Vein Images Research Articles

Finger Vein Recognition Model for Biometric Authentication Using Intelligent Deep Learning

In recent years, biometric authentication systems have remained a hot research topic, as they can recognize or authenticate a person by comparing their data to other biometric data stored in a database. Fingerprints, palm prints, hand vein, finger vein, palm vein, and other anatomic or behavioral features have all been used to develop a variety of biometric approaches. Finger vein recognition (FVR) is a common method of examining the patterns of the finger veins for proper authentication among the various biometrics. Finger vein acquisition, preprocessing, feature extraction, and authentication are all part of the proposed intelligent deep learning-based FVR (IDL-FVR) model. Infrared imaging devices have primarily captured the use of finger veins. Furthermore, a region of interest extraction process is carried out in order to save the finger part. The shark smell optimization algorithm is used to tune the hyperparameters of the bidirectional long–short-term memory model properly. Finally, an authentication process based on Euclidean distance is performed, which compares the features of the current finger vein image to those in the database. The IDL-FVR model surpassed the earlier methods by accomplishing a maximum accuracy of 99.93%. Authentication is successful when the Euclidean distance is small and vice versa.

A multimodal architecture using Adapt‐HKFCT segmentation and feature‐based chaos integrated deep neural networks (Chaos‐DNN‐SPOA) for contactless biometricpalm vein recognition system

In the recent past, the fusion of various unimodal biometrics has gained increasing attention from researchers dedicated to the use of practical biometrics. In this paper, a Chaos Integrated Deep Neural Networks (Chaos-DNN) using Sandpiper Optimization Algorithm (SPOA) is proposed to enhance the performance of the multimodal contactless biometric pattern recognition system. The most significant advantage of palm vein and palm dorsal vein recognition is that it provides relatively high accuracy, reliability and also liveness detection. In this recognition process, the palm and its dorsal vein images are segmented using multilevel segmentation of region of interest (ROI) depending on Adaptive Hyper Kernel Optimized Fuzzy Clustering Technique (Adapt-HKFCT), and then ROI extracted based on the valley points among the index and ring finger using the Knuckle points. Then, the features of the ROI amid the index and ring finger are extracted using centre-symmetric local binary pattern algorithm and fed to the Chaos-DNN for training and classification. The efficiency of the features is optimized using SPOA. The extracted feature points help in authenticating a person using his/her chosen biometric traits for security purposes. The objective of this system is to increase the efficiency of the biometric system, and the performance is calculated depending on equal error rate, correct recognition rate, decidability index, false acceptance rate and false rejection rate. The proposed algorithm has been implemented in MATLAB platform. The experimental outcomes demonstrate that the proposed palm vein recognition depending Adapt-HKFCT with Chaos-DNN using SPOA method (PVR-AHKFCT-CDNN-SPOA) attains higher performance metrics with an accuracy of 98.3% and is also more efficient likened to the other existing methods, such as K Nearest Neighbour with Particle Swarm Optimization (PVR-KNN-PSO), Fuzzy Brain Storm Optimization based on Adaptive Thresholding (PVR-FBSO-AT), multiscale local binary pattern with ant colony optimization (PVR-MSLBP-ACO) and Deep Learning with Bayesian Optimization (PVR-DL-BO).

Explainable AI: A Multispectral Palm-Vein Identification System with New Augmentation Features

Recently, as one of the most promising biometric traits, the vein has attracted the attention of both academia and industry because of its living body identification and the convenience of the acquisition process. State-of-the-art techniques can provide relatively good performance, yet they are limited to specific light sources. Besides, it still has poor adaptability to multispectral images. Despite the great success achieved by convolutional neural networks (CNNs) in various image understanding tasks, they often require large training samples and high computation that are infeasible for palm-vein identification. To address this limitation, this work proposes a palm-vein identification system based on lightweight CNN and adaptive multi-spectral method with explainable AI. The principal component analysis on symmetric discrete wavelet transform (SMDWT-PCA) technique for vein images augmentation method is adopted to solve the problem of insufficient data and multispectral adaptability. The depth separable convolution (DSC) has been applied to reduce the number of model parameters in this work. To ensure that the experimental result demonstrates accurately and robustly, a multispectral palm image of the public dataset (CASIA) is also used to assess the performance of the proposed method. As result, the palm-vein identification system can provide superior performance to that of the former related approaches for different spectrums.

