Feature Selection-Based Hierarchical Deep Network for Image Classification
In this paper, a novel hierarchical deep network is proposed to combine the deep convolutional neural network and the feature selection-based tree classifier efficiently for image classification. First, the concept ontology is built for organizing large-scale image classes hierarchically in a coarse...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8959205/ |
id |
doaj-95ef087863c74ab490d8e7cd539eae0b |
---|---|
record_format |
Article |
spelling |
doaj-95ef087863c74ab490d8e7cd539eae0b2021-03-30T03:05:42ZengIEEEIEEE Access2169-35362020-01-018154361544710.1109/ACCESS.2020.29666518959205Feature Selection-Based Hierarchical Deep Network for Image ClassificationGuiqing He0Jiaqi Ji1https://orcid.org/0000-0003-2693-1172Haixi Zhang2Yuelei Xu3Jianping Fan4School of Electronics and Information, Northwestern Polytechnical University, Xi’an, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an, ChinaUnmanned System Research Institute, Northwestern Polytechnical University, Xi’an, ChinaDepartment of Computer Science, University of North Carolina at Charlotte, Charlotte, NC, USAIn this paper, a novel hierarchical deep network is proposed to combine the deep convolutional neural network and the feature selection-based tree classifier efficiently for image classification. First, the concept ontology is built for organizing large-scale image classes hierarchically in a coarse-to-fine fashion. Second, a novel selective orthogonal algorithm is proposed to make sure deep features extracted for each level classifiers more in line with the requirements of different classification tasks. Also, the role of useful feature components in multi-level deep features are improved. The experimental results on three datasets show that adding a feature selection module in a hierarchical deep network can perform better performance in large-scale image classification.https://ieeexplore.ieee.org/document/8959205/Feature selectionmulti-level tree classifiersimage classificationselective orthogonal |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Guiqing He Jiaqi Ji Haixi Zhang Yuelei Xu Jianping Fan |
spellingShingle |
Guiqing He Jiaqi Ji Haixi Zhang Yuelei Xu Jianping Fan Feature Selection-Based Hierarchical Deep Network for Image Classification IEEE Access Feature selection multi-level tree classifiers image classification selective orthogonal |
author_facet |
Guiqing He Jiaqi Ji Haixi Zhang Yuelei Xu Jianping Fan |
author_sort |
Guiqing He |
title |
Feature Selection-Based Hierarchical Deep Network for Image Classification |
title_short |
Feature Selection-Based Hierarchical Deep Network for Image Classification |
title_full |
Feature Selection-Based Hierarchical Deep Network for Image Classification |
title_fullStr |
Feature Selection-Based Hierarchical Deep Network for Image Classification |
title_full_unstemmed |
Feature Selection-Based Hierarchical Deep Network for Image Classification |
title_sort |
feature selection-based hierarchical deep network for image classification |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
In this paper, a novel hierarchical deep network is proposed to combine the deep convolutional neural network and the feature selection-based tree classifier efficiently for image classification. First, the concept ontology is built for organizing large-scale image classes hierarchically in a coarse-to-fine fashion. Second, a novel selective orthogonal algorithm is proposed to make sure deep features extracted for each level classifiers more in line with the requirements of different classification tasks. Also, the role of useful feature components in multi-level deep features are improved. The experimental results on three datasets show that adding a feature selection module in a hierarchical deep network can perform better performance in large-scale image classification. |
topic |
Feature selection multi-level tree classifiers image classification selective orthogonal |
url |
https://ieeexplore.ieee.org/document/8959205/ |
work_keys_str_mv |
AT guiqinghe featureselectionbasedhierarchicaldeepnetworkforimageclassification AT jiaqiji featureselectionbasedhierarchicaldeepnetworkforimageclassification AT haixizhang featureselectionbasedhierarchicaldeepnetworkforimageclassification AT yueleixu featureselectionbasedhierarchicaldeepnetworkforimageclassification AT jianpingfan featureselectionbasedhierarchicaldeepnetworkforimageclassification |
_version_ |
1724184108805914624 |