Research on finger vein recognition based on capsule network

This paper propose a finger vein recognition algorithm based on the CapsNets(Capsule Network for short) to solve the problem of the information loss of the finger vein in the Convolution Neural Network(CNN). The CapsNets is transferred from the bottom to the high level in the form of capsule in the...

Full description

Bibliographic Details
Main Authors: Yu Chengbo, Xiong Dien
Format: Article
Language:zho
Published: National Computer System Engineering Research Institute of China 2018-10-01
Series:Dianzi Jishu Yingyong
Subjects:
Online Access:http://www.chinaaet.com/article/3000091554
_version_ 1818516220181217280
author Yu Chengbo
Xiong Dien
author_facet Yu Chengbo
Xiong Dien
author_sort Yu Chengbo
collection DOAJ
description This paper propose a finger vein recognition algorithm based on the CapsNets(Capsule Network for short) to solve the problem of the information loss of the finger vein in the Convolution Neural Network(CNN). The CapsNets is transferred from the bottom to the high level in the form of capsule in the whole learning process, so that the multidimensional characteristics of the finger vein are encapsulated in the form of vector, and the features will be preserved in the network, but not in the network after the loss is recovered. In this paper, 60 000 images are used as training set, and 10 000 images are used as test set. The experimental results show that the network structure features of CapsNets are more obvious than that of CNN, the accuracy of VGG is increased by 13.6%, and the value of loss converges to 0.01.
first_indexed 2024-12-11T00:39:11Z
format Article
id doaj.art-0e5bcf8a6ec24708b7bf3dcb84378bf6
institution Directory Open Access Journal
issn 0258-7998
language zho
last_indexed 2024-12-11T00:39:11Z
publishDate 2018-10-01
publisher National Computer System Engineering Research Institute of China
record_format Article
series Dianzi Jishu Yingyong
spelling doaj.art-0e5bcf8a6ec24708b7bf3dcb84378bf62022-12-22T01:27:00ZzhoNational Computer System Engineering Research Institute of ChinaDianzi Jishu Yingyong0258-79982018-10-014410151810.16157/j.issn.0258-7998.1822363000091554Research on finger vein recognition based on capsule networkYu Chengbo0Xiong Dien1School of Electrical and Electronic Engineering,Chongqing University of Techology,Chongqing 400050,ChinaSchool of Electrical and Electronic Engineering,Chongqing University of Techology,Chongqing 400050,ChinaThis paper propose a finger vein recognition algorithm based on the CapsNets(Capsule Network for short) to solve the problem of the information loss of the finger vein in the Convolution Neural Network(CNN). The CapsNets is transferred from the bottom to the high level in the form of capsule in the whole learning process, so that the multidimensional characteristics of the finger vein are encapsulated in the form of vector, and the features will be preserved in the network, but not in the network after the loss is recovered. In this paper, 60 000 images are used as training set, and 10 000 images are used as test set. The experimental results show that the network structure features of CapsNets are more obvious than that of CNN, the accuracy of VGG is increased by 13.6%, and the value of loss converges to 0.01.http://www.chinaaet.com/article/3000091554capsnetsfinger vein recognitiondeep learningcnn
spellingShingle Yu Chengbo
Xiong Dien
Research on finger vein recognition based on capsule network
Dianzi Jishu Yingyong
capsnets
finger vein recognition
deep learning
cnn
title Research on finger vein recognition based on capsule network
title_full Research on finger vein recognition based on capsule network
title_fullStr Research on finger vein recognition based on capsule network
title_full_unstemmed Research on finger vein recognition based on capsule network
title_short Research on finger vein recognition based on capsule network
title_sort research on finger vein recognition based on capsule network
topic capsnets
finger vein recognition
deep learning
cnn
url http://www.chinaaet.com/article/3000091554
work_keys_str_mv AT yuchengbo researchonfingerveinrecognitionbasedoncapsulenetwork
AT xiongdien researchonfingerveinrecognitionbasedoncapsulenetwork