Robust ISAR Target Recognition Based on ADRISAR-Net
Due to the inherent unknown image deformation among the training and test samples, performance of the deep convolutional neural network (CNN) will be degraded for Inverse Synthetic Aperture Radar (ISAR) automatic target recognition. Meanwhile, traditional CNN only captures the local spatial informat...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on aerospace and electronic systems 2022-12, Vol.58 (6), p.5494-5505 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 5505 |
|---|---|
| container_issue | 6 |
| container_start_page | 5494 |
| container_title | IEEE transactions on aerospace and electronic systems |
| container_volume | 58 |
| creator | Zhou, Xuening Bai, Xueru Wang, Li Zhou, Feng |
| description | Due to the inherent unknown image deformation among the training and test samples, performance of the deep convolutional neural network (CNN) will be degraded for Inverse Synthetic Aperture Radar (ISAR) automatic target recognition. Meanwhile, traditional CNN only captures the local spatial information due to small receptive fields, thus, neglects the global information useful for recognition. To tackle these issues, this article proposes the attention-augmented deformation robust ISAR image recognition network, dubbed as ADRISAR-Net. The model adopts the inverse compositional spatial transformer for automatic image deformation adjustment, and performs joint local and global feature extractions by the attention-augmented CNN. Finally, the softmax classifier outputs the recognition results. The proposed ADRISAR-Net is end-to-end trainable, and achieves higher recognition accuracy for the four-satellite and three-airplane ISAR image data sets generated by electromagnetic computing. |
| doi_str_mv | 10.1109/TAES.2022.3174826 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_ieee_primary_9774276</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9774276</ieee_id><sourcerecordid>2747611322</sourcerecordid><originalsourceid>FETCH-LOGICAL-c293t-ca0ac038cfadd5f0e9b57cc7e9ad40287b213901fda007d7edc072907b5a61c83</originalsourceid><addsrcrecordid>eNo9kMFKw0AQhhdRsFYfQLwEPCfO7CaZ7DHWqoWikNbzstndlBRt6m568O2b0OJp_oHvn4GPsXuEBBHk07qcrxIOnCcCKS14fsEmmGUUyxzEJZsAYBFLnuE1uwlhO6xpkYoJE1VXH0IfLVZlFa2137g-qpzpNru2b7td9KyDs9EQypdqZOIP19-yq0Z_B3d3nlP29Tpfz97j5efbYlYuY8Ol6GOjQRsQhWm0tVkDTtYZGUNOapsCL6jmKCRgYzUAWXLWAHEJVGc6R1OIKXs83d377vfgQq-23cHvhpeKU0o5ouB8oPBEGd-F4F2j9r790f5PIajRjRrdqNGNOrsZOg-nTuuc--clUcopF0dKFl2C</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2747611322</pqid></control><display><type>article</type><title>Robust ISAR Target Recognition Based on ADRISAR-Net</title><source>IEEE Electronic Library (IEL)</source><creator>Zhou, Xuening ; Bai, Xueru ; Wang, Li ; Zhou, Feng</creator><creatorcontrib>Zhou, Xuening ; Bai, Xueru ; Wang, Li ; Zhou, Feng</creatorcontrib><description>Due to the inherent unknown image deformation among the training and test samples, performance of the deep convolutional neural network (CNN) will be degraded for Inverse Synthetic Aperture Radar (ISAR) automatic target recognition. Meanwhile, traditional CNN only captures the local spatial information due to small receptive fields, thus, neglects the global information useful for recognition. To tackle these issues, this article proposes the attention-augmented deformation robust ISAR image recognition network, dubbed as ADRISAR-Net. The model adopts the inverse compositional spatial transformer for automatic image deformation adjustment, and performs joint local and global feature extractions by the attention-augmented CNN. Finally, the softmax classifier outputs the recognition results. The proposed ADRISAR-Net is end-to-end trainable, and achieves higher recognition accuracy for the four-satellite and three-airplane ISAR image data sets generated by electromagnetic computing.