Finger vein recognition based on finger crease location

Lu, Zhiying; Ding, Shumeng; Yin, Jing

doi:10.1117/1.jei.25.4.043004

Cited by 12 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The energy modeling of vein image segmentation is completed by GMM modeling of foreground and background. The image after feature fusion can be transformed into a weighted graph [4,36] with two endpoints G = (V, E). Where V is the set of image pixels and endpoints (s, t); E is the set of edges, which includes the weighted similar edges of the pixels belonging to the foreground and background, and the penalty weights between the pixels and the neighboring edges.…”

Section: Finger Vein Image Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Research on Finger Vein Image Segmentation and Blood Sampling Point Location in Automatic Blood Collection

Yang

et al. 2020

Sensors

View full text Add to dashboard Cite

In the fingertip blood automatic sampling process, when the blood sampling point in the fingertip venous area, it will greatly increase the amount of bleeding without being squeezed. In order to accurately locate the blood sampling point in the venous area, we propose a new finger vein image segmentation approach basing on Gabor transform and Gaussian mixed model (GMM). Firstly, Gabor filter parameter can be set adaptively according to the differential excitation of image and we use the local binary pattern (LBP) to fuse the same-scale and multi-orientation Gabor features of the image. Then, finger vein image segmentation is achieved by Gabor-GMM system and optimized by the max flow min cut method which is based on the relative entropy of the foreground and the background. Finally, the blood sampling point can be localized with corner detection. The experimental results show that the proposed approach has significant performance in segmenting finger vein images which the average accuracy of segmentation images reach 91.6%.

show abstract

Section: Finger Vein Image Segmentationmentioning

confidence: 99%

“…Image segmentation methods have made great achievements in many fields, but some gaps still exist in finger vein segmentation [ 4 , 5 ]. Due to the poor contrast of finger vein images, current segmentation methods cannot effectively distinguish the vein from non-venous areas [ 6 , 7 ].…”

Section: Introductionmentioning

confidence: 99%

Research on Finger Vein Image Segmentation and Blood Sampling Point Location in Automatic Blood Collection

Yang

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…In other words, they focus on extracting the whole vein network accurately from the complex background. Contrast-based methods [12]- [14] were proposed to enhance the contrast between the vein network and background. Since the cross section of vein networks shows valley forms, different valley detected methods [15]- [18] were presented to directly detect the entire vein network using repeated line and curvature.…”

Section: A Related Workmentioning

confidence: 99%

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

Zhang

et al. 2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

Deep learning methods, and especially convolutional neural networks (CNNs), have made a considerable breakthrough in various fields of machine vision, basically by employing large-scale labeled databases. However, deep learning methods applied in finger-vein area are basically implemented on small-scale datasets, which are probably faced with challenges such as overfitting, susceptible to finger position, unstable performance on various datasets and son on. In this study, we present a lightweight and fully convolutional Generative Adversarial Network (GAN) architecture, which is named FCGAN, using preliminary batch normalization, and tightly-constrained loss function for implementing finger-vein image augmentation. In addition, we present a novel scheme FCGAN-CNN for finger-vein classification, which reveals that synthetic finger-vein images using FCGAN are capable of improving the property of CNN for finger-vein image classification. The experiment of sample augmentation shows that the training accuracy using FCGAN-augmented samples could go beyond 99%, which is higher than 96.34% obtained using only classic sample augmentation. Furthermore, the well-trained CNN is further evaluated on a totally different dataset, which indicates that the proposed scheme FCGAN-CNN is capable of improving the ability of CNN to extract deep features. We consider that the proposed method for sample augmentation could be extended to other biometric systems. INDEX TERMS Sample augmentation, convolutional neural networks, generative adversarial networks, finger-vein classification.

show abstract

“…Identification from the vein pattern; hand vein pattern [4], finger vein pattern [5], palm vein pattern [6] and wrist vein pattern [7]. Among the four different types of vein patterns, the hand-wrist vascular network structure provides a clear image because it is located very close to the outer skin and is densely located.…”

Section: Introductionmentioning

confidence: 99%

Subject Review: Hand Vascular Pattern Technology

Salah¹,

Khalaf²

2022

IJERAT

View full text Add to dashboard Cite

Vein recognition systems are a form of biometric recognition that can distinguish people according to their vascular structure. Identification from hand-wrist vein pattern is one of these systems. In this study, hand-wrist vein images taken from people using an infrared light source with a wavelength of 850 nm were segmented by passing through various image processing algorithms. Scale-independent feature transformation (SURF) method was used for key point extraction from segmented images. The features obtained by the SURF method are rotation, camera angle, ambient light intensity, etc. This method has been preferred because it is invariant against situations. In the identification process, the Euclidean distance method was used by making use of the extracted key points. The accuracy rate was determined as 97% as a result of the matching processes using the hand-wrist vein patterns in the database.

show abstract

Finger vein recognition based on finger crease location

Cited by 12 publications

References 0 publications

Research on Finger Vein Image Segmentation and Blood Sampling Point Location in Automatic Blood Collection

Research on Finger Vein Image Segmentation and Blood Sampling Point Location in Automatic Blood Collection

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

Subject Review: Hand Vascular Pattern Technology

Contact Info

Product

Resources

About