Performance Of Finger Vein Recognition Research Articles

A Comparative Study of Cross-Device Finger Vein Recognition Using Classical and Deep Learning Approaches

Finger vein recognition is gaining popularity in the field of biometrics, yet the inter-operability of finger vein patterns has received limited attention. This study aims to fill this gap by introducing a cross-device finger vein dataset and evaluating the performance of finger vein recognition across devices using a classical method, a convolutional neural network, and our proposed patch-based convolutional auto-encoder (CAE). The findings emphasise the importance of standardisation of finger vein recognition, similar to that of fingerprints or irises, crucial for achieving inter-operability. Despite the inherent challenges of cross-device recognition, the proposed CAE architecture in this study demonstrates promising results in finger vein recognition, particularly in the context of cross-device comparisons.

Multi scale-aware attention for pyramid convolution network on finger vein recognition

In recent years, biometrics has been the most popular style of personal identification. The finger vein is an intrinsic and stable trait, and with the ability to detect liveness, it receives academic and industry attention. However, convolution neural networks (CNNs) based finger vein recognition generally can only cover a small input region by using small kernels. Hence, the performance is poor, facing low-quality finger vein images. It is a challenge to effectively use the critical feature of multi-scale for finger veins. In this article, we extract multi-scale features via pyramid convolution. We propose scale attention, namely, the scale-aware attention (SA) module, which enables dynamic adjustment of the weight of each scale to information aggregation. Utilize the complementation of different scale detail features to enhance the discriminativeness of extracted features, thus improving the finger vein recognition performance. In order to verify the present method’s efficiency, we carried out experiments on two public data sets and one internal data, and the wide range of experimental results proves the proposed method’s effectiveness.

Raspberry PI-Based Finger Vein Recognition System Using PCA NET

Despite simultaneously ignoring the intensity distribution that is formed by the finger tissue and, in some instances, processing it as background noise, the majority of finger vein feature extraction algorithms achieve satisfactory performance due to their ability to represent texture. This project makes use of two- directional two-dimensional Fisher Principal Component Analysis, also known as (2D) 2 FPCA, for feature extraction. This type of "noise" is presented as a novel soft biometric trait for improving finger vein recognition performance. In order to demonstrate that the intensity distribution that is formed by the finger tissue in the background can be extracted as a soft biometric trait for recognition, a comprehensive analysis of the finger vein imaging principle and the characteristics of the image are first presented.

Residual Gabor convolutional network and FV-Mix exponential level data augmentation strategy for finger vein recognition

Using deep learning to improve the performance of finger vein recognition has become a part of mainstream research. Currently, the performance of finger vein recognition systems is limited by insufficient finger vein training samples, which leads to insufficient feature learning and weak model generalization. To solve the above problems, first, we propose a simple and effective finger vein data augmentation strategy named FV-Mix, which uses fine finger vein image region of interest (ROI) for grayscale normalization and linear mixing. It can accomplish exponential-level data augmentation on the training sample, and the augmented data will represent a more complete dataset. Second, a residual Gabor convolutional network (RGCN) is designed for finger vein recognition, which includes a residual Gabor convolutional layer (RGCL) and a dense semantic analysis module (DSAM). The RGCL is designed to replace the shallow convolutional layer in the network by using the characteristics of the Gabor filter to enhance the scale and direction of information in the shallow pattern features. And then, the enhanced deep features are further extracted and analyzed by DSAM, which is used to assist the final model in classifying and recognizing finger vein images. To verify the effectiveness of our work, five publicly available finger vein image datasets are used, and a large number of comparison experiments are designed. The proposed FV-Mix strategy, RGCL module, and DSAM module were fully validated. The experimental results show that on these five datasets, the average recognition accuracy and equal error rate (EER) of the proposed RGCN are 99.22% and 0.188%, respectively, achieves competitive performance compared with current state-of-the-art work.

