Loading…
Hierarchical Attention Siamese Network for Thermal Infrared Target Tracking
Thermal infrared (TIR) target tracking is an important topic in the computer vision area. The TIR images are not affected by ambient light and have strong environmental adaptability, making them widely used in battlefield perception, video surveillance, and assisted driving. However, TIR target trac...
Saved in:
Published in: | IEEE transactions on instrumentation and measurement 2024, Vol.73, p.1-11 |
---|---|
Main Authors: | , , , , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | |
---|---|
cites | cdi_FETCH-LOGICAL-c175t-1a093548775d48b2df7dd6d194294119cef69300dc89f982222b9f93d9a8fd373 |
container_end_page | 11 |
container_issue | |
container_start_page | 1 |
container_title | IEEE transactions on instrumentation and measurement |
container_volume | 73 |
creator | Yuan, Di Liao, Donghai Huang, Feng Qiu, Zhaobing Shu, Xiu Tian, Chunwei Liu, Qiao |
description | Thermal infrared (TIR) target tracking is an important topic in the computer vision area. The TIR images are not affected by ambient light and have strong environmental adaptability, making them widely used in battlefield perception, video surveillance, and assisted driving. However, TIR target tracking faces problems such as relatively insufficient information and lack of target texture information, which significantly affects the tracking accuracy of the TIR tracking methods. To solve the above problems, we propose a TIR target tracking method based on a Siamese network with a hierarchical attention mechanism (called: SiamHAN). Specifically, the CIoU Loss is introduced to make full use of the regression box information to calculate the loss function more accurately. The global context network (GCNet) attention mechanism is introduced to reconstruct the feature extraction structure of fine-grained information for the fine-grained information of TIR images. Meanwhile, for the feature information of the hierarchical backbone network of the Siamese network, the ECANet attention mechanism is used for hierarchical feature fusion, so that it can fully utilize the feature information of the multilayer backbone network to represent the target. On the LSOTB-TIR, the hierarchical attention Siamese network achieved a 2.9% increase in success rate and a 4.3% increase in precision relative to the baseline tracker. Experiments show that the proposed SiamHAN method has achieved competitive tracking results on the TIR testing datasets. |
doi_str_mv | 10.1109/TIM.2024.3462973 |
format | article |
fullrecord | <record><control><sourceid>proquest_ieee_</sourceid><recordid>TN_cdi_proquest_journals_3112186490</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>10683783</ieee_id><sourcerecordid>3112186490</sourcerecordid><originalsourceid>FETCH-LOGICAL-c175t-1a093548775d48b2df7dd6d194294119cef69300dc89f982222b9f93d9a8fd373</originalsourceid><addsrcrecordid>eNpNkDlPAzEQRi0EEiHQU1BYot7ga32UUQQk4ipYasuxx8nm2A3ejRD_HkdJwTTfFO-bkR5Ct5SMKCXmoZq9jRhhYsSFZEbxMzSgZakKIyU7RwNCqC6MKOUluuq6FSFESaEG6GVaQ3LJL2vvNnjc99D0ddvgz9ptoQP8Dv1Pm9Y4tglXS0jbTM2amCsQcOXSAnpcJefXdbO4RhfRbTq4OeUQfT09VpNp8frxPJuMXwtPVdkX1BHDS6GVKoPQcxaiCkEGagQzglLjIUrDCQlem2g0yzPPCw_G6Ri44kN0f7y7S-33Hrrertp9avJLyyllVEthSKbIkfKp7boE0e5SvXXp11JiD8psVmYPyuxJWa7cHSs1APzDpeZKc_4H_PJmqQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3112186490</pqid></control><display><type>article</type><title>Hierarchical Attention Siamese Network for Thermal Infrared Target Tracking</title><source>IEEE Electronic Library (IEL) Journals</source><creator>Yuan, Di ; Liao, Donghai ; Huang, Feng ; Qiu, Zhaobing ; Shu, Xiu ; Tian, Chunwei ; Liu, Qiao</creator><creatorcontrib>Yuan, Di ; Liao, Donghai ; Huang, Feng ; Qiu, Zhaobing ; Shu, Xiu ; Tian, Chunwei ; Liu, Qiao</creatorcontrib><description>Thermal infrared (TIR) target tracking is an important topic in the computer vision area. The TIR images are not affected by ambient light and have strong environmental adaptability, making them widely used in battlefield perception, video