Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN)

Standard

Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN). / He, Jing; Shen, Lei; Yao, Yudong; Wang, Huaxia; Zhao, Guodong; Gu, Xiaowei; Ding, Weiping.

In: IEEE Transactions on Emerging Topics in Computational Intelligence, 23.09.2021.

Research output: Contribution to journalArticlepeer-review

Harvard

He, J, Shen, L, Yao, Y, Wang, H, Zhao, G, Gu, X & Ding, W 2021, 'Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN)', IEEE Transactions on Emerging Topics in Computational Intelligence. https://doi.org/10.1109/TETCI.2021.3097734

APA

He, J., Shen, L., Yao, Y., Wang, H., Zhao, G., Gu, X., & Ding, W. (2021). Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN). IEEE Transactions on Emerging Topics in Computational Intelligence. https://doi.org/10.1109/TETCI.2021.3097734

Vancouver

He J, Shen L, Yao Y, Wang H, Zhao G, Gu X et al. Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN). IEEE Transactions on Emerging Topics in Computational Intelligence. 2021 Sep 23. https://doi.org/10.1109/TETCI.2021.3097734

Author

He, Jing ; Shen, Lei ; Yao, Yudong ; Wang, Huaxia ; Zhao, Guodong ; Gu, Xiaowei ; Ding, Weiping. / Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN). In: IEEE Transactions on Emerging Topics in Computational Intelligence. 2021.

Bibtex - Download

@article{fe7113c6b0414ed397891098a54a846b,
title = "Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN)",
abstract = "Vein contraction and venous compression typically caused by low temperature and excessive placement pressure can blur the captured finger vein images and severely impair the quality of extracted features. To improve the quality of captured finger vein image, this paper proposes a 26-layer generator network constrained by Neighbors-based Binary Patterns (NBP) texture loss to recover the clear image (guessing the original clear image). Firstly, by analyzing various types and degrees of blurred finger vein images captured in real application scenarios, a method to mathematically model the local and global blurriness using a pair of defocused and mean blur kernels is proposed. By iteratively and alternatively convoluting clear images with both kernels in a multi-scale window, a polymorphic blur training set is constructed for network training. Then, NBP texture loss is used for training the generator to enhance the deblurring ability of the network on images. Lastly, a novel network structure is proposed to retain more vein texture feature information, and two residual connections are added on both sides of the residual module of the 26-layer generator network to prevent degradation and overfitting. Theoretical analysis and simulation results show that the proposed neighbors-based binary-GAN (NB-GAN) can achieve better deblurring performance than the the-state-of-the-art approaches.",
keywords = "Convolution, Finger vein, GAN, Generators, image deblurring, Image restoration, Kernel, Mathematical models, texture loss, Training, Veins",
author = "Jing He and Lei Shen and Yudong Yao and Huaxia Wang and Guodong Zhao and Xiaowei Gu and Weiping Ding",
note = "Publisher Copyright: IEEE",
year = "2021",
month = sep,
day = "23",
doi = "10.1109/TETCI.2021.3097734",
language = "English",
journal = "IEEE Transactions on Emerging Topics in Computational Intelligence",
issn = "2471-285X",
publisher = "IEEE Press",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Finger Vein Image Deblurring Using Neighbors-Based Binary-GAN (NB-GAN)

AU - He, Jing

AU - Shen, Lei

AU - Yao, Yudong

AU - Wang, Huaxia

AU - Zhao, Guodong

AU - Gu, Xiaowei

AU - Ding, Weiping

N1 - Publisher Copyright: IEEE

PY - 2021/9/23

Y1 - 2021/9/23

N2 - Vein contraction and venous compression typically caused by low temperature and excessive placement pressure can blur the captured finger vein images and severely impair the quality of extracted features. To improve the quality of captured finger vein image, this paper proposes a 26-layer generator network constrained by Neighbors-based Binary Patterns (NBP) texture loss to recover the clear image (guessing the original clear image). Firstly, by analyzing various types and degrees of blurred finger vein images captured in real application scenarios, a method to mathematically model the local and global blurriness using a pair of defocused and mean blur kernels is proposed. By iteratively and alternatively convoluting clear images with both kernels in a multi-scale window, a polymorphic blur training set is constructed for network training. Then, NBP texture loss is used for training the generator to enhance the deblurring ability of the network on images. Lastly, a novel network structure is proposed to retain more vein texture feature information, and two residual connections are added on both sides of the residual module of the 26-layer generator network to prevent degradation and overfitting. Theoretical analysis and simulation results show that the proposed neighbors-based binary-GAN (NB-GAN) can achieve better deblurring performance than the the-state-of-the-art approaches.

AB - Vein contraction and venous compression typically caused by low temperature and excessive placement pressure can blur the captured finger vein images and severely impair the quality of extracted features. To improve the quality of captured finger vein image, this paper proposes a 26-layer generator network constrained by Neighbors-based Binary Patterns (NBP) texture loss to recover the clear image (guessing the original clear image). Firstly, by analyzing various types and degrees of blurred finger vein images captured in real application scenarios, a method to mathematically model the local and global blurriness using a pair of defocused and mean blur kernels is proposed. By iteratively and alternatively convoluting clear images with both kernels in a multi-scale window, a polymorphic blur training set is constructed for network training. Then, NBP texture loss is used for training the generator to enhance the deblurring ability of the network on images. Lastly, a novel network structure is proposed to retain more vein texture feature information, and two residual connections are added on both sides of the residual module of the 26-layer generator network to prevent degradation and overfitting. Theoretical analysis and simulation results show that the proposed neighbors-based binary-GAN (NB-GAN) can achieve better deblurring performance than the the-state-of-the-art approaches.

KW - Convolution

KW - Finger vein

KW - GAN

KW - Generators

KW - image deblurring

KW - Image restoration

KW - Kernel

KW - Mathematical models

KW - texture loss

KW - Training

KW - Veins

UR - http://www.scopus.com/inward/record.url?scp=85115814601&partnerID=8YFLogxK

U2 - 10.1109/TETCI.2021.3097734

DO - 10.1109/TETCI.2021.3097734

M3 - Article

AN - SCOPUS:85115814601

JO - IEEE Transactions on Emerging Topics in Computational Intelligence

JF - IEEE Transactions on Emerging Topics in Computational Intelligence

SN - 2471-285X

ER -

Show download statistics
View graph of relations
Citation formats