</description><identifier>ISSN: 0018-9251</identifier><identifier>EISSN: 1557-9603</identifier><identifier>DOI: 10.1109/TAES.2022.3174826</identifier><identifier>CODEN: IEARAX</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Artificial neural networks ; Attention ; Automatic target recognition ; automatic target recognition (ATR) ; Convolution ; convolutional neural network (CNN) ; Deformable models ; Deformation ; Feature extraction ; Generators ; image deformation ; Inverse synthetic aperture radar ; inverse synthetic aperture radar (ISAR) ; Object recognition ; Robustness ; Satellite imagery ; Spatial data ; Strain ; Target recognition</subject><ispartof>IEEE transactions on aerospace and electronic systems, 2022-12, Vol.58 (6), p.5494-5505</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c293t-ca0ac038cfadd5f0e9b57cc7e9ad40287b213901fda007d7edc072907b5a61c83</citedby><cites>FETCH-LOGICAL-c293t-ca0ac038cfadd5f0e9b57cc7e9ad40287b213901fda007d7edc072907b5a61c83</cites><orcidid>0000-0001-9283-1810 ; 0000-0002-1514-7393</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9774276$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,780,784,796,27915,27916,54749</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9774276$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Zhou, Xuening</creatorcontrib><creatorcontrib>Bai, Xueru</creatorcontrib><creatorcontrib>Wang, Li</creatorcontrib><creatorcontrib>Zhou, Feng</creatorcontrib><title>Robust ISAR Target Recognition Based on ADRISAR-Net</title><title>IEEE transactions on aerospace and electronic systems</title><addtitle>T-AES</addtitle><description>Due to the inherent unknown image deformation among the training and test samples, performance of the deep convolutional neural network (CNN) will be degraded for Inverse Synthetic Aperture Radar (ISAR) automatic target recognition. Meanwhile, traditional CNN only captures the local spatial information due to small receptive fields, thus, neglects the global information useful for recognition. To tackle these issues, this article proposes the attention-augmented deformation robust ISAR image recognition network, dubbed as ADRISAR-Net. The model adopts the inverse compositional spatial transformer for automatic image deformation adjustment, and performs joint local and global feature extractions by the attention-augmented CNN. Finally, the softmax classifier outputs the recognition results. The proposed ADRISAR-Net is end-to-end trainable, and achieves higher recognition accuracy for the four-satellite and three-airplane ISAR image data sets generated by electromagnetic computing.</description><subject>Artificial neural networks</subject><subject>Attention</subject><subject>Automatic target recognition</subject><subject>automatic target recognition (ATR)</subject><subject>Convolution</subject><subject>convolutional neural network (CNN)</subject><subject>Deformable models</subject><subject>Deformation</subject><subject>Feature extraction</subject><subject>Generators</subject><subject>image deformation</subject><subject>Inverse synthetic aperture radar</subject><subject>inverse synthetic aperture radar (ISAR)</subject><subject>Object recognition</subject><subject>Robustness</subject><subject>Satellite imagery</subject><subject>Spatial data</subject><subject>Strain</subject><subject>Target recognition</subject><issn>0018-9251</issn><issn>1557-9603</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNo9kMFKw0AQhhdRsFYfQLwEPCfO7CaZ7DHWqoWikNbzstndlBRt6m568O2b0OJp_oHvn4GPsXuEBBHk07qcrxIOnCcCKS14fsEmmGUUyxzEJZsAYBFLnuE1uwlhO6xpkYoJE1VXH0IfLVZlFa2137g-qpzpNru2b7td9KyDs9EQypdqZOIP19-yq0Z_B3d3nlP29Tpfz97j5efbYlYuY8Ol6GOjQRsQhWm0tVkDTtYZGUNOapsCL6jmKCRgYzUAWXLWAHEJVGc6R1OIKXs83d377vfgQq-23cHvhpeKU0o5ouB8oPBEGd-F4F2j9r790f5PIajRjRrdqNGNOrsZOg-nTuuc--clUcopF0dKFl2C</recordid><startdate>20221201</startdate><enddate>20221201</enddate><creator>Zhou, Xuening</creator><creator>Bai, Xueru</creator><creator>Wang, Li</creator><creator>Zhou, Feng</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>7TB</scope><scope>8FD</scope><scope>FR3</scope><scope>H8D</scope><scope>L7M</scope><orcidid>https://orcid.org/0000-0001-9283-1810</orcidid><orcidid>https://orcid.org/0000-0002-1514-7393</orcidid></search><sort><creationdate>20221201</creationdate><title>Robust ISAR Target Recognition Based on ADRISAR-Net</title><author>Zhou, Xuening ; Bai, Xueru ; Wang, Li ; Zhou, Feng</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c293t-ca0ac038cfadd5f0e9b57cc7e9ad40287b213901fda007d7edc072907b5a61c83</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Artificial neural networks</topic><topic>Attention</topic><topic>Automatic target