Palm Vein Identification Based on Hybrid Feature Selection Model

Palm Vein Identification (PVI) is a modern biometric security technique used for enhancing security and authentication systems. The key characteristics of palm vein patterns include its uniqueness to each individual, its unforgettability, non-intrusiveness and its ability for disallowing unauthorized persons. However, the extracted features from the palm vein patterns are huge with high redundancy. In this paper, we propose a combined model of two-Dimensional Discrete Wavelet Transform, Principal Component Analysis (PCA), and Particle Swarm Optimization (PSO) (2D-DWTPP) that feeds wrapper model with an optimal subset of features to enhance the prediction accuracy of -palm vein patterns. The 2D-DWT extract features from palm vein images, using the PCA to reduce the redundancy in palm vein features. The system has been trained to select high recognition features based on the wrapper model. The proposed system uses four classifiers as an objective function to determine PVI which include Support Vector Machine (SVM), K Nearest Neighbor (KNN), Decision Tree (DT) and Naive Bayes (NB). The empirical results proved that the proposed model has the best results with SVM. Moreover, our proposed 2D-DWTPP model has been evaluated and the results show remarkable efficiency in comparison with AlexNet and other classifiers without feature selection. Experimentally, the proposed model has better accuracy as reflected by 98.65% whereas AlexNet has 63.5% accuracy and the classifier without feature selection process has 78.79% accuracy.

Infrared Vein Imaging for Insertion of Peripheral Intravenous Catheter for Patients Requiring Isolation for Severe Acute Respiratory Syndrome Coronavirus 2 Infection: A Nonrandomized Clinical Trial

IntroductionEstablishing intravenous access is essential but may be difficult to achieve for patients requiring isolation for severe acute respiratory syndrome coronavirus 2 infection. This study aimed to investigate the effectiveness of an infrared vein visualizer on peripheral intravenous catheter therapy in patients with coronavirus disease 2019. MethodsA nonrandomized clinical trial was performed. In total, 122 patients with coronavirus disease 2019 who required peripheral intravenous cannulation were divided into 2 groups with 60 in the control group and 62 in the intervention group. A conventional venipuncture method was applied to the control group, whereas an infrared vein imaging device was applied in the intervention group. The first attempt success rate, total procedure time, and patients’ satisfaction score were compared between the 2 groups using chi-square, t test, and z test (also known as Mann-Whitney U test) statistics. ResultsThe first attempt success rate in the intervention group was significantly higher than that of control group (91.94% vs 76.67%, ꭓ2 = 5.41, P = .02). The procedure time was shorter in the intervention group (mean [SD], 211.44 [68.58] seconds vs 388.27 [88.97] seconds, t = 12.27, P < .001). Patients from the intervention group experienced a higher degree of satisfaction (7.5 vs 6, z = −3.31, P < .001). DiscussionPeripheral intravenous catheter insertion assisted by an infrared vein visualizer could improve the first attempt success rate of venipuncture, shorten the procedure time, and increase patients’ satisfaction.

Two-phase non-invasive multi-disease detection via sublingual region.

Non-invasive multi-disease detection is an active technology that detects human diseases automatically. By observing images of the human body, computers can make inferences on disease detection based on artificial intelligence and computer vision techniques. The sublingual vein, lying on the lower part of the human tongue, is a critical identifier in non-invasive multi-disease detection, reflecting health status. However, few studies have fully investigated non-invasive multi-disease detection via the sublingual vein using a quantitative method. In this paper, a two-phase sublingual-based disease detection framework for non-invasive multi-disease detection was proposed. In this framework, sublingual vein region segmentation was performed on each image in the first phase to achieve the region with the highest probability of covering the sublingual vein. In the second phase, features in this region were extracted, and multi-class classification was applied to these features to output a detection result. To better represent the characterisation of the obtained sublingual vein region, multi-feature representations were generated of the sublingual vein region (based on color, texture, shape, and latent representation). The effectiveness of sublingual-based multi-disease detection was quantitatively evaluated, and the proposed framework was based on 1103 sublingual vein images from patients in different health status categories. The best multi-feature representation was generated based on color, texture, and latent representation features with the highest accuracy of 98.05%.

Dorsal Hand Vein Image Enhancement Using Fusion of CLAHE and Fuzzy Adaptive Gamma.

