary, and magnitude contrast. The double quasar fraction increases toward smaller separations by a factor of ∼5 over these scales. The division between physical quasar pairs and lensed quasars in our sample is currently unknown, requiring dedicated follow-up observations (in particular, deep, subarcsecond-resolution IR imaging for the closest pairs). Intriguingly, at this point, the observed pair statistics are in rough agreement with theoretical predictions both for the lensed quasar population in mock catalogs and for dual quasars in cosmological hydrodynamic simulations. Upcoming wide-field imaging/spectroscopic space missions such as Euclid, CSST, and Roman, combined with targeted follow-up observations, will conclusively measure the abundances and host galaxy properties of galactic-scale quasar pairs, offset AGNs, and subarcsecond lensed quasars across cosmic time.

Infrared Small Target Detection Algorithm Using an Augmented Intensity and Density-Based Clustering

In infrared search and tracking (IRST) systems, small target detection is challenging because IR imaging lacks feature information and has a low signal-to-noise ratio. The recently studied small IR target detection methods have achieved high detection performance without considering execution time. We propose a fast and robust single-frame IR small target detection algorithm while maintaining excellent detection performance. The augmented infrared intensity map based on the standard deviation speeds up small target detection and improves detection accuracy. Density-based clustering helps to detect the shape of objects and makes it easy to identify centroid points. By incorporating these two approaches, the proposed method has a novel approach to the small target detection algorithm. We have self-built 300 images with various scenes and experimented with comparing other methods. Experimental results demonstrate that the proposed method is suitable for real-time detection and effective even when the target size is as small as 2 pixels.

Distribution of Domains Formed by Lateral Packing of Intercellular Lipid in the Stratum Corneum.

Intercellular lipids fill the interstices of corneocytes and serve a barrier function. The amount of transdermal water evaporation varies depending on the packing structure of intercellular lipids, as this structure is important for maintaining barrier efficacy. This packing structure consists of a mixture of crystals (orthorhombic and hexagonal) and liquid crystals (fluid phase), and the proportion of these phases is thought to affect barrier function. However, there have been no methods to visualize the actual distribution of the domains formed by packing structure in intercellular lipids. In this study, the planar distribution of intercellular lipid structures was determined using focal plane array (FPA)-based Fourier transform (FT) IR imaging analysis of stratum corneum cell units obtained by grid stripping. The lipid composition of ceramides was revealed by electrospray ionization tandem mass spectrometry (ESI-MS/MS)-based shotgun lipidomics. The distribution of domains formed by packing structures and the lipid composition of ceramides was compared in skin with high- or low-transepidermal water loss (TEWL). The orthorhombic proportion was lower in high-TEWL skin than in low-TEWL skin. ESI-MS/MS-based shotgun lipidomics analysis showed that the alpha-hydroxyceramide content in the low- and high-TEWL groups differed regarding the distribution of fatty acid chain lengths. The evaluation of stratum corneum cell units using FPA-based FTIR imaging is an innovative technology that can visualize the distribution of domains formed by intercellular lipid-packing structures. Increased proportions of alpha-hydroxyceramide subclasses such as alpha-hydroxy-sphingosine ceramide and alpha-hydroxy-phytosphingosine ceramide were associated with a reduced proportion of the orthorhombic packing structure domain.