ViT-Cap: A Novel Vision Transformer-Based Capsule Network Model for Finger Vein Recognition

Finger vein recognition has been widely studied due to its advantages, such as high security, convenience, and living body recognition. At present, the performance of the most advanced finger vein recognition methods largely depends on the quality of finger vein images. However, when collecting finger vein images, due to the possible deviation of finger position, ambient lighting and other factors, the quality of the captured images is often relatively low, which directly affects the performance of finger vein recognition. In this study, we proposed a new model for finger vein recognition that combined the vision transformer architecture with the capsule network (ViT-Cap). The model can explore finger vein image information based on global and local attention and selectively focus on the important finger vein feature information. First, we split-finger vein images into patches and then linearly embedded each of the patches. Second, the resulting vector sequence was fed into a transformer encoder to extract the finger vein features. Third, the feature vectors generated by the vision transformer module were fed into the capsule module for further training. We tested the proposed method on four publicly available finger vein databases. Experimental results showed that the average recognition accuracy of the algorithm based on the proposed model was above 96%, which was better than the original vision transformer, capsule network, and other advanced finger vein recognition algorithms. Moreover, the equal error rate (EER) of our model achieved state-of-the-art performance, especially reaching less than 0.3% under the test of FV-USM datasets which proved the effectiveness and reliability of the proposed model in finger vein recognition.

Restoration of Motion Blurred Image by Modified DeblurGAN for Enhancing the Accuracies of Finger-Vein Recognition.

Among many available biometrics identification methods, finger-vein recognition has an advantage that is difficult to counterfeit, as finger veins are located under the skin, and high user convenience as a non-invasive image capturing device is used for recognition. However, blurring can occur when acquiring finger-vein images, and such blur can be mainly categorized into three types. First, skin scattering blur due to light scattering in the skin layer; second, optical blur occurs due to lens focus mismatching; and third, motion blur exists due to finger movements. Blurred images generated in these kinds of blur can significantly reduce finger-vein recognition performance. Therefore, restoration of blurred finger-vein images is necessary. Most of the previous studies have addressed the restoration method of skin scattering blurred images and some of the studies have addressed the restoration method of optically blurred images. However, there has been no research on restoration methods of motion blurred finger-vein images that can occur in actual environments. To address this problem, this study proposes a new method for improving the finger-vein recognition performance by restoring motion blurred finger-vein images using a modified deblur generative adversarial network (modified DeblurGAN). Based on an experiment conducted using two open databases, the Shandong University homologous multi-modal traits (SDUMLA-HMT) finger-vein database and Hong Kong Polytechnic University finger-image database version 1, the proposed method demonstrates outstanding performance that is better than those obtained using state-of-the-art methods.

Finger-Vein Recognition Using Bidirectional Feature Extraction and Transfer Learning

Accuracy and efficiency are essential topics in the current biometric feature recognition and security research. This paper proposes a deep neural network using bidirectional feature extraction and transfer learning to improve finger-vein recognition performance. Above all, we make a new finger-vein database with the opposite position information of the original one and adopt transfer learning to make the network suitable for our overall recognition framework. Next, the feature extractor is constructed by adjusting the unidirectional database’s parameters, capturing vein features from top to bottom and vice versa. Correspondingly, we concatenate the above two features to form the finger-veins’ bidirectional features, which are trained and classified by Support Vector Machines (SVM) to realize recognition. Experiments are conducted on the Malaysian Polytechnic University’s published database (FV-USM) and finger veins of Signal and Information Processing Laboratory (FV-SIPL). The accuracy of our proposed algorithm reaches 99.67% and 99.31%, which is significantly higher than the unidirectional recognition under each database. Compared with the algorithms cited in this paper, our proposed model based on bidirectional feature enjoys higher accuracy, faster recognition speed than the state-of-the-art frameworks, and excellent practical value.

