{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T23:28:10Z","timestamp":1773271690903,"version":"3.50.1"},"reference-count":60,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2021,3,23]],"date-time":"2021-03-23T00:00:00Z","timestamp":1616457600000},"content-version":"vor","delay-in-days":81,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Director\u2019s Foundation of Institute of Microelectronics, Chinese Academy of Sciences","award":["E0518101"],"award-info":[{"award-number":["E0518101"]}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Journal of Sensors"],"published-print":{"date-parts":[[2021,1]]},"abstract":"<jats:p>Ship detection on synthetic aperture radar (SAR) imagery has many valuable applications for both civil and military fields and has received extraordinary attention in recent years. The traditional detection methods are insensitive to multiscale ships and usually time\u2010consuming, results in low detection accuracy and limitation for real\u2010time processing. To balance the accuracy and speed, an end\u2010to\u2010end ship detection method for complex inshore and offshore scenes based on deep convolutional neural networks (CNNs) is proposed in this paper. First, the SAR images are divided into different grids, and the anchor boxes are predefined based on the responsible grids for dense ship prediction. Then, Darknet\u201053 with residual units is adopted as a backbone to extract features, and a top\u2010down pyramid structure is added for multiscale feature fusion with concatenation. By this means, abundant hierarchical features containing both spatial and semantic information are extracted. Meanwhile, the strategies such as soft non\u2010maximum suppression (Soft\u2010NMS), mix\u2010up and mosaic data augmentation, multiscale training, and hybrid optimization are used for performance enhancement. Besides, the model is trained from scratch to avoid learning objective bias of pretraining. The proposed one\u2010stage method adopts end\u2010to\u2010end inference by a single network, so the detection speed can be guaranteed due to the concise paradigm. Extensive experiments are performed on the public SAR ship detection dataset (SSDD), and the results show that the method can detect both inshore and offshore ships with higher accuracy than other mainstream methods, yielding the accuracy with an average of 95.52%, and the detection speed is quite fast with about 72 frames per second (FPS). The actual Sentinel\u20101 and Gaofen\u20103 data are utilized for verification, and the detection results also show the effectiveness and robustness of the method.<\/jats:p>","DOI":"10.1155\/2021\/8893182","type":"journal-article","created":{"date-parts":[[2021,3,23]],"date-time":"2021-03-23T18:51:52Z","timestamp":1616525512000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":16,"title":["End\u2010to\u2010End Ship Detection in SAR Images for Complex Scenes Based on Deep CNNs"],"prefix":"10.1155","volume":"2021","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9656-1448","authenticated-orcid":false,"given":"Yao","family":"Chen","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tao","family":"Duan","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Changyuan","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuanyuan","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6241-487X","authenticated-orcid":false,"given":"Mo","family":"Huang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2021,3,23]]},"reference":[{"key":"e_1_2_8_1_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2010.2071879"},{"key":"e_1_2_8_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/MGRS.2013.2248301"},{"key":"e_1_2_8_3_2","volume-title":"Synthetic Aperture Radar Marine User\u2032s Manual","author":"Jackson C.","year":"2004"},{"key":"e_1_2_8_4_2","doi-asserted-by":"publisher","DOI":"10.1155\/2019\/1246548"},{"key":"e_1_2_8_5_2","doi-asserted-by":"crossref","unstructured":"SantamariaC.andGreidanusH. Ambiguity discrimination for ship detection using Sentinel-1 repeat acquisition operations Proc. 2015 IEEE Int. Geosci. Remote Sens. Symp. (IGARSS) 2015 Milan 2477\u20132480.","DOI":"10.1109\/IGARSS.2015.7326312"},{"key":"e_1_2_8_6_2","doi-asserted-by":"publisher","DOI":"10.1109\/LGRS.2018.2838043"},{"key":"e_1_2_8_7_2","doi-asserted-by":"publisher","DOI":"10.1109\/LGRS.2005.845033"},{"key":"e_1_2_8_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/36.368224"},{"key":"e_1_2_8_9_2","doi-asserted-by":"publisher","DOI":"10.1155\/2015\/437083"},{"key":"e_1_2_8_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2004.842196"},{"key":"e_1_2_8_11_2","doi-asserted-by":"publisher","DOI":"10.1109\/7.805463"},{"key":"e_1_2_8_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/TAES.2014.130683"},{"key":"e_1_2_8_13_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2019.2931353"},{"key":"e_1_2_8_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/TAES.1982.309210"},{"key":"e_1_2_8_15_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2008.2006504"},{"key":"e_1_2_8_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/7.78295"},{"key":"e_1_2_8_17_2","doi-asserted-by":"publisher","DOI":"10.1109\/LGRS.2019.2960095"},{"key":"e_1_2_8_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2019.2906054"},{"key":"e_1_2_8_19_2","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"e_1_2_8_20_2","volume-title":"Deep