Loading…

A lightweight deep neural network with higher accuracy

To improve accuracy of the MobileNet network, a new lightweight deep neural network is designed based on the MobileNetV2 network. Firstly, it modifies the network depth of MobileNetV2 to balance the image resolution, network width and depth to keep the gradient stable, which reduces the generation o...

Full description

Saved in:
Bibliographic Details
Published in:PloS one 2022-08, Vol.17 (8), p.e0271225-e0271225
Main Authors: Zhao, Liquan, Wang, Leilei, Jia, Yanfei, Cui, Ying
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c669t-ad8223e67b29c39a09d2ac67f41118fec676198a5645a95ccf3867b0907a2ec93
cites cdi_FETCH-LOGICAL-c669t-ad8223e67b29c39a09d2ac67f41118fec676198a5645a95ccf3867b0907a2ec93
container_end_page e0271225
container_issue 8
container_start_page e0271225
container_title PloS one
container_volume 17
creator Zhao, Liquan
Wang, Leilei
Jia, Yanfei
Cui, Ying
description To improve accuracy of the MobileNet network, a new lightweight deep neural network is designed based on the MobileNetV2 network. Firstly, it modifies the network depth of MobileNetV2 to balance the image resolution, network width and depth to keep the gradient stable, which reduces the generation of gradient vanishing or gradient exploding. Secondly, it proposes an improved Bottleneck module by introducing channel attention mechanism. It assigns different weights for different channels according to the degree of relevance between the object features and channels. Therefore, the network can extract more effective features from a complex background. In the end, a new usage strategy of the improved Bottleneck is proposed. It uses the improved Bottleneck module in the second, fourth and fifth stages of MobileNetV2, and uses the original Bottleneck module in other states. Compared with MobileNetV2, MobileNetV3, ShuffleNetV2, GhostNet and HBONetmethods, the proposed method has the highest classification accuracy on the ImageNet-1K dataset, CIFAR-10 and CIFAR-100. Compared with YOLOV4-Lite methods based on these lightweight network networks, YOLOV4-Lite based on our proposed network also has the highest detection accuracy on the PASCAL VOC07+12 dataset.
doi_str_mv 10.1371/journal.pone.0271225
format article
fullrecord <record><control><sourceid>gale_plos_</sourceid><recordid>TN_cdi_plos_journals_2697325010</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><galeid>A712277937</galeid><doaj_id>oai_doaj_org_article_0f5d585b21a94732bf912fceffcc1c1c</doaj_id><sourcerecordid>A712277937</sourcerecordid><originalsourceid>FETCH-LOGICAL-c669t-ad8223e67b29c39a09d2ac67f41118fec676198a5645a95ccf3867b0907a2ec93</originalsourceid><addsrcrecordid>eNqNkm2L1DAQx4so3nn6DQQLguiLXfOwSZo3wnL4sHBw4NPbkKaTbdZusyap6317U7fKVe6FBJIw85t_ZiZTFE8xWmIq8OudH0Kvu-XB97BERGBC2L3iHEtKFpwgev_W_ax4FOMOIUYrzh8WZ5RJLCjG5wVfl53btukI4142AIeyhyHoLh_p6MO38uhSW7bZDaHUxmSfuXlcPLC6i_BkOi-KL-_efr78sLi6fr-5XF8tDOcyLXRTEUKBi5pIQ6VGsiHacGFXGOPKQr5yLCvN-IppyYyxOUFRI4mEJmAkvSienXQPnY9qKjkqwqWghCGMMrE5EY3XO3UIbq_DjfLaqd8GH7ZKh-RMBwpZ1rCK1QRrucrxtZWYWAPWGoPzylpvpteGeg-NgT7lRsxE557etWrrfyhJV4wymgVeTgLBfx8gJrV30UDX6R78MOXNK8lJRp__g95d3URtdS7A9dbnd80oqtbjjwshqcjU8g4qrwb2zuT5sC7bZwGvZgGZSfAzbfUQo9p8-vj_7PXXOfviFtuC7lIbfTck5_s4B1cn0AQfYwD7t8kYqXG8_3RDjeOtpvGmvwCJmunE</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2697325010</pqid></control><display><type>article</type><title>A lightweight deep neural network with higher accuracy</title><source>Publicly Available Content Database</source><source>PubMed Central</source><creator>Zhao, Liquan ; Wang, Leilei ; Jia, Yanfei ; Cui, Ying</creator><contributor>Saha, Sriparna</contributor><creatorcontrib>Zhao, Liquan ; Wang, Leilei ; Jia, Yanfei ; Cui, Ying ; Saha, Sriparna</creatorcontrib><description>To improve accuracy of the MobileNet network, a new lightweight deep neural network is designed based on the MobileNetV2 network. Firstly, it