Enhancement of captured hand vein images is essential for a number of purposes, such as accurate biometric identification and ease of medical intravenous access. This paper presents an improved hand vein image enhancement technique based on weighted average fusion of contrast limited adaptive histogram equalization (CLAHE) and fuzzy adaptive gamma (FAG). The proposed technique is applied using three stages. Firstly, grey level intensities with CLAHE are locally applied to image pixels for contrast enhancement. Secondly, the grey level intensities are then globally transformed into membership planes and modified with FAG operator for the same purposes. Finally, the resultant images from CLAHE and FAG are fused using improved weighted averaging methods for clearer vein patterns. Then, matched filter with first-order derivative Gaussian (MF-FODG) is employed to segment vein patterns. The proposed technique was tested on self-acquired dorsal hand vein images as well as images from the SUAS databases. The performance of the proposed technique is compared with various other image enhancement techniques based on mean square error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index measurement (SSIM). The proposed enhancement technique’s impact on the segmentation process has also been evaluated using sensitivity, accuracy, and dice coefficient. The experimental results show that the proposed enhancement technique can significantly enhance the hand vein patterns and improve the detection of dorsal hand veins.

DESIGN AND FABRICATION OF HIGH QUALITY IMAGING OBJECTIVE WORKING IN THE NEAR-INFRARED SPECTRAL REGION FOR VEIN FINDER DEVICES

This work demonstrates a new design of an imaging objective operating in the near-infrared spectral region for vein finder devices. The objective has a simple structure, consisting of 6 elements divided into 2 groups and a long-pass optical filter that allows a near-infrared spectrum to pass through. Due to the filter, the objective only creates images in the 850-940 nm spectral region produced by the infrared illuminator and therefore it can eliminate optical noise in the visible region. The objective image quality is very high, suitable for vein imaging: large aperture, f/2.0, allows creating clear vein images from weak signal reflected from the human skin surface; modulation transfer function (MTF) of 0.5 at a frequency of 72.5 lines/mm; field curvature and image distortion less than 50 μm and 0.1%, respectively. The experimental result indicates that the objective can provide clear vein images when combining with a Sony IMX287 CMOS image sensor.

Integration of Two-Dimensional Kernel Principal Component Analysis Plus Two-Dimensional Linear Discriminant Analysis with Convolutional Neural Network for Finger Vein Recognition

High complexity and low recognition rate are two common problems with the current finger vein recognition methods. To solve these problems, this paper integrates two-dimensional kernel principal component analysis (K2DPCA) plus two-dimensional linear discriminant analysis (2DLDA) (K2DPCA+2DLDA) into convolutional neural network (CNN) to recognize finger veins. Considering the row and column correlations of the finger vein image matrix and the classes of finger vein images, the authors adopted K2DPCA and 2DLDA separately for dimensionality reduction and extraction of nonlinear features in row and column directions, producing a dimensionally reduced compressed image without row or column correlation. Taking the dimensionally reduced compressed image as the input, the CNN was introduced to learn higher-level features, making finger vein recognition more accurate and robust. The public dataset of Finger Vein USM (FV-USM) Database was adopted for experimental verification. The results show that the proposed approach effectively overcome the common defects of original image feature extraction: the insufficient feature description, and the redundancy of information. When the training reached 120 epochs, the model basically realized stable convergence, with the loss approaching zero and the recognition rate reaching 97.3%. Compared with two-directional two-dimensional Fisher principal component analysis ((2D)2FPCA), our strategy, which integrates K2DPCA+2DLDA with CNN, achieved a very high recognition rate of finger vein images.

Fast Correction of “Finite Aperture Effect” in Photoacoustic Tomography Based on Spatial Impulse Response

Photoacoustic computed tomography (PACT) is a fast-developing imaging technique, which can provide structural and functional information in biological tissues with high-resolution beyond the depth of the optical diffusion limit. However, the most current PACT reconstruction method generally employs a point detector assumption, whereas in most PAT systems with circular or spherical scanning modes, the transducer is mostly flat and with a finite size. This model mismatch leads to a notable deterioration in the lateral direction in regions far from the rotation center, which is known as the “finite aperture effect”. In this work, we propose to compensate a novel Back-projection (BP) method based on the transducer’s spatial impulse response (SIR) for fast correction of the “finite aperture effect”. The SIR accounts for the waveform change of the transducer for an arbitrary point source due to the geometry of the detection surface. Simulation results showed that the proposed SIR-BP method can effectively improve the lateral resolution and signal to noise ratio (SNR) in the off-center regions. For a target 4.5 mm far from the rotation center, this new method improved the lateral resolution about five times along with a 7 dB increase in the SNR. Experimental results also showed that this SIR-BP method can well restore the image angular blur to recover small structures, as demonstrated by the imaging of leaf veins. This new method offers a valuable alternative to the conventional BP method, and can guide the design of PAT systems based on circular/spherical scan.

