Improving Biometric Identification Performance Using PCANet Deep Learning and Multispectral Palmprint

Meraoumia, Abdallah; Farid, Kadri; Bendjenna, Hakim; Chitroub, Salim; Bouridane, Ahmed

doi:10.1007/978-3-319-47301-7_3

Cited by 38 publications

(34 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have analyzed and compared these rates. The methods studied are the methods of [34], [54], [55], [56] which use the multispectral palmprint CASIA database. The best performance results of these methods are found 95% in terms of Classification Accuracy (ACC) for identification and between [0.02; 3.12] % in terms of EER for the verification mode.…”

Section: Comparison and Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Improving Palmprint based Biometric System Performance using Novel Multispectral Image Fusion Scheme

Thamri¹,

Aloui²,

Naceur³

2020

IJACSA

View full text Add to dashboard Cite

Nowadays, there are several identification systems which are based on different biometric modalities. In particular, multispectral images of palmprints captured in different spectral bands have a very distinctive biometric identifier. This paper proposes a novel fusion scheme of a biometric recognition system by multiSpectral palmprint. This system is composed of three blocks: (1) extraction of the region of interest (ROI) from multispectral images, (2) a new image fusion architecture based on the measurement of decorrelation, and (3) a scheme of dimension reduction and classification. The proposed image fusion system combines the information from the same left and right spectral band using the 2D discrete wavelet (DWT) transform technique. In addition, a feature extraction using the Log-Gabor transform is performed, while the feature size has been reduced using the Kernel Principal Component Analysis technique (KPCA). In Our experiments we use CASIA multispectral palmprint database. We obtained an accuracy rate (ACC) of 99.50% for the spectral bands WHT (white light) and 940 nm and an equal error rate EER = 0.05%.These results show that our system is robust against spoofing.

show abstract

Section: Comparison and Discussionmentioning

confidence: 99%

“…EER rates for this method are 0.16% and 0.73% for two databases. In [56], the method seeks to represent a biometric identification system utilizing PCANet 2 deep learning. This method is done by the 4 classifiers (RFT, SVM, KNN and RBF).…”

Section: Comparison and Discussionmentioning

confidence: 99%

Improving Palmprint based Biometric System Performance using Novel Multispectral Image Fusion Scheme

Thamri¹,

Aloui²,

Naceur³

2020

IJACSA

View full text Add to dashboard Cite

show abstract

“…It is worth mentioning that existing approaches such as those in [28] [66] did not use an optimization algorithm to automate the selection of the hyperparameters. As a result, an excessive amount of computation time was invested in randomly training multiple models in the hope of finding a fit model for palm vein authentication.…”

Section: Figure 14 Roc Curves For All Training/test Foldsmentioning

confidence: 99%

“…Moreover, feature extraction and selection methods are performed manually, which, in turn, can result in low performance if those features are not carefully picked out and, consequently, will not be very representative. Existing approaches that use deep CNN for palm vein verification and identification, including [27] [28] [29] [30], are very few and they employ fixed structures and training options for their CNN models. Moreover, studies that employ convolutional neural networks for similar applications with Bayesian optimization, such as [31] [32], use fixed structures or pretrained CNN models, and consider optimizing only the training options, such as learning rate and momentum, while ignoring the possibility of optimizing the network structure, such as the number of convolutional layers.…”

Section: Introductionmentioning

confidence: 99%

Contactless Palm Vein Authentication Using Deep Learning With Bayesian Optimization

2021

View full text Add to dashboard Cite

“…The neural network technology, as a kind of typical machine learning method, is proven to be a successful tool for artificial intelligence (AI). With the rapid development of computer hardware, deep neural network techniques also achieve a huge success in various kinds of recognition tasks [1, 2].…”

Section: Introductionmentioning

confidence: 99%

Hypercomplex extreme learning machine with its application in multispectral palmprint recognition

Zhang

2019

PLoS ONE

View full text Add to dashboard Cite

An extreme learning machine (ELM) is a novel training method for single-hidden layer feedforward neural networks (SLFNs) in which the hidden nodes are randomly assigned and fixed without iterative tuning. ELMs have earned widespread global interest due to their fast learning speed, satisfactory generalization ability and ease of implementation. In this paper, we extend this theory to hypercomplex space and attempt to simultaneously consider multisource information using a hypercomplex representation. To illustrate the performance of the proposed hypercomplex extreme learning machine (HELM), we have applied this scheme to the task of multispectral palmprint recognition. Images from different spectral bands are utilized to construct the hypercomplex space. Extensive experiments conducted on the PolyU and CASIA multispectral databases demonstrate that the HELM scheme can achieve competitive results. The source code together with datasets involved in this paper can be available for free download at https://figshare.com/s/01aef7d48840afab9d6d .

show abstract

Improving Biometric Identification Performance Using PCANet Deep Learning and Multispectral Palmprint

Cited by 38 publications

References 28 publications

Improving Palmprint based Biometric System Performance using Novel Multispectral Image Fusion Scheme

Improving Palmprint based Biometric System Performance using Novel Multispectral Image Fusion Scheme

Contactless Palm Vein Authentication Using Deep Learning With Bayesian Optimization

Hypercomplex extreme learning machine with its application in multispectral palmprint recognition

Contact Info

Product

Resources

About