modifies the network depth of MobileNetV2 to balance the image resolution, network width and depth to keep the gradient stable, which reduces the generation of gradient vanishing or gradient exploding. Secondly, it proposes an improved Bottleneck module by introducing channel attention mechanism. It assigns different weights for different channels according to the degree of relevance between the object features and channels. Therefore, the network can extract more effective features from a complex background. In the end, a new usage strategy of the improved Bottleneck is proposed. It uses the improved Bottleneck module in the second, fourth and fifth stages of MobileNetV2, and uses the original Bottleneck module in other states. Compared with MobileNetV2, MobileNetV3, ShuffleNetV2, GhostNet and HBONetmethods, the proposed method has the highest classification accuracy on the ImageNet-1K dataset, CIFAR-10 and CIFAR-100. Compared with YOLOV4-Lite methods based on these lightweight network networks, YOLOV4-Lite based on our proposed network also has the highest detection accuracy on the PASCAL VOC07+12 dataset.</description><identifier>ISSN: 1932-6203</identifier><identifier>EISSN: 1932-6203</identifier><identifier>DOI: 10.1371/journal.pone.0271225</identifier><identifier>PMID: 35917311</identifier><language>eng</language><publisher>San Francisco: Public Library of Science</publisher><subject>Accuracy ; Analysis ; Artificial neural networks ; Biology and Life Sciences ; Channels ; Classification ; Computer and Information Sciences ; Datasets ; Engineering and Technology ; Fault diagnosis ; Feature extraction ; Image processing ; Image resolution ; Lightweight ; Modules ; Neural networks ; Power ; Research and Analysis Methods ; Social Sciences</subject><ispartof>PloS one, 2022-08, Vol.17 (8), p.e0271225-e0271225</ispartof><rights>COPYRIGHT 2022 Public Library of Science</rights><rights>2022 Zhao et al. This is an open access article distributed under the terms of the Creative Commons Attribution License: http://creativecommons.org/licenses/by/4.0/ (the “License”), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>2022 Zhao et al 2022 Zhao et al</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c669t-ad8223e67b29c39a09d2ac67f41118fec676198a5645a95ccf3867b0907a2ec93</citedby><cites>FETCH-LOGICAL-c669t-ad8223e67b29c39a09d2ac67f41118fec676198a5645a95ccf3867b0907a2ec93</cites><orcidid>0000-0002-9499-1911</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.proquest.com/docview/2697325010/fulltextPDF?pq-origsite=primo$$EPDF$$P50$$Gproquest$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.proquest.com/docview/2697325010?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>230,314,727,780,784,885,25753,27924,27925,37012,37013,44590,53791,53793,75126</link.rule.ids></links><search><contributor>Saha, Sriparna</contributor><creatorcontrib>Zhao, Liquan</creatorcontrib><creatorcontrib>Wang, Leilei</creatorcontrib><creatorcontrib>Jia, Yanfei</creatorcontrib><creatorcontrib>Cui, Ying</creatorcontrib><title>A lightweight deep neural network with higher accuracy</title><title>PloS one</title><description>To improve accuracy of the MobileNet network, a new lightweight deep neural network is designed based on the MobileNetV2 network. Firstly, it modifies the network depth of MobileNetV2 to balance the image resolution, network width and depth to keep the gradient stable, which reduces the generation of gradient vanishing or gradient exploding. Secondly, it proposes an improved Bottleneck module by introducing channel attention mechanism. It assigns different weights for different channels according to the degree of relevance between the object features and channels. Therefore, the network can extract more effective features from a complex background. In the end, a new usage strategy of the improved Bottleneck is proposed. It uses the improved Bottleneck module in the second, fourth and fifth stages of MobileNetV2, and uses the original Bottleneck module in other states. Compared with MobileNetV2, MobileNetV3, ShuffleNetV2, GhostNet and HBONetmethods, the proposed method has the highest classification accuracy on the ImageNet-1K dataset, CIFAR-10 and CIFAR-100. Compared with YOLOV4-Lite methods based on these lightweight network networks, YOLOV4-Lite based on our proposed network also has the highest detection accuracy on the PASCAL VOC07+12 dataset.