The biometric recognition system based on near-infrared finger vein image

It is a difficult task to extract vein features accurately since the finger-vein images captured by near infrared light are always poor in quality. This paper proposes a novel finger vein feature representation scheme based on pyramid histograms of oriented gradients and local phase quantization. As the vein networks consist of abundant texture and orientation features, a texture feature description operator at various scales is employed on the finger vein image to reduce the effects of geometric deformation occurred image acquisition due to the different posture and position of fingers. To solve the adverse effects of image blurring caused by uneven illumination, local phase quantization is then introduced to extract vein features. Finally, the above mentioned extracted two kinds of texture characteristics of vein image are fused at feature level by concatenated histograms to obtain accurate vein feature named pyramid local phase quantization histogram (PLPQ). In this way, we encode the vein image information not only in frequency domain but also among different orientations and scales. We perform rigorous experiments on two publicly available databases named FV-USM and MMCBUN, and the results of the experiments reveal that proposed fusion system can make promising improvement of finger vein recognition performance.

Fusion loss and inter-class data augmentation for deep finger vein feature learning

Finger vein recognition (FVR) based on deep learning (DL) has gained rising attention in recent years. However, the performance of FVR is limited by the insufficient amount of finger vein training data and the weak generalization of learned features. To address these limitations and improve the performance, we propose a simple framework by jointly considering intensive data augmentation, loss function design and network architecture selection. Firstly, we propose a simple inter-class data augmentation technique that can double the number of finger vein training classes with new vein patterns via vertical flipping. Then, we combine it with conventional intra-class data augmentation methods to achieve highly diversified expansion, thereby effectively resolving the data shortage problem. In order to enhance the discrimination of deep features, we design a fusion loss by incorporating the classification loss and the metric learning loss. We find that the fusion of these two penalty signals will lead to a good trade-off between the intra-class similarity and inter-class separability, thereby greatly improving the generalization ability of learned features. We also investigate various network architectures for FVR application in terms of performances and model complexities. To examine the reliability and efficiency of our proposed framework, we implement a real-time FVR system to perform end-to-end verification in a near-realworld working condition. In challenging open-set evaluation protocol, extensive experiments conducted on three public finger vein databases and an in-house database confirm the effectiveness of the proposed method.

Finger Vein Recognition via Sparse Reconstruction Error Constrained Low-Rank Representation

Vein pattern-based methods have powerfully promoted the performance of finger vein recognition. However, it is not easy to precisely extract vein patterns from images, especially from low-quality images, and the non-vein area have been proved to be helpful for recognition. This paper proposes to use low-rank representation to extract as much noiseless discriminative information as possible from finger vein images. However, image deformation and image quality variations weaken the correlation of genuine images, and therefore damage the low-rank linear representation. To further deal with this problem, the class labels of training images and the local geometric structure between testing images and training images, reflected by sparse reconstruction errors of testing images, are used as constraints of low-rank coefficients. In particular, vein backbone decomposition based sparse representation is proposed to fast compute the deformation-robust reconstruction errors of each testing image. The reconstruction errors on sub-backbones of one training image are summed and modified as the constraint of the low-rank coefficient on this training image. We evaluate the proposed method on three widely used finger vein databases, and experimental results show that the proposed method performs well in finger vein recognition.

Finger Vein Recognition Based on Fusion of Deformation Information

As a convenient and fraud-proof biometric, finger vein has received increasing attention from researchers. However, the deformation problem in finger vein images is still a challenge preventing the performance of finger vein recognition from prefect. In our opinion, the deformations are distinguishable between genuine and imposter fingers rather than performing as disturbing noises. It is because that the diversity between genuine images is mainly caused by different finger postures, and this difference can be characterized and utilized. While in the imposter matching, it is not the case. In this paper, we design a framework to ensemble the encoded deformation information with traditional texture-based method to improve the recognition performance. First, the pixel-level features are extracted to represent the finger vein images. Next, the best matching for each pixel is discovered by an optimized matching. Consequently, the displacement of each pixel is obtained, and the texture features of deformation is extracted from the displacement matrices. Furthermore, the results of direct and optimized matching for pixel-level features, together with the encoded deformations are fused using weights learned by the Support Vector Machine (SVM) model. Evaluated on the publicly available databases HKPU and SDU-MLA, extensive experiments demonstrate the superiority of the proposed method, with the Equal Error Rates (EERs) of 0.0040 and 0.0266 respectively on the two databases, which are much lower than recognition solely with deformation information. The proposed method is also applicable in high security scenarios, the False Reject Rate at zero False Accept Rate (FRR-at-0-FAR) can be reduced to 0.0363 on the HKPU database through adjusting the fusion parameters.