surveillance, and assisted driving. However, TIR target tracking faces problems such as relatively insufficient information and lack of target texture information, which significantly affects the tracking accuracy of the TIR tracking methods. To solve the above problems, we propose a TIR target tracking method based on a Siamese network with a hierarchical attention mechanism (called: SiamHAN). Specifically, the CIoU Loss is introduced to make full use of the regression box information to calculate the loss function more accurately. The global context network (GCNet) attention mechanism is introduced to reconstruct the feature extraction structure of fine-grained information for the fine-grained information of TIR images. Meanwhile, for the feature information of the hierarchical backbone network of the Siamese network, the ECANet attention mechanism is used for hierarchical feature fusion, so that it can fully utilize the feature information of the multilayer backbone network to represent the target. On the LSOTB-TIR, the hierarchical attention Siamese network achieved a 2.9% increase in success rate and a 4.3% increase in precision relative to the baseline tracker. Experiments show that the proposed SiamHAN method has achieved competitive tracking results on the TIR testing datasets.</description><identifier>ISSN: 0018-9456</identifier><identifier>EISSN: 1557-9662</identifier><identifier>DOI: 10.1109/TIM.2024.3462973</identifier><identifier>CODEN: IEIMAO</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Accuracy ; Artificial neural networks ; Attention ; Attention mechanism ; Computer vision ; Convolution ; Feature extraction ; feature fusion ; Image reconstruction ; Infrared imagery ; Infrared tracking ; Interference ; Multilayers ; Siamese network ; Support vector machines ; Target tracking ; thermal infrared (TIR) target tracking ; Training</subject><ispartof>IEEE transactions on instrumentation and measurement, 2024, Vol.73, p.1-11</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c175t-1a093548775d48b2df7dd6d194294119cef69300dc89f982222b9f93d9a8fd373</cites><orcidid>0000-0003-4652-4312 ; 0000-0001-9403-1112 ; 0000-0003-3834-3802 ; 0000-0003-0885-7976 ; 0000-0002-5610-8147 ; 0009-0004-9372-7686 ; 0000-0002-1205-9662</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/10683783$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,780,784,4024,27923,27924,27925,54796</link.rule.ids></links><search><creatorcontrib>Yuan, Di</creatorcontrib><creatorcontrib>Liao, Donghai</creatorcontrib><creatorcontrib>Huang, Feng</creatorcontrib><creatorcontrib>Qiu, Zhaobing</creatorcontrib><creatorcontrib>Shu, Xiu</creatorcontrib><creatorcontrib>Tian, Chunwei</creatorcontrib><creatorcontrib>Liu, Qiao</creatorcontrib><title>Hierarchical Attention Siamese Network for Thermal Infrared Target Tracking</title><title>IEEE transactions on instrumentation and measurement</title><addtitle>TIM</addtitle><description>Thermal infrared (TIR) target tracking is an important topic in the computer vision area. The TIR images are not affected by ambient light and have strong environmental adaptability, making them widely used in battlefield perception, video surveillance, and assisted driving. However, TIR target tracking faces problems such as relatively insufficient information and lack of target texture information, which significantly affects the tracking accuracy of the TIR tracking methods. To solve the above problems, we propose a TIR target tracking method based on a Siamese network with a hierarchical attention mechanism (called: SiamHAN). Specifically, the CIoU Loss is introduced to make full use of the regression box information to calculate the loss function more accurately. The global context network (GCNet) attention mechanism is introduced to reconstruct the feature extraction structure of fine-grained information for the fine-grained information of TIR images. Meanwhile, for the feature information of the hierarchical backbone network of the Siamese network, the ECANet attention mechanism is used for hierarchical feature fusion, so that it can fully utilize the feature information of the multilayer backbone network to represent the target. On the LSOTB-TIR, the hierarchical attention Siamese network achieved a 2.9% increase in success rate and a 4.3% increase in precision relative to the baseline tracker. Experiments show that the proposed SiamHAN method has achieved competitive tracking results on the TIR testing datasets.