Integration of discriminative features and similarity-preserving encoding for finger vein image retrieval

Wang, Kuikui; Yang, Gongping; Yang, Geng; Yin, Yilong

doi:10.1109/icip.2017.8296938

Cited by 8 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, Fei et al utilized Euclidean and Hamming distance to evaluate the similarities of feature vector, respectively. In addition, Wang et al [26] divided a finger-vein image into 96 patches and each patch encoded as a 19-dimensional decimal vector, then the whole 1824-dimensional decimal feature was compacted and encoded as a 128-dimensional binary feature. Based on the one-to-many matching scheme above, we calculate the false acceptance rate (FAR) and the false rejection rate (FRR), and by adjusting the recognition threshold we draw the ROC curves of various approaches, as shown in Fig.…”

Section: ) Test Resultsmentioning

confidence: 99%

“…Local binary pattern (LBP) [20], local line binary pattern (LLBP) [21] and pyramid histogram of double competitive pattern (PHDCP) [22] encoded the vein pattern using local statistical information. Relying on the bifurcation and termination of veins, various methods [23]- [26] were proposed to extract such key points with different feature descriptor. Furthermore, the scale invariant feature transform (SIFT) [27], [28] was employed to acquire minute features that are more robust to rotation and translation.…”

Section: A Related Workmentioning

confidence: 99%

See 1 more Smart Citation

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

Zhang

et al. 2019

IEEE Access

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

Section: ) Test Resultsmentioning

confidence: 99%

Section: A Related Workmentioning

confidence: 99%

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

Zhang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To learn a feature mapping relationship to enhance the recognition ability of local features, a discriminative binary descriptor (DBD) [27] method is proposed. Recently, there are some classical methods of encoding and extracting features, such as k-means hashing-based method (KMHM) [28], iterative quantization-based method (ITQM) [29], weighted vein code indexing [30], anatomy structure analysis-based vein extraction (ASVAE) [31].…”

Section: Related Work a Finger Vein Recognition Methodsmentioning

confidence: 99%

“…This proves that our method can update the parameters of Gabor function better and extract the main vein information of finger vein image. And our method has the best performance compared with other latest methods and other classical methods, such as k-means hashingbased method (KMHM) [28], iterative quantization-based method (ITQM) [29], Deep representation-based feature extraction [36], Repeated line tracking, Maximum curvature, LLBP, ELBP. However, compared with Combining primary and soft biometric traits [57], Weighted Vein code indexing [30], our method performs worse on SDUMAL and FV-USM datasets.…”

Section: Comparison With the State-of-the-art Algorithmsmentioning

confidence: 96%

Adaptive Learning Gabor Filter for Finger-Vein Recognition

Zhang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Presently, finger-vein recognition is a new research direction in the field of biometric recognition. The Gabor filter has been extensively used for finger-vein recognition; however, its parameters are difficult to adjust. To solve this problem, an adaptive-learning Gabor filter is presented herein. We combine convolutional neural networks with a Gabor filter to calculate the gradient of the Gabor-filter parameters, based on the objective function, and to then optimize its parameters via back-propagation. The parameter θ of Gabor filter can be trained to the same angle as the vein texture of finger vein image. The parameter σ of Gabor filter has a certain relation with λ, and the parameter λ of Gabor filter can converge to the optimal value well. Using this method, we not only select appropriate and effective Gabor filter parameters to design the filter banks, we also consider the relationship between those parameters. Finally, we perform experiments on four public finger-vein datasets. Experimental results demonstrate that our method outperforms state-ofthe-art methods in finger-vein classification.

show abstract

A Local Descriptor with Physiological Characteristic for Finger Vein Recognition

Zhang

Ning

et al. 2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Integration of discriminative features and similarity-preserving encoding for finger vein image retrieval

Cited by 8 publications

References 17 publications

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

Adaptive Learning Gabor Filter for Finger-Vein Recognition

A Local Descriptor with Physiological Characteristic for Finger Vein Recognition

Contact Info

Product

Resources

About