An Improved Multispectral Palmprint System Using Deep CNN-based Palm-Features

Autor: Abdallah Meraoumia, Djamel Samai, Selma Trabelsi, Khaled Bensid, Abdelmalik Taleb-Ahmed
Přispěvatelé: Université de Ouargla, Laboratoire LAMIS, Université Larbi Tébessi [Tebessa], Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 (IEMN), Centrale Lille-Institut supérieur de l'électronique et du numérique (ISEN)-Université de Valenciennes et du Hainaut-Cambrésis (UVHC)-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Université Polytechnique Hauts-de-France (UPHF), COMmunications NUMériques - IEMN (COMNUM - IEMN), Institut d’Électronique, de Microélectronique et de Nanotechnologie - Département Opto-Acousto-Électronique - UMR 8520 (IEMN-DOAE), Centrale Lille-Institut supérieur de l'électronique et du numérique (ISEN)-Université de Valenciennes et du Hainaut-Cambrésis (UVHC)-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Université Polytechnique Hauts-de-France (UPHF)-Centrale Lille-Institut supérieur de l'électronique et du numérique (ISEN)-Université de Valenciennes et du Hainaut-Cambrésis (UVHC)-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Université Polytechnique Hauts-de-France (UPHF)-Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 (IEMN), Centrale Lille-Institut supérieur de l'électronique et du numérique (ISEN)-Université de Valenciennes et du Hainaut-Cambrésis (UVHC)-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Université Polytechnique Hauts-de-France (UPHF)-Centrale Lille-Institut supérieur de l'électronique et du numérique (ISEN)-Université de Valenciennes et du Hainaut-Cambrésis (UVHC)-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Université Polytechnique Hauts-de-France (UPHF), A C K N O W L E D G M E N T :The authors are grateful to the anonymous referees for their valuable and helpful comments. This research has been carried out within the PRFU projects: Grant: A25N01UN300120190003 of the Departmentof electronics and telecommunication, University KasdiMerbah of Ouargla. Grant: A01L08UN120120180001 of the Department ofElectrical Engineering, University Laarbi Tebessi ofTebessa.The authors thank the staff of Laboratory of electrical engineering (LAGE), and LAMIS laboratory for helpful commentsand suggestions
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Biometrics
Computer science
Physiology
Multispectral image
Data_MISCELLANEOUS
ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION
Convolutional neural network
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Databases
[INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI]
[SPI]Engineering Sciences [physics]
ComputerApplications_MISCELLANEOUS
Machine learning
[INFO]Computer Science [cs]
Deep cnn
Modality (human–computer interaction)
business.industry
Deep learning
Pattern recognition
Sensor fusion
[SPI.TRON]Engineering Sciences [physics]/Electronics
Identification (information)
ComputingMethodologies_PATTERNRECOGNITION
Security
Feature extraction
Convolutional neural networks
Artificial intelligence
Biometrics (access control)
business
[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Zdroj: International Conference on Advanced Electrical Engineering (ICAEE 2019)
International Conference on Advanced Electrical Engineering (ICAEE 2019), Nov 2019, Algiers, Algeria. pp.1-6, ⟨10.1109/ICAEE47123.2019.9015074⟩
Popis: International audience; Due to security imperatives, biometrics has attracted a lot of attention in recent decades. Biometric recognition refers to the identification of individuals based on their physiological and/or behavioral traits. Among the various physiological traits, the palmprint modality, which contains rich biometric features, has become one of the essential features that prove their effectiveness in improving the biometric recognition system accuracy. In addition to the palmprint texture features, infrared light can capture the vein-net of the palm, an independent biometric trait called palm-vein. Fortunately, these two biometric modalities can be easily obtained with a multispectral device and thus used together to enhance the biometric system. In this paper, we attempt to extract deep biometric features using a Convolutional Neural Network (CNN) to develop an effective deep-learning based multispectral palmprint recognition system. The tests results of extensive experiments conducted on a large and public palmprint multispectral database show that the proposed scheme effectively improves recognition results, mainly when fusing spectral bands of biometric modality.
Databáze: OpenAIRE
Externí odkaz: