Finger vein verification system based on sparse representation

Recently, finger vein sensors have been embedded in all kinds of electronic devices for personal identification, such as intelligent door locks and attendance machines. The embedded sensors are generally small, thus capturing only part of the finger vein. However, prior studies have focused on near-full finger vein recognition, without considering the partial finger vein image caused by the small imaging window of the finger vein sensor. This paper aims to study personal identification based on partial finger vein images, known as small-area finger vein recognition. The effect of the small-area finger vein on recognition performance is first analyzed by cutting out the local part from the near-full finger vein image to model a small-area finger vein image. Second, a small-area finger vein database is built using a commercial finger vein imaging device, in which the vein pattern from approximately one-third of one adult finger is captured. To explore more discriminative information from small-area finger vein images, we propose a locality-constrained consistent dictionary learning (LCDL) method to fuse multiple features for small-area finger vein recognition. Finally, the proposed method is evaluated on the self-built small-area finger vein database and four synthetic small-area finger vein databases. Experimental results show the promising recognition performance of the proposed method.

Recent years have seen a plenty of personal identification methods with different biometrics such as finger pattern, face, palm-print and vein. The majority of these methods focus on complex image data projections and transforms in Fourier space, wavelet space or other domains, which usually bring heavy load in computation and difficult understanding in perceptual intuition. Moreover, these methods, oriented to multiple samples learning, are constricted usually in application. Among so much biometrics, vein, as a living feature with high anti-counterfeiting capability, has attracted considerable attention. In this paper, we propose a structured personal identification approach using finger vein Location and Direction Coding(LDC). First of all, we design a finger vein imaging device with near-infrared(NIR) light source, by which a database for finger vein images is established. Subsequently, we make use of the brightness difference in the finger vein image to extract the vein pattern. Furthermore, finger vein LDC is proposed and performed, which creates a structured feature image for each finger vein. Finally, the structured feature image is utilized to conduct the personal identification on our image database for finger vein, which includes 440 vein images from 220 different fingers. The equal error rate of our method for this database is 0.44%.

Finger-vein recognition has received increased attention recently. However, the finger-vein images are always captured in poor quality. This certainly makes finger-vein feature representation unreliable, and further impairs the accuracy of finger-vein recognition. In this paper, we first give an analysis of the intrinsic factors causing finger-vein image degradation, and then propose a simple but effective image restoration method based on scattering removal. To give a proper description of finger-vein image degradation, a biological optical model (BOM) specific to finger-vein imaging is proposed according to the principle of light propagation in biological tissues. Based on BOM, the light scattering component is sensibly estimated and properly removed for finger-vein image restoration. Finally, experimental results demonstrate that the proposed method is powerful in enhancing the finger-vein image contrast and in improving the finger-vein image matching accuracy.

Trilateral Filterative Hermitian feature transformed deep perceptive fuzzy neural network for finger vein verification

Exploring finger vein based personal authentication for secure IoT

Finger Vein Recognition System (FVRS) is a biometric technology that identifies or verifies an individual based on unique vein patterns. Region of interest (ROI) extraction is one of the essential steps in FVRS. Current ROI extraction methods cannot extract an accurate ROI for a finger vein image with non-uniform background illumination. In this paper, we propose a new ROI extraction method that is immune to non-uniform background illumination. To detect the edge of the finger for the ROI extraction, we utilise watershed segmentation with distance transform and Canny edge detector. Experimental results show that the proposed ROI extraction method can extract consistent ROI from a finger vein image with non-uniform background illumination.

Finger vein images are rich in orientation and edge features. Inspired by the edge histogram descriptor proposed in MPEG-7, this paper presents an efficient orientation-based local descriptor, named histogram of salient edge orientation map (HSEOM). HSEOM is based on the fact that human vision is sensitive to edge features for image perception. For a given image, HSEOM first finds oriented edge maps according to predefined orientations using a well-known edge operator and obtains a salient edge orientation map by choosing an orientation with the maximum edge magnitude for each pixel. Then, subhistograms of the salient edge orientation map are generated from the nonoverlapping submaps and concatenated to build the final HSEOM. In the experiment of this paper, eight oriented edge maps were used to generate a salient edge orientation map for HSEOM construction. Experimental results on our available finger vein image database, MMCBNU_6000, show that the performance of HSEOM outperforms that of state-of-the-art orientation-based methods (e.g., Gabor filter, histogram of oriented gradients, and local directional code). Furthermore, the proposed HSEOM has advantages of low feature dimensionality and fast implementation for a real-time finger vein recognition system.