From Noise to Feature: Exploiting Intensity Distribution as a Novel Soft Biometric Trait for Finger Vein Recognition

Most finger vein feature extraction algorithms achieve satisfactory performance due to their texture representation abilities, despite simultaneously ignoring the intensity distribution that is formed by the finger tissue, and in some cases, processing it as background noise. In this paper, we exploit this kind of noise as a novel soft biometric trait for achieving better finger vein recognition performance. First, a detailed analysis of the finger vein imaging principle and the characteristics of the image are presented to show that the intensity distribution that is formed by the finger tissue in the background can be extracted as a soft biometric trait for recognition. Then, two finger vein background layer extraction algorithms and three soft biometric trait extraction algorithms are proposed for intensity distribution feature extraction. Finally, a hybrid matching strategy is proposed to solve the issue of dimension difference between the primary and soft biometric traits on the score level. A series of rigorous contrast experiments on three open-access databases demonstrates that our proposed method is feasible and effective for finger vein recognition.

Weber Local Descriptors With Variable Curvature Gabor Filter for Finger Vein Recognition

Since captured finger vein images are usually of low quality, it is challenging to extract reliable finger vein features directly from the original finger vein images. Most current methods utilize texture change information and rich line features to extract features from finger vein images while neglecting the curvature of the finger veins. In this paper, a Weber local descriptor (WLD) with variable curvature Gabor filters is proposed for finger vein recognition. First, the differential excitation operator in the original WLD is improved by adding directional information, thereby allowing the local texture changes in an image to be better characterized and enhancing the differences between heterogeneous finger veins. Then, variable curvature Gabor filters are introduced to extract finger vein features that can simultaneously reflect the directional information and the curvature of the finger veins. In fact, two response values from the proposed Gabor filters are employed as features for each pixel; this approach is equivalent to defining intervals for the line features, rather than single values, and makes the results more robust to the rotation. The extensive experiments on the SDUMLA-FV and PolyU databases demonstrate that the proposed method can effectively improve the performance of finger vein recognition and show good robustness to translation, rotation, and illumination.

Geometric shape analysis based finger vein deformation detection and correction

Abstract Image deformation degrades the performance of finger vein recognition seriously. Some deformation-robust feature extraction and matching methods have been proposed to deal with this problem. However, the deformed image cannot be detected and corrected by these methods. So, it is essential to carry out the image deformation detection and correction in the preprocessing stage. This paper proposes a geometric analysis based detection and correction (GADC) method for deformable finger vein recognition. Specifically, the geometric shape of the finger is analyzed firstly. And then, the deformed finger vein image is detected based on the variation of the finger geometric shape. Lastly, the deformed image is corrected by a linear transformation and a nonlinear transformation. The experiments, on two opened databases and one self-built deformation database, show that our approach can improve the recognition performance of the deformed finger vein images.

A comparative study of finger vein recognition by using Learning Vector Quantization

¾ This paper presents a comparative study of finger vein recognition using various features with Learning Vector Quantization (LVQ) as a classification method. For the purpose of this study, two main features are employed: Scale Invariant Feature Transform (SIFT) and Local Extensive Binary Pattern (LEBP). The other features that formed LEBP features: Local Multilayer Binary Pattern (LmBP) and Local Directional Binary Pattern (LdBP) are also employed. The type of images are also become the base of comparison. The SIFT features will be extracted from two types of images which are grayscale and binary images. The feature that have been extracted become the input for recognition stage. In recognition stage, LVQ classifier is used. LVQ will classify the images into two class which are the recognizable images and non recognizable images. The accuracy, false positive rate (FPR), and true positive rate (TPR) value are used to evaluate the performance of finger vein recognition. The performance result of finger vein recognition becomes the main study for comparison stage. From the experiments result, it can be found which feature is the best for finger vein reconition using LVQ. The performance of finger vein recognition that use SIFT feature from binary images give a slightly better result than uisng LmBP, LdBP, or LEBP feature. The accuracy value could achieve 97,45%, TPR at 0,9000 and FPR at 0,0129.