Improvement of confusion matrix for Hand Vein Recognition Based On Deep- Learning multi-classifier Decisions

In this paper, recognition of the hand vein patterns approach is proposed employing the Convolutional Neural Network (CNN). This approach is routinely well-learned in what way to get features from the main pattern using Region of Interest (ROI). Though, the poor quality of the hand vein image still attitudes an unlimited strain to the extension leads of its usability. Firstly, by applying the method of Generative adversarial networks (GAN) data augmentation the performance gain of adding GAN generated data exceeds that of adding more true images, and apply ROI in a hand vein image feature extraction is studied initially. Secondly, the suggested approach is tested on the data sets of hand veins to decrease the overfitting in the fully connecting layer of CNN which this model proves the most effective one. In total, 1575 hand vein images from 100 subjects are applied to authorize the proposed approach for hand vein. A high accuracy (>99.8%) and low False Rejection Rate(FRR) (<0.99%) were achieved by applying the suggested approach, when compared with the existing CNN classifiers, indicating the efficiency of the suggested approach.

Efficient broadband near-infrared luminescence of Cr3+ doped fluoride K2NaInF6 and its NIR-LED application toward veins imaging

High performance broadband Near-infrared (NIR) phosphors with high efficiency and excellent thermal stability are highly demanded for constructing the next generation of NIR radiation sources. Herein, a broadband NIR emitting fluoride phosphor K2NaInF6:Cr3+ peaking at 774 nm with a Full width at half maximum (FWHM) of 116 nm (1906 cm−1) was demonstrated. A high Internal quantum efficiency (IQE) value of 70.2% can be obtained in the as-synthesized K2NaInF6:Cr3+ phosphor. Additionally, benefiting from the relatively weak electron phonon coupling effect, the phosphor also exhibits excellent anti-thermal quenching behavior with the emission intensity at 150 °C retaining 78.3% of that at room temperature. A high power NIR pc-LED with NIR radiant flux of 492.85 mW@1050 mA is obtained by combing the as-synthesized K2NaInF6:Cr3+ phosphor and a blue LED chip. Utilizing this high power NIR pc-LED as NIR radiation source, high-resolution human hand veins imaging is realized with the assistance of a NIR camera, which indicates that K2NaInF6:Cr3+ phosphor has high potential in biological applications.

Vortex Vein Imaging: What Can It Tell Us?

This review article summarizes the patho-anatomy of the vortex veins, the major drainage channels for the choroid, and describes the various pathways of diseases associated with vortex vein abnormalities. This report also details the technical advancements to image the vortex veins, such as ultra-widefield indocyanine green angiography, which are critical to elucidate the importance of the vortices in various retino-choroidal disorders. Future applications of these advanced imaging systems to better understand the role of the vortex veins in health and disease are also discussed.

INTELLIGENT sYSTEM FOR HUMAN AUTHENTICATION USING FUSION OF DORSAL HAND, PALM AND FINGER VEINS

Multimodal biometric systems roughly used to achieve extreme recognition accuracy. This paper reports a novel multimodal biometric system employing intelligent technique to authenticate human by fusion of dorsal hand, palm and finger veins pattern. We improved an image analysis technique to separate region of interest (ROI) from dorsal hand, palm and finger veins image. After separating ROI we construct a sequence of preprocessing steps to enhance dorsal hand, palm and finger veins images using Median filter, Wiener filter, Contrast Limited Adaptive Histogram Equalization (CLAHE) and Homomorphic filter to improve vein image. Our intelligent technique is based on the following intelligent algorithms, namely; principal component analysis (PCA) algorithm for feature extraction and k-Nearest Neighbors (K-NN) classifier for matching operation. The database selected was Bosphorus Hand Vein Database, CASIA Multi-Spectral Palmprint Image Database V1.0 (CASIA database) and the Shandong University Machine Learning and Applications - Homologous Multi-modal Traits (SDUMLA-HMT). The accomplished result for the fusion of three biometric traits was Correct Recognition Rate (CRR) is 99.21% with False Reject Rate (FRR) 0.04%.