</description><subject>Accuracy</subject><subject>Analysis</subject><subject>Artificial neural networks</subject><subject>Biology and Life Sciences</subject><subject>Channels</subject><subject>Classification</subject><subject>Computer and Information Sciences</subject><subject>Datasets</subject><subject>Engineering and Technology</subject><subject>Fault diagnosis</subject><subject>Feature extraction</subject><subject>Image processing</subject><subject>Image resolution</subject><subject>Lightweight</subject><subject>Modules</subject><subject>Neural networks</subject><subject>Power</subject><subject>Research and Analysis Methods</subject><subject>Social Sciences</subject><issn>1932-6203</issn><issn>1932-6203</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><sourceid>DOA</sourceid><recordid>eNqNkm2L1DAQx4so3nn6DQQLguiLXfOwSZo3wnL4sHBw4NPbkKaTbdZusyap6317U7fKVe6FBJIw85t_ZiZTFE8xWmIq8OudH0Kvu-XB97BERGBC2L3iHEtKFpwgev_W_ax4FOMOIUYrzh8WZ5RJLCjG5wVfl53btukI4142AIeyhyHoLh_p6MO38uhSW7bZDaHUxmSfuXlcPLC6i_BkOi-KL-_efr78sLi6fr-5XF8tDOcyLXRTEUKBi5pIQ6VGsiHacGFXGOPKQr5yLCvN-IppyYyxOUFRI4mEJmAkvSienXQPnY9qKjkqwqWghCGMMrE5EY3XO3UIbq_DjfLaqd8GH7ZKh-RMBwpZ1rCK1QRrucrxtZWYWAPWGoPzylpvpteGeg-NgT7lRsxE557etWrrfyhJV4wymgVeTgLBfx8gJrV30UDX6R78MOXNK8lJRp__g95d3URtdS7A9dbnd80oqtbjjwshqcjU8g4qrwb2zuT5sC7bZwGvZgGZSfAzbfUQo9p8-vj_7PXXOfviFtuC7lIbfTck5_s4B1cn0AQfYwD7t8kYqXG8_3RDjeOtpvGmvwCJmunE</recordid><startdate>20220802</startdate><enddate>20220802</enddate><creator>Zhao, Liquan</creator><creator>Wang, Leilei</creator><creator>Jia, Yanfei</creator><creator>Cui, Ying</creator><general>Public Library of Science</general><general>Public Library of Science (PLoS)</general><scope>AAYXX</scope><scope>CITATION</scope><scope>IOV</scope><scope>ISR</scope><scope>3V.</scope><scope>7QG</scope><scope>7QL</scope><scope>7QO</scope><scope>7RV</scope><scope>7SN</scope><scope>7SS</scope><scope>7T5</scope><scope>7TG</scope><scope>7TM</scope><scope>7U9</scope><scope>7X2</scope><scope>7X7</scope><scope>7XB</scope><scope>88E</scope><scope>8AO</scope><scope>8C1</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>8FH</scope><scope>8FI</scope><scope>8FJ</scope><scope>8FK</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>ATCPS</scope><scope>AZQEC</scope><scope>BBNVY</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>BHPHI</scope><scope>C1K</scope><scope>CCPQU</scope><scope>D1I</scope><scope>DWQXO</scope><scope>FR3</scope><scope>FYUFA</scope><scope>GHDGH</scope><scope>GNUQQ</scope><scope>H94</scope><scope>HCIFZ</scope><scope>K9.</scope><scope>KB.</scope><scope>KB0</scope><scope>KL.</scope><scope>L6V</scope><scope>LK8</scope><scope>M0K</scope><scope>M0S</scope><scope>M1P</scope><scope>M7N</scope><scope>M7P</scope><scope>M7S</scope><scope>NAPCQ</scope><scope>P5Z</scope><scope>P62</scope><scope>P64</scope><scope>PATMY</scope><scope>PDBOC</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PTHSS</scope><scope>PYCSY</scope><scope>RC3</scope><scope>7X8</scope><scope>5PM</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0002-9499-1911</orcidid></search><sort><creationdate>20220802</creationdate><title>A lightweight deep neural network with higher accuracy</title><author>Zhao, Liquan ; Wang, Leilei ; Jia, Yanfei ; Cui, Ying</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c669t-ad8223e67b29c39a09d2ac67f41118fec676198a5645a95ccf3867b0907a2ec93</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Accuracy</topic><topic>Analysis</topic><topic>Artificial