A finger vein image quality assessment method which uses object and human visual system index is proposed. This is based on both the human visual characteristics and finger vein image characteristics captured by the use of contactless and near infrared rays. We present an HSNR (signal to Noise ratio based on human visual system) finger vein image quality evaluation index by simulating the human visual system, and integrating the HSNR with effective area score, clarity score, finger shifting score and contrast score to obtain the total image quality score of the finger vein image. Experimental results demonstrate that the proposed method is consistent with subjective assessment by humans, and thus can be used to describe the visual perception of the image effectively.

In this study, the authors propose a multimodal biometric system by combining the finger knuckle and finger vein images at feature‐level fusion using fractional firefly (FFF) optimisation. Biometric characteristics, like finger knuckle and finger vein are unique and secure. Initially, the features are extracted from the finger knuckle and finger vein images using repeated line tracking method. Then, a newly developed method of feature‐level fusion using FFF optimisation is used. This method is utilised to find out the optimal weight score to fuse the extracted feature sets of finger knuckle and finger vein images. Thus, the recognition is carried out by the fused feature set using layered k‐SVM (k‐support vector machine) which is newly developed by combining the layered SVM classifier and k‐neural network classifier. The experimental results are evaluated and the performance is analysed with false acceptance ratio, false rejection ratio and accuracy. The outcome of the proposed FFF optimisation system obtains a higher accuracy of 96%.

Finger vein recognition has become a research hotspot in the field of biometrics due to its advantages of non-contact acquisition, unique information, and difficulty in terms of forging or pirating. However, in the real-world application process, the extraction of image features for the biometric remains a significant challenge when the captured finger vein images suffer from blur, noise, or missing feature information. To address the above challenges, we propose a novel deep reinforcement learning-based finger vein image recovery method, DRL-FVRestore, which trained an agent that adaptively selects the appropriate restoration behavior according to the state of the finger vein image, enabling continuous restoration of the image. The behaviors of image restoration are divided into three tasks: deblurring restoration, defect restoration, and denoising and enhancement restoration. Specifically, a DeblurGAN-v2 based on the Inception-Resnet-v2 backbone is proposed to achieve deblurring restoration of finger vein images. A finger vein feature-guided restoration network is proposed to achieve defect image restoration. The DRL-FVRestore is proposed to deal with multi-image problems in complex situations. In this paper, extensive experimental results are conducted based on using four publicly accessible datasets. The experimental results show that for restoration with single image problems, the EER values of the deblurring network and damage restoration network are reduced by an average of 4.31% and 1.71%, respectively, compared to other methods. For images with multiple vision problems, the EER value of the proposed DRL-FVRestore is reduced by an average of 3.98%.

Finger vein recognition is a promising biometric authentication technique that depends on the unique features of vein patterns in the finger for recognition. The existing finger vein recognition methods are based on minutiae features or binary features such as LBP, LLBP, PBBM etc. or from the entire vein pattern. However, the minutiae-based features cannot accurately represent the structural or anatomical aspects of the vein pattern. These issues with the minutia feature led to increased false matches. Recognition based on binary features have limitations such as increased false matches, sensitivity to the translation and rotation, security and privacy issues etc. A feature representation based on the anatomy of vein patterns can be an alternative solution to improve the recognition performance. In the IJCB 2020 conference, we showed that every finger vein image contains one or more of a kind of 4 special vein patterns which we refereed as Fork, Eye, Bridge, and Arch (FEBA). In this paper, we further enlarge this set to 6 vein patterns (F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> EB <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> B <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> A) by identifying two variations in the Fork and Bridge vein patterns. Based on 6 anatomical features of the possible 6 vein patterns in a vein image, we define a 6 × 6 feature matrix representation for finger vein images. Since this feature representation is based on the anatomical properties of the local vein patterns, it provides template security. Further we show that, the proposed feature representation is invariant to scaling, translation, and rotation changes. The experimental results using two open datasets and an in-house dataset show that the proposed method has a better recognition performance when compared to the existing approaches with an EER around 0.02% and an average recognition accuracy of 98%.