Portal vein thrombosis following elective laparoscopic splenectomy: incidence and analysis of risk factors.

Minimally invasive splenectomy is now well established for a wide range of pathologies. Portal vein thrombosis (PVT) is increasingly being recognised as a complication of splenectomy. The aim was to determine the incidence and risk factors for PVT after laparoscopic splenectomy. All cases of elective laparoscopic splenectomy performed from 1993 to 2020 were reviewed. Parameters recorded included demographics, diagnostic criterion and post-operative outcomes. Data were analysed using Minitab V18 with a p < 0.05 considered significant. 210 patients (103 female, 107 male) underwent laparoscopic splenectomy (14 to 85years). A major proportion of cases were performed for ITP (n = 77, p = 0.012) followed by lymphoma (n = 28), indeterminate lesions (n = 21) and myelofibrosis (n = 19). Ten patients developed symptomatic portal vein thrombosis (4.8%). Patients presented most commonly with pain and fever and diagnosis was confirmed by computed tomography (CT) or ultrasonography (USS). There were 10 conversions (4.8%) to open and two postoperative deaths, one from PVT and one from pneumonia. The remaining nine patients were successfully treated with anticoagulation. Of 19 patients with myelofibrosis, six patients developed PVT (p = 0.0002). Patients who developed PVT had significantly greater specimen weights (1773g vs 348g, p < 0.001). Forty-three patients had a specimen weight of 1kg or greater, and of these 9 developed portal vein thrombosis (21%), versus one with PVT of 155 with a specimen weight of less than 1kg (p < 0.0001). Myelofibrosis (p = 0.0039), specimen weight (p < 0.001) and mean platelet count (p = 0.0049) were predictive of PVT. A high index of suspicion for this complication should be maintained and prompt treatment with anticoagulation. High-risk patients should be considered for prophylactic anticoagulation and routine imaging of the portal vein.

Effect of Interventional Therapy on Iliac Venous Compression Syndrome Evaluated and Diagnosed by Artificial Intelligence Algorithm-Based Ultrasound Images.

In order to explore the efficacy of using artificial intelligence (AI) algorithm-based ultrasound images to diagnose iliac vein compression syndrome (IVCS) and assist clinicians in the diagnosis of diseases, the characteristics of vein imaging in patients with IVCS were summarized. After ultrasound image acquisition, the image data were preprocessed to construct a deep learning model to realize the position detection of venous compression and the recognition of benign and malignant lesions. In addition, a dataset was built for model evaluation. The data came from patients with thrombotic chronic venous disease (CVD) and deep vein thrombosis (DVT) in hospital. The image feature group of IVCS extracted by cavity convolution was the artificial intelligence algorithm imaging group, and the ultrasound images were directly taken as the control group without processing. Digital subtraction angiography (DSA) was performed to check the patient's veins one week in advance. Then, the patients were rolled into the AI algorithm imaging group and control group, and the correlation between May–Thurner syndrome (MTS) and AI algorithm imaging was analyzed based on DSA and ultrasound results. Satisfaction of intestinal venous stenosis (or occlusion) or formation of collateral circulation was used as a diagnostic index for MTS. Ultrasound showed that the AI algorithm imaging group had a higher percentage of good treatment effects than that of the control group. The call-up rate of the DMRF-convolutional neural network (CNN), precision, and accuracy were all superior to those of the control group. In addition, the degree of venous swelling of patients in the artificial intelligence algorithm imaging group was weak, the degree of pain relief was high after treatment, and the difference between the artificial intelligence algorithm imaging group and control group was statistically considerable (p < 0.005). Through grouped experiments, it was found that the construction of the AI imaging model was effective for the detection and recognition of lower extremity vein lesions in ultrasound images. To sum up, the ultrasound image evaluation and analysis using AI algorithm during MTS treatment was accurate and efficient, which laid a good foundation for future research, diagnosis, and treatment.

