Finger-Vein Image Enhancement Based on Pulse Coupled Neural Network

Lei, Lei; Xi, Feng; Chen, Shengyao

doi:10.1109/access.2019.2914229

Cited by 16 publications

(3 citation statements)

References 29 publications

(37 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finger-vein recognition technology is a burgeoning research field, and meanwhile it faces enormous challenges. Since the acquisition process is inherently affected by various factors such as uneven illumination [5], [6], light scattering inside finger tissues [7], [8] and ambient temperature [4], [9], a majority of finger-vein images inevitably contain blurred areas where venous and non-venous regions cannot be distinguished easily. Furthermore, since the acquisition system is mainly designed for non-contact, different finger postures can lead to displacement or deformation of acquired finger-vein images [10], [11].…”

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

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: 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

“…It has been widely used in image segmentation [3,4], edge detection [5,6], image denoising [7,8], image fusion [9,10], recognition and classification [11], noise filtering [12], etc. For example, Wang et al [13] proposed an improved pulse-coupled neural network based on multi-hybrid feature grey wolf optimizer for multimodal medical image segmentation; Liu et al [14] proposed a multi-focus image fusion method based on image texture using an improved PCNN model, which has good image fusion performance while preserving the original image information; Prieto et al [15] proposed a detection method based on pulse-coupled neural network to generate time matrix, which uses time matrix to extract the differences between image gray values to achieve edge segmentation; Zhang et al [16] proposed an adaptive genetic algorithm based on PCNN to suppress additive Gaussian white noise; Lei et al [17] proposed a finger vein image quality enhancement model based on pulse-coupled neural network, which further improves the reliability of image recognition; Deng et al [18] proposed an improved non-coupled PCNN model, which introduces the coupling effect of neighboring neurons into the dynamic threshold subsystem, making it easier for the network to obtain globally optimal segmentation. Due to the excellent synchronous pulse bursts characteristics and the capture ability, the PCNN model has been widely used in various fields of image processing.…”

Section: Introductionmentioning

confidence: 99%

Time Domain Characteristic Analysis of Non-coupled PCNN

Deng,

Yu,

Huang

2023

Preprint

View full text Add to dashboard Cite

The Pulse-Coupled Neural Network (PCNN) model is a multi-parameter neural network, the input-output characteristics are greatly affected by parameters setting. The traditional non-coupled PCNN model, which only uses the gray value as input, results in a certain inhibition of the network's pulse firing characteristics. In this paper, based on the traditional non-coupled PCNN model, from the perspective of system equations, the firing characteristics of the non-simplified uncoupled PCNN model with coupled linking term is studied, and the mathematical expressions for firing time and interval were summarized. By constructing different neighborhood linking weight matrix patterns, the impact of the linking weight matrix and coefficients on network characteristics was analyzed, and the constraint conditions for parameters aE, VE, and VF were provided. Finally, through experiments simulation, the correctness of the theoretical analysis is verified, providing theoretical support for researching on fully parameterized PCNN and the application of the studied model in image processing.

show abstract

“…Reference [22] proposed a convolutional neural network (CNN) to train and restore the vein patterns, which tackles the problem of vein pattern loss caused by overexposure or blurred region in the finger vein images. Reference [23] proposed a new model based on the pulse coupled neural network (PCNN) to enhance finger vein image quality and further to improve the reliability of image recognition. Reference [24] applied CNN to finger vein recognition and achieved better performance than traditional algorithms.…”

Section: Introductionmentioning

confidence: 99%

Finger Vein Image Inpainting With Gabor Texture Constraints

et al. 2020

View full text Add to dashboard Cite

The texture edge continuity of a finger vein image is very important for the accuracy of feature extraction. However, the traditional inpainting methods which, without accurate texture constraints, are easy to cause the vein texture of the inpainted image to be blurred and break. A finger vein image inpainting method with Gabor texture constraints is proposed. The proposed method effectively protects the texture edge continuity of the inpainted image. Firstly, using the proposed vertical phase difference coding method, the Gabor texture feature matrix of the finger vein image, which can accurately describe the texture information, can be extracted from the Gabor filtering responses. Then, according to the local texture continuity of the finger vein image, the known pixels, which have different texture orientations with the center pixel in the patch, are filtered out using the Gabor texture constraining mechanism during the inpainting process. The proposed method eliminates irrelevant information interference in the inpainting process and has a more precise texture propagation. Simulation experiments of artificially synthetic images and acquired images show that the finger vein images inpainted by the proposed method have better texture continuity and higher image quality than the traditional methods which do not have accurate texture constraints. The proposed method improves the recognition performance of the finger vein identification system with the acquired damaged images.INDEX TERMS Finger vein image, Gabor filter, image inpainting, texture feature, vertical phase difference coding.HUAXIA WANG (Member, IEEE) received the B.Eng. degree in information engineering from

show abstract

Finger-Vein Image Enhancement Based on Pulse Coupled Neural Network

Cited by 16 publications

References 29 publications

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

GAN-Based Image Augmentation for Finger-Vein Biometric Recognition

Time Domain Characteristic Analysis of Non-coupled PCNN

Finger Vein Image Inpainting With Gabor Texture Constraints

Contact Info

Product

Resources

About