recognition</topic><topic>automatic target recognition (ATR)</topic><topic>Convolution</topic><topic>convolutional neural network (CNN)</topic><topic>Deformable models</topic><topic>Deformation</topic><topic>Feature extraction</topic><topic>Generators</topic><topic>image deformation</topic><topic>Inverse synthetic aperture radar</topic><topic>inverse synthetic aperture radar (ISAR)</topic><topic>Object recognition</topic><topic>Robustness</topic><topic>Satellite imagery</topic><topic>Spatial data</topic><topic>Strain</topic><topic>Target recognition</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zhou, Xuening</creatorcontrib><creatorcontrib>Bai, Xueru</creatorcontrib><creatorcontrib>Wang, Li</creatorcontrib><creatorcontrib>Zhou, Feng</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 (IEL)</collection><collection>CrossRef</collection><collection>Electronics & Communications Abstracts</collection><collection>Mechanical & Transportation Engineering Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>Aerospace Database</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>IEEE transactions on aerospace and electronic systems</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhou, Xuening</au><au>Bai, Xueru</au><au>Wang, Li</au><au>Zhou, Feng</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Robust ISAR Target Recognition Based on ADRISAR-Net</atitle><jtitle>IEEE transactions on aerospace and electronic systems</jtitle><stitle>T-AES</stitle><date>2022-12-01</date><risdate>2022</risdate><volume>58</volume><issue>6</issue><spage>5494</spage><epage>5505</epage><pages>5494-5505</pages><issn>0018-9251</issn><eissn>1557-9603</eissn><coden>IEARAX</coden><abstract>Due to the inherent unknown image deformation among the training and test samples, performance of the deep convolutional neural network (CNN) will be degraded for Inverse Synthetic Aperture Radar (ISAR) automatic target recognition. Meanwhile, traditional CNN only captures the local spatial information due to small receptive fields, thus, neglects the global information useful for recognition. To tackle these issues, this article proposes the attention-augmented deformation robust ISAR image recognition network, dubbed as ADRISAR-Net. The model adopts the inverse compositional spatial transformer for automatic image deformation adjustment, and performs joint local and global feature extractions by the attention-augmented CNN. Finally, the softmax classifier outputs the recognition results. The proposed ADRISAR-Net is end-to-end trainable, and achieves higher recognition accuracy for the four-satellite and three-airplane ISAR image data sets generated by electromagnetic computing.</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TAES.2022.3174826</doi><tpages>12</tpages><orcidid>https://orcid.org/0000-0001-9283-1810</orcidid><orcidid>https://orcid.org/0000-0002-1514-7393</orcidid></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 0018-9251 |
| ispartof | IEEE transactions on aerospace and electronic systems, 2022-12, Vol.58 (6), p.5494-5505 |
| issn | 0018-9251 1557-9603 |
| language | eng |
| recordid | cdi_ieee_primary_9774276 |
| source | IEEE Electronic Library (IEL) |
| subjects | Artificial neural networks Attention Automatic target recognition automatic target recognition (ATR) Convolution convolutional neural network (CNN) Deformable models Deformation Feature extraction Generators image deformation Inverse synthetic aperture radar inverse synthetic aperture radar (ISAR) Object recognition Robustness Satellite imagery Spatial data Strain Target recognition |
| title | Robust ISAR Target Recognition Based on ADRISAR-Net |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-15T06%3A38%3A28IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Robust%20ISAR%20Target%20Recognition%20Based%20on%20ADRISAR-Net&rft.jtitle=IEEE%20transactions%20on%20aerospace%20and%20electronic%20systems&rft.au=Zhou,%20Xuening&rft.date=2022-12-01&rft.volume=58&rft.issue=6&rft.spage=5494&rft.epage=5505&rft.pages=5494-5505&rft.issn=0018-9251&rft.eissn=1557-9603&rft.coden=IEARAX&rft_id=info:doi/10.1109/TAES.2022.3174826&rft_dat=%3Cproquest_RIE%3E2747611322%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2747611322&rft_id=info:pmid/&rft_ieee_id=9774276&rfr_iscdi=true |