Finger-vein recognition with modified binary tree model

Finger-vein recognition is an increasingly promising biometric identification technology in terms of its high identification accuracy and prominent security performance. The main challenge faced by finger-vein recognition is the low recognition performance caused by segmentation error and local difference. To tackle this challenge, a finger-vein recognition method with modified binary tree (MBT) model is proposed in this paper. MBT model is used to describe the relationship and spatial structure of vein branches quantitatively. Based on the MBT model, four stages including rough selection, model correction, segment matching, and comprehensive judgment are presented to achieve a robust matching for finger-vein. Experiments demonstrate that the proposed method can boost the performance of finger-vein recognition that is degraded by segmentation error and local difference. While maintaining low complexity, the proposed method achieves 0.12 % equal error rate in the introduced dataset with 8,100 finger-vein images from 150 participants, which outperforms the state-of-the-art methods.

Exploring soft biometric trait with finger vein recognition

Soft biometric trait has been used as ancillary information to enhance the recognition accuracy for face, fingerprint, gait, iris, etc. In this paper, we present a new investigation of soft biometric trait to improve the performance of finger vein recognition. We first propose some extraction criteria of soft biometric trait for comprehensively understanding this kind of ancillary information. And then based on these criteria, the width of phalangeal joint is employed as a novel soft biometric trait, which can be directly extracted from finger vein image. Finally, three frameworks are developed to conduct the combination of the width measurement and finger vein pattern, i.e., the fusion framework, the filter framework and the hybrid framework. We perform rigorous experiments both on the open and self-built finger vein databases, and experimental results illustrate that soft biometric trait can make promising improvement of finger vein recognition performance.

Fusion of Band Limited Phase Only Correlation and Width Centroid Contour Distance for finger based biometrics

In this paper, a new approach of multimodal finger biometrics based on the fusion of finger vein and finger geometry recognition is presented. In the proposed method, Band Limited Phase Only Correlation (BLPOC) is utilized to measure the similarity of finger vein images. Unlike previous methods, BLPOC is resilient to noise, occlusions and rescaling factors; thus can enhance the performance of finger vein recognition. As for finger geometry recognition, a new type of geometrical features called Width-Centroid Contour Distance (WCCD) is proposed. This WCCD combines the finger width with Centroid Contour Distance (CCD). As compared with the single type of feature, the fusion of W and CCD can improve the accuracy of finger geometry recognition. Finally, we integrate the finger vein and finger geometry recognitions by a score-level fusion method based on the weighted SUM rule. Experimental evaluation using our own database which was collected from 123 volunteers resulted in an efficient recognition performance where the equal error rate (EER) was 1.78% with a total processing time of 24.22ms.

Finger vein image quality evaluation using support vector machines

In an automatic finger-vein recognition system, finger-vein image quality is significant for segmentation, enhancement, and matching processes. In this paper, we propose a finger-vein image quality evaluation method using support vector machines (SVMs). We extract three features including the gradient, image contrast, and information capacity from the input image. An SVM model is built on the training images with annotated quality labels (i.e., high/low) and then applied to unseen images for quality evaluation. To resolve the class-imbalance problem in the training data, we perform oversampling for the minority class with random-synthetic minority oversampling technique. Cross-validation is also employed to verify the reliability and stability of the learned model. Our experimental results show the effectiveness of our method in evaluating the quality of finger-vein images, and by discarding low-quality images detected by our method, the overall finger-vein recognition performance is considerably improved.