</description><subject>Accuracy</subject><subject>Artificial neural networks</subject><subject>Attention</subject><subject>Attention mechanism</subject><subject>Computer vision</subject><subject>Convolution</subject><subject>Feature extraction</subject><subject>feature fusion</subject><subject>Image reconstruction</subject><subject>Infrared imagery</subject><subject>Infrared tracking</subject><subject>Interference</subject><subject>Multilayers</subject><subject>Siamese network</subject><subject>Support vector machines</subject><subject>Target tracking</subject><subject>thermal infrared (TIR) target tracking</subject><subject>Training</subject><issn>0018-9456</issn><issn>1557-9662</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNpNkDlPAzEQRi0EEiHQU1BYot7ga32UUQQk4ipYasuxx8nm2A3ejRD_HkdJwTTfFO-bkR5Ct5SMKCXmoZq9jRhhYsSFZEbxMzSgZakKIyU7RwNCqC6MKOUluuq6FSFESaEG6GVaQ3LJL2vvNnjc99D0ddvgz9ptoQP8Dv1Pm9Y4tglXS0jbTM2amCsQcOXSAnpcJefXdbO4RhfRbTq4OeUQfT09VpNp8frxPJuMXwtPVdkX1BHDS6GVKoPQcxaiCkEGagQzglLjIUrDCQlem2g0yzPPCw_G6Ri44kN0f7y7S-33Hrrertp9avJLyyllVEthSKbIkfKp7boE0e5SvXXp11JiD8psVmYPyuxJWa7cHSs1APzDpeZKc_4H_PJmqQ</recordid><startdate>2024</startdate><enddate>2024</enddate><creator>Yuan, Di</creator><creator>Liao, Donghai</creator><creator>Huang, Feng</creator><creator>Qiu, Zhaobing</creator><creator>Shu, Xiu</creator><creator>Tian, Chunwei</creator><creator>Liu, Qiao</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SP</scope><scope>7U5</scope><scope>8FD</scope><scope>L7M</scope><orcidid>https://orcid.org/0000-0003-4652-4312</orcidid><orcidid>https://orcid.org/0000-0001-9403-1112</orcidid><orcidid>https://orcid.org/0000-0003-3834-3802</orcidid><orcidid>https://orcid.org/0000-0003-0885-7976</orcidid><orcidid>https://orcid.org/0000-0002-5610-8147</orcidid><orcidid>https://orcid.org/0009-0004-9372-7686</orcidid><orcidid>https://orcid.org/0000-0002-1205-9662</orcidid></search><sort><creationdate>2024</creationdate><title>Hierarchical Attention Siamese Network for Thermal Infrared Target Tracking</title><author>Yuan, Di ; Liao, Donghai ; Huang, Feng ; Qiu, Zhaobing ; Shu, Xiu ; Tian, Chunwei ; Liu, Qiao</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c175t-1a093548775d48b2df7dd6d194294119cef69300dc89f982222b9f93d9a8fd373</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Accuracy</topic><topic>Artificial neural networks</topic><topic>Attention</topic><topic>Attention mechanism</topic><topic>Computer vision</topic><topic>Convolution</topic><topic>Feature extraction</topic><topic>feature fusion</topic><topic>Image reconstruction</topic><topic>Infrared imagery</topic><topic>Infrared tracking</topic><topic>Interference</topic><topic>Multilayers</topic><topic>Siamese network</topic><topic>Support vector machines</topic><topic>Target tracking</topic><topic>thermal infrared (TIR) target tracking</topic><topic>Training</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Yuan, Di</creatorcontrib><creatorcontrib>Liao, Donghai</creatorcontrib><creatorcontrib>Huang, Feng</creatorcontrib><creatorcontrib>Qiu, Zhaobing</creatorcontrib><creatorcontrib>Shu, Xiu</creatorcontrib><creatorcontrib>Tian, Chunwei</creatorcontrib><creatorcontrib>Liu, Qiao</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library Online</collection><collection>CrossRef</collection><collection>Electronics & Communications Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>Technology Research Database</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>IEEE transactions on instrumentation and