Compared with single biometric recognition, multimodal biometric recognition based on fingerprint and finger vein has been widely considered because of its convenient sample collection, high security and accurate recognition. However, according to our investigation, there is no public dataset of fingerprint and finger vein collected at the same time. The existing work uses fingerprint datasets and finger vein datasets from different sources for research, besides the researchers data from building their own equipment, which lacks consideration of practical applications. This is not conducive to the promotion of multibiometric technology based on finger. To promote research on multimodal biometric recognition based on fingerprint and finger vein, we design a finger collection device and introduce a new dataset, NUPT-FPV. It is the first public dataset to collect fingerprint and finger vein simultaneously in real-world applications. NUPT-FPV obtained 840 finger information from 140 volunteers, each finger was collected 20 times (collected in two sessions), and 33600 fingerprint and finger vein images were obtained. In addition, we propose a novel multimodal fusion method based on a convolutional neural network as a benchmark. Extensive experiments were conducted to verify the necessity of our dataset. Through the released dataset and benchmark, we hope to further promote the development of multimodal biometrics based on fingerprint and finger vein.

The modern-day finger vein based human recognition techniques provide good performance, yet they are highly finger vein image quality dependent. To address this problem, a novel deep learning-based approach using convolution-neural-network (CNN) for finger vein identification has been introduced here. The prime objective of our work is to achieve a stable response with accurate performance keeping varying quality finger vein images in account. The proposed approach is tested on the considered publicly available dataset and reported experiment results show that with effective training and testing strategy high identification accuracy can be achieved.

Low-quality finger vein images seriously affect the performance of the recognition system. Aiming at filtering high- and low-quality finger vein images, this paper presents a finger-vein image quality assessment method based on light-CNN (convolutional neural network). The previous quality assessment method aims to establish a model that is consistent with the evaluation effect of the human visual system. In contrast, our model further evaluates the quality of the finger vein image for the purpose of distinguishing whether the image contains rich and stable vein characteristics. The rich and stable vein characteristics in finger vein images play a more important role in practical applications. First, the finger vein image is automatically annotated based on a traditional finger vein image quality assessment method; secondly, the finger vein image is cut into image blocks to expand the training set, and the image block instead of the entire image is input to the CNN; then, the network and its variants are trained; in the final test, the average quality score of multiple image blocks from an image is the final quality score of the image. The experimental results show that the proposed light-CNN accurately identifies high and low quality images and significantly outperforms existing approaches in terms of distinguishing whether there are rich and stable vein characteristics in the image.

To carry out the finger vein recognition effectively and quickly, according to the characteristics of horizontal and vertical two-dimensional linear discriminant analysis ((2D)2LDA) reducing the dimensions, an algorithm of finger vein recognition is proposed. The algorithm is two direction weighted (2D)2LDA ((W2D)2LDA) based on preprocessing image of the figure vein image. Firstly, the original finger vein image is pre-processed. Secondly, the normalized vein image is gotten. Finally, the algorithm of this paper is used to be recognized. The effect of the rate of cumulate eigenvalue for 2DLDA and (2D)2LDA is analyzed. The effect of the weighted value and the rate of cumulate eigenvalue on W2DLDA, W(2D)2LDA and (W2D)2LDA are analyzed as well. Experimental results on our database of finger vein images show that the method presented achieves high recognition accuracy. The redundant information of eigenvectors extracted by (2D)2LDA is restrained strongly, and the bi-direction weighted effect is better than the one direction weighted effect. The average recognition rate of (W2D)2LDA is higher than 2DLDA, W2DLDA, (2D)2LDA and W(2D)2LDA.