neural networks</topic><topic>Biology and Life Sciences</topic><topic>Channels</topic><topic>Classification</topic><topic>Computer and Information Sciences</topic><topic>Datasets</topic><topic>Engineering and Technology</topic><topic>Fault diagnosis</topic><topic>Feature extraction</topic><topic>Image processing</topic><topic>Image resolution</topic><topic>Lightweight</topic><topic>Modules</topic><topic>Neural networks</topic><topic>Power</topic><topic>Research and Analysis Methods</topic><topic>Social Sciences</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zhao, Liquan</creatorcontrib><creatorcontrib>Wang, Leilei</creatorcontrib><creatorcontrib>Jia, Yanfei</creatorcontrib><creatorcontrib>Cui, Ying</creatorcontrib><collection>CrossRef</collection><collection>Gale In Context: Opposing Viewpoints</collection><collection>Gale In Context: Science</collection><collection>ProQuest Central (Corporate)</collection><collection>Animal Behavior Abstracts</collection><collection>Bacteriology Abstracts (Microbiology B)</collection><collection>Biotechnology Research Abstracts</collection><collection>Proquest Nursing &amp; Allied Health Source</collection><collection>Ecology Abstracts</collection><collection>Entomology Abstracts (Full archive)</collection><collection>Immunology Abstracts</collection><collection>Meteorological &amp; Geoastrophysical Abstracts</collection><collection>Nucleic Acids Abstracts</collection><collection>Virology and AIDS Abstracts</collection><collection>Agricultural Science Collection</collection><collection>Health &amp; Medical Collection</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>Medical Database (Alumni Edition)</collection><collection>ProQuest Pharma Collection</collection><collection>Public Health Database</collection><collection>Technology Research Database</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Natural Science Collection</collection><collection>Hospital Premium Collection</collection><collection>Hospital Premium Collection (Alumni Edition)</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>Advanced Technologies &amp; Aerospace Collection</collection><collection>Agricultural &amp; Environmental Science Collection</collection><collection>ProQuest Central Essentials</collection><collection>Biological Science Collection</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>Natural Science Collection</collection><collection>Environmental Sciences and Pollution Management</collection><collection>ProQuest One Community College</collection><collection>ProQuest Materials Science Collection</collection><collection>ProQuest Central Korea</collection><collection>Engineering Research Database</collection><collection>Health Research Premium Collection</collection><collection>Health Research Premium Collection (Alumni)</collection><collection>ProQuest Central Student</collection><collection>AIDS and Cancer Research Abstracts</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Health &amp; Medical Complete (Alumni)</collection><collection>Materials Science Database</collection><collection>Nursing &amp; Allied Health Database (Alumni Edition)</collection><collection>Meteorological &amp; Geoastrophysical Abstracts - Academic</collection><collection>ProQuest Engineering Collection</collection><collection>ProQuest Biological Science Collection</collection><collection>Agricultural Science Database</collection><collection>Health &amp; Medical Collection (Alumni Edition)</collection><collection>Medical Database</collection><collection>Algology Mycology and Protozoology Abstracts (Microbiology C)</collection><collection>Biological Science Database</collection><collection>Engineering Database</collection><collection>Nursing &amp; Allied Health Premium</collection><collection>Advanced Technologies &amp; Aerospace Database</collection><collection>ProQuest Advanced Technologies &amp; Aerospace Collection</collection><collection>Biotechnology and BioEngineering Abstracts</collection><collection>Environmental Science Database</collection><collection>Materials Science Collection</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>Engineering Collection</collection><collection>Environmental Science Collection</collection><collection>Genetics Abstracts</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><collection>DOAJ Directory of Open Access Journals</collection><jtitle>PloS one</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Zhao, Liquan</au><au>Wang, Leilei</au><au>Jia, Yanfei</au><au>Cui, Ying</au><au>Saha, Sriparna</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A lightweight deep neural network with higher accuracy</atitle><jtitle>PloS one</jtitle><date>2022-08-02</date><risdate>2022</risdate><volume>17</volume><issue>8</issue><spage>e0271225</spage><epage>e0271225</epage><pages>e0271225-e0271225</pages><issn>1932-6203</issn><eissn>1932-6203</eissn><abstract>To improve accuracy of the MobileNet network, a new lightweight deep neural network is designed based on the MobileNetV2 network. Firstly, it modifies the network depth of MobileNetV2 to balance the image resolution, network width and depth to keep the gradient stable, which reduces the generation of gradient vanishing or gradient exploding. Secondly, it proposes an improved Bottleneck module by introducing channel attention mechanism. It assigns different weights for different channels according to the degree of relevance between the object features and channels. Therefore, the network can extract more effective features from a complex background. In the end, a new usage strategy of the improved Bottleneck is proposed. It uses the improved Bottleneck module in the second, fourth and fifth stages of MobileNetV2, and uses the original Bottleneck module in other states. Compared with MobileNetV2, MobileNetV3, ShuffleNetV2, GhostNet and HBONetmethods, the proposed method has the highest classification accuracy on the ImageNet-1K dataset, CIFAR-10 and CIFAR-100. Compared with YOLOV4-Lite methods based on these lightweight network networks, YOLOV4-Lite based on our proposed network also has the highest detection accuracy on the PASCAL VOC07+12 dataset.</abstract><cop>San Francisco</cop><pub>Public Library of Science</pub><pmid>35917311</pmid><doi>10.1371/journal.pone.0271225</doi><tpages>e0271225</tpages><orcidid>https://orcid.org/0000-0002-9499-1911</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1932-6203
ispartof PloS one, 2022-08, Vol.17 (8), p.e0271225-e0271225
issn 1932-6203
1932-6203
language eng
recordid cdi_plos_journals_2697325010
source Publicly Available Content Database; PubMed Central
subjects Accuracy
Analysis
Artificial neural networks
Biology and Life Sciences
Channels
Classification
Computer and Information Sciences
Datasets
Engineering and Technology
Fault diagnosis
Feature extraction
Image processing
Image resolution
Lightweight
Modules
Neural networks
Power
Research and Analysis Methods
Social Sciences
title A lightweight deep neural network with higher accuracy
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-07T05%3A59%3A46IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-gale_plos_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20lightweight%20deep%20neural%20network%20with%20higher%20accuracy&rft.jtitle=PloS%20one&rft.au=Zhao,%20Liquan&rft.date=2022-08-02&rft.volume=17&rft.issue=8&rft.spage=e0271225&rft.epage=e0271225&rft.pages=e0271225-e0271225&rft.issn=1932-6203&rft.eissn=1932-6203&rft_id=info:doi/10.1371/journal.pone.0271225&rft_dat=%3Cgale_plos_%3EA712277937%3C/gale_plos_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c669t-ad8223e67b29c39a09d2ac67f41118fec676198a5645a95ccf3867b0907a2ec93%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2697325010&rft_id=info:pmid/35917311&rft_galeid=A712277937&rfr_iscdi=true