measurement</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Yuan, Di</au><au>Liao, Donghai</au><au>Huang, Feng</au><au>Qiu, Zhaobing</au><au>Shu, Xiu</au><au>Tian, Chunwei</au><au>Liu, Qiao</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Hierarchical Attention Siamese Network for Thermal Infrared Target Tracking</atitle><jtitle>IEEE transactions on instrumentation and measurement</jtitle><stitle>TIM</stitle><date>2024</date><risdate>2024</risdate><volume>73</volume><spage>1</spage><epage>11</epage><pages>1-11</pages><issn>0018-9456</issn><eissn>1557-9662</eissn><coden>IEIMAO</coden><abstract>Thermal infrared (TIR) target tracking is an important topic in the computer vision area. The TIR images are not affected by ambient light and have strong environmental adaptability, making them widely used in battlefield perception, video surveillance, and assisted driving. However, TIR target tracking faces problems such as relatively insufficient information and lack of target texture information, which significantly affects the tracking accuracy of the TIR tracking methods. To solve the above problems, we propose a TIR target tracking method based on a Siamese network with a hierarchical attention mechanism (called: SiamHAN). Specifically, the CIoU Loss is introduced to make full use of the regression box information to calculate the loss function more accurately. The global context network (GCNet) attention mechanism is introduced to reconstruct the feature extraction structure of fine-grained information for the fine-grained information of TIR images. Meanwhile, for the feature information of the hierarchical backbone network of the Siamese network, the ECANet attention mechanism is used for hierarchical feature fusion, so that it can fully utilize the feature information of the multilayer backbone network to represent the target. On the LSOTB-TIR, the hierarchical attention Siamese network achieved a 2.9% increase in success rate and a 4.3% increase in precision relative to the baseline tracker. Experiments show that the proposed SiamHAN method has achieved competitive tracking results on the TIR testing datasets.</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TIM.2024.3462973</doi><tpages>11</tpages><orcidid>https://orcid.org/0000-0003-4652-4312</orcidid><orcidid>https://orcid.org/0000-0001-9403-1112</orcidid><orcidid>https://orcid.org/0000-0003-3834-3802</orcidid><orcidid>https://orcid.org/0000-0003-0885-7976</orcidid><orcidid>https://orcid.org/0000-0002-5610-8147</orcidid><orcidid>https://orcid.org/0009-0004-9372-7686</orcidid><orcidid>https://orcid.org/0000-0002-1205-9662</orcidid></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0018-9456 |
ispartof | IEEE transactions on instrumentation and measurement, 2024, Vol.73, p.1-11 |
issn | 0018-9456 1557-9662 |
language | eng |
recordid | cdi_proquest_journals_3112186490 |
source | IEEE Electronic Library (IEL) Journals |
subjects | Accuracy Artificial neural networks Attention Attention mechanism Computer vision Convolution Feature extraction feature fusion Image reconstruction Infrared imagery Infrared tracking Interference Multilayers Siamese network Support vector machines Target tracking thermal infrared (TIR) target tracking Training |
title | Hierarchical Attention Siamese Network for Thermal Infrared Target Tracking |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T15%3A38%3A48IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_ieee_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Hierarchical%20Attention%20Siamese%20Network%20for%20Thermal%20Infrared%20Target%20Tracking&rft.jtitle=IEEE%20transactions%20on%20instrumentation%20and%20measurement&rft.au=Yuan,%20Di&rft.date=2024&rft.volume=73&rft.spage=1&rft.epage=11&rft.pages=1-11&rft.issn=0018-9456&rft.eissn=1557-9662&rft.coden=IEIMAO&rft_id=info:doi/10.1109/TIM.2024.3462973&rft_dat=%3Cproquest_ieee_%3E3112186490%3C/proquest_ieee_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c175t-1a093548775d48b2df7dd6d194294119cef69300dc89f982222b9f93d9a8fd373%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=3112186490&rft_id=info:pmid/&rft_ieee_id=10683783&rfr_iscdi=true |