Trustworthy authorization method for security in Industrial Internet of Things

Industrial Internet of Things (IIoT) realizes machine-to-machine communication and human–computer interaction (HCI) through communication network, which makes industrial production automatic and intelligent. Security is critical in IIoT because of the interconnection of intelligent industrial equipment. In IIoT environment, legitimate human–computer interaction can only be performed by authorized professionals, and unauthorized access is not tolerated. In this paper, a reliable authentication method based on biological information is proposed. Specifically, the complete local binary pattern (CLPB) and the statistical local binary pattern (SLPB) are introduced to describe the local vein texture characteristics. Meanwhile, the contrast energy and frequency domain information are regarded as auxiliary information to interpret the finger vein. The distance between the features of the registration image and the test image is used to recognize the finger vein image, so as to realize identity authentication. The experiments are carried out on SDUMLA-FV database and FV-USM database, and results show that the presented method has achieved high recognition accuracy.

Intelligent Technique for Human Authentication using Fusion of Finger and Dorsal Hand Veins

Multimodal biometric systems have been widely used to achieve high recognition accuracy. This paper presents a new multimodal biometric system using intelligent technique to authenticate human by fusion of finger and dorsal hand veins pattern. We developed an image analysis technique to extract region of interest (ROI) from finger and dorsal hand veins image. After extracting ROI we design a sequence of preprocessing steps to improve finger and dorsal hand veins images using Median filter, Wiener filter and Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance vein image. Our smart technique is based on the following intelligent algorithms, namely; principal component analysis (PCA) algorithm for feature extraction and k-Nearest Neighbors (K-NN) classifier for matching operation. The database chosen was the Shandong University Machine Learning and Applications - Homologous Multi-modal Traits (SDUMLA-HMT) and Bosphorus Hand Vein Database. The achieved result for the fusion of both biometric traits was Correct Recognition Rate (CRR) is 96.8%.

Hydrothermal fluid evolution in the Escondida porphyry copper deposit, northern Chile: evidence from SEM-CL imaging of quartz veins and LA-ICP-MS of fluid inclusions

The origin of hypogene alteration and mineralization features in the Escondida porphyry Cu deposit resulted from intense overprinting related to three main hydrothermal stages. The beginning of each stage is recorded by deposition of bright quartz crystals on vein walls that precipitated from high-temperature fluids. In the deepest zones of the deposit, the earliest stage started with exsolution of intermediate-density fluids, which transported high concentrations of Cu, whereas Mo was not detected. However, in the shallow zones, the early stage began with depressurization and unmixing of intermediate-density fluids, which generated both a hypersaline and vapour-rich fluid phase, producing an important geochemical segregation between both phases. The transitional stage also started with circulation of intermediate-density fluids but never experienced unmixing. These fluids transported the highest Mo concentrations in the deposit; however, Cu displays lower concentrations relative to intermediate-density fluids from the early stage. The beginning of the late stage was also associated with intermediate-density fluids; however, Cu and Mo were below the detection limits in most of the analysed fluids. During the evolution of the three stages, the fluids experienced gradual cooling, which promoted the precipitation of euhedral and zoned quartz crystals that overgrew the early high-luminescence quartz generations. Sometimes, sulfide minerals display euhedral crystal boundaries with zoned quartz, suggesting that mineralization started during these phases. The latest events detected in each main stage are linked to cooler and low-salinity fluids, from which dark quartz and hypogene sulfides precipitated along microfractures and interstitial spaces developed in the earlier quartz generations.

Towards Protecting and Enhancing Vascular Biometric Recognition Methods via Biohashing and Deep Neural Networks

Biometric template protection has been a crucial concern in biometric recognition systems. This is because biometric characteristics are irreplaceable, so the compromised templates can also be used in other applications. On the other side, deploying template protection algorithms often affects the performance of biometric systems. In this paper, we consider both raw and pre-processed finger vein images and propose a novel deep-learning-based framework to protect biometric templates and enhance recognition performance. We use a deep convolutional auto-encoder structure to reduce the dimension of the feature space, and then secure templates by applying the Biohashing algorithm on the features extracted at the bottleneck layer of our auto-encoder. The experimental results indicate that the protected templates through our framework achieve superior performance than Biohash protected templates of the raw features in the normal scenario. In the stolen scenario, where the Biohashing key is stolen, our model yields far better performance than Biohashing of raw features extracted by previous recognition methods. We also evaluate the generalization of our proposed framework on other vascular biometric modalities. It is worth mentioning that we provide an open-source implementation of our framework so that other researchers can verify our findings and build upon our work.