Finger vein recognition

Wang, Kejun; Ma, Hui; Popoola, Oluwatoyin P.; Liu, Jingyu

doi:10.5772/18025

Cited by 23 publications

(8 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Later, Rosdi et al [9] proposed a new texture descriptor called the local line binary pattern, which shows better results than the LBP and LDP. To further improve the performance, Yang et al proposed the personalised best bit map [10], and personalised weight maps [11] which assign different weight values for different bits according to their stability. Minutiae-based methods: The minutiae point in finger vein verification refers to bifurcation points and endpoints.…”

Section: Pattern-based Methodsmentioning

confidence: 99%

Finger vein verification using a Siamese CNN

Zhou

Kang

et al. 2019

IET biom.

View full text Add to dashboard Cite

Finger vein verification has received more attention recently due to its unique advantages. However, most existing algorithms rely on handcrafted features, making them less robust to finger rotation and offsets. To alleviate these problems, the authors propose a novel method to extract more discriminative features from finger vein images. First, facing the issue of insufficient training data, they adopt a heavy image augmentation strategy and develop a pretrained-weights based convolutional neural network (CNN). Second, focusing on the characteristics of finger vein verification, they construct a Siamese structure combining with a modified contrastive loss function for training the above CNN, which effectively improves the network's performance. Finally, considering the feasibility of deploying the above CNN on embedded devices, they construct a lightweight CNN with depthwise separable convolution and adopt a knowledge distillation method to learn the knowledge from the pretrained-weights based CNN, which makes it small but effective. The experimental results show that the size of the lightweight CNN shrinks to 1/6th of the pretrained-weights based CNN, while its equal error rates achieved in the MMCBNU_6000, FV-USM and SDUMLA-HMT datasets are 0.08, 0.11 and 0.75% respectively, which nearly stays the same with the pretrained-weights based CNN and surpasses state-of-the-art methods.

show abstract

Section: Pattern-based Methodsmentioning

confidence: 99%

Finger vein verification using a Siamese CNN

Zhou

Kang

et al. 2019

IET biom.

View full text Add to dashboard Cite

show abstract

“…In these two techniques, the author followed a statistical approach which was computationally intensive. K. Wang et al [52] suggested a particular approach for matching finger vein images through utilizing relative distances and angles. The intersection points were derived from the image of the thinned finger vein by measuring the number of arms emanating from the pixels.…”

Section: Related Workmentioning

confidence: 99%

Feature extraction of finger-vein patterns based on boosting evolutionary algorithm and its application for loT identity and access management

Darwish

2021

Multimed Tools Appl

View full text Add to dashboard Cite

With billions of devices being connected, how to make sure that our information stays secure is becoming a hot topic of IoT. Traditional approaches to personal authentication are inadequate and ineffective in the IoT era. Finger vein technology is the current biometric system that utilizes the vein structure for recognition. As such patterns are veiled under the skin surface, they have significant privacy protection and are therefore incredibly difficult to forge. Finger vein recognition has gained a great deal of publicity because earlier approaches experienced significant pitfalls, such as its inability to handle the imbalanced collection of finger veins samples and the detection of distinguishing features in low-quality images. Such disadvantages have triggered a lack of consistency of the optimization algorithm or have contributed to a decrease in its efficiency. The key objective of the research discussed in this paper is to examine the impact of the genetic algorithm in the selection of the optimum vector characteristics of the finger vein. This is done by incorporating a Niching model in the form of a Context-Based Clearing (CBC) procedure to increase the heterogeneity of the features within the features' vector, with the goal of minimizing the association between them. It also offers the idea of a reduction of the feature set to reduce duplication without reducing accuracy. The performance study of the proposed model is carried out through multiple tests and the findings indicate an overall increase of 6% in the accuracy relative to some of the state-of-the-art finger vein recognition systems present in the literature.

show abstract

“…al. [5] proposed another method for matching finger vein images using relative distance and angles. The intersecting points were extracted from the thinned finger vein image by computing number of arms originating from a pixel.…”

Section: A Related Workmentioning

confidence: 99%

A novel finger vein feature extraction technique for authentication

Rajan¹,

Indu²

2014

2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)

View full text Add to dashboard Cite

Biometrics is mainly used for human identification using different physical traits. The traits that can be used as biometrics are face, palm print, voice, signature, gait etc. But use of these traits in biometrics is not perfectly reliable or secure. So, in order to overcome the security issue, a non-forgeable pattern has to be used. In terms of security and convenience, the fingervein is a promising biometric pattern for human identification. Since the vein images can be taken from live body only and these patterns being attributes present inside the human body, they are extremely complex to forge. In this paper, the finger vein images are enhanced by incorporating the concept of local histogram equalization, which improves the local contrast of an image. The features are extracted from the enhanced images using a combination of Frangi filter, FAST(Features from Accelerated Segment Test) algorithm and FREAK (Fast Retina Keypoint)descriptors .

show abstract

Finger vein recognition

Cited by 23 publications

References 7 publications

Finger vein verification using a Siamese CNN

Finger vein verification using a Siamese CNN

Feature extraction of finger-vein patterns based on boosting evolutionary algorithm and its application for loT identity and access management

A novel finger vein feature extraction technique for authentication

Contact Info

Product

Resources

About