Learning","author":"Goodfellow I.","year":"2016"},{"key":"e_1_2_8_21_2","doi-asserted-by":"crossref","unstructured":"GirshickR. DonahueJ. DarrellT. andMalikJ. Rich feature hierarchies for accurate object detection and semantic segmentation Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR) 2014 Columbus OH USA 580\u2013587.","DOI":"10.1109\/CVPR.2014.81"},{"key":"e_1_2_8_22_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2577031"},{"key":"e_1_2_8_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2844175"},{"key":"e_1_2_8_24_2","unstructured":"DaiJ. LiY. HeK. andSunJ. R-FCN: object detection via region-based fully convolutional networks Proc. Int. Conf. Neural Inf. Process. Syst. 2016 Barcelona Spain 379\u2013387 https:\/\/arxiv.org\/abs\/1605.06409."},{"key":"e_1_2_8_25_2","doi-asserted-by":"crossref","unstructured":"RedmonJ. DivvalaS. GirshickR. andFarhadiA. You only look once: unified real-time object detection Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR) 2016 Las Vegas NV USA 779\u2013788.","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_2_8_26_2","doi-asserted-by":"crossref","unstructured":"LiuW. AnguelovD. ErhanD. SzegedyC. FuC.-Y. andBergA. C. SSD: single shot multibox detector Proc. Eur. Conf. Comput. Vis. (ECCV) 2016 Amsterdam The Netherlands 21\u201337.","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"e_1_2_8_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2858826"},{"key":"e_1_2_8_28_2","unstructured":"RedmonJ.andFarhadiA. YOLOv3: an incremental improvement 2018 https:\/\/arxiv.org\/abs\/1804.02767."},{"key":"e_1_2_8_29_2","unstructured":"BochkovskiyA. WangC. andLiaoH. YOLOv4: optimal speed and accuracy of object detection 2020 https:\/\/arxiv.org\/pdf\/2004.10934."},{"key":"e_1_2_8_30_2","doi-asserted-by":"crossref","unstructured":"LiJ. QuC. andShaoJ. Ship detection in SAR images based on an improved faster R-CNN Proc. BIGSARDATA 2017 Beijing China 1\u20136.","DOI":"10.1109\/BIGSARDATA.2017.8124934"},{"key":"e_1_2_8_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2825376"},{"key":"e_1_2_8_32_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2869289"},{"key":"e_1_2_8_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2019.2923988"},{"key":"e_1_2_8_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.2985637"},{"key":"e_1_2_8_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2019.2920534"},{"key":"e_1_2_8_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2019.2951030"},{"key":"e_1_2_8_37_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2018.2889353"},{"key":"e_1_2_8_38_2","doi-asserted-by":"publisher","DOI":"10.3390\/rs11070786"},{"key":"e_1_2_8_39_2","doi-asserted-by":"publisher","DOI":"10.3390\/rs11101206"},{"key":"e_1_2_8_40_2","doi-asserted-by":"publisher","DOI":"10.3390\/rs11212483"},{"key":"e_1_2_8_41_2","doi-asserted-by":"publisher","DOI":"10.1109\/LGRS.2016.2616187"},{"key":"e_1_2_8_42_2","doi-asserted-by":"publisher","DOI":"10.1109\/MGRS.2017.2762307"},{"key":"e_1_2_8_43_2","unstructured":"ZouZ. ShiZ. GuoY. andYeJ. Object detection in 20 years: a survey 2019 https:\/\/arxiv.org\/abs\/1905.05055."},{"key":"e_1_2_8_44_2","doi-asserted-by":"crossref","unstructured":"HeK. ZhangX. RenS. andSunJ. Deep residual learning for image recognition Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR) 2016 Las Vegas NV USA 770\u2013778.","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_8_45_2","doi-asserted-by":"crossref","unstructured":"LinT.-Y. Doll\u00e1rP. GirshickR. HeK. HariharanB. andBelongieS. Feature pyramid networks for object detection Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR) 2017 Honolulu HI USA 936\u2013944.","DOI":"10.1109\/CVPR.2017.106"},{"key":"e_1_2_8_46_2","doi-asserted-by":"crossref","unstructured":"SzegedyC. LiuW. JiaY. SermanetP. ReedS. AnguelovD. ErhanD. VanhouckeV. andRabinovichA. Going deeper with convolutions Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR) 2015 Boston MA USA 1\u20139.","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_2_8_47_2","unstructured":"XuB. WangN. ChenT. andLiM. Empirical evaluation of rectified activations in convolutional network 2018 https:\/\/arxiv.org\/abs\/1505.00853."},{"key":"e_1_2_8_48_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.2978753"},{"key":"e_1_2_8_49_2","doi-asserted-by":"crossref","unstructured":"BodlaN. SinghB. ChellappaR. andDavisL. S. Soft-NMS\u2013improving object detection with one line of code Proc. IEEE Conf. Comput. Vis. (ICCV) 2017 Venice Italy 5562\u20135570.","DOI":"10.1109\/ICCV.2017.593"},{"key":"e_1_2_8_50_2","first-page":"1097","volume-title":"Advances in Neural Information Processing Systems","author":"Krizhevsky A.","year":"2012"},{"key":"e_1_2_8_51_2","unstructured":"HeK. GirshickR. andDoll\u00e1rP. Rethinking ImageNet pre-training 2018 https:\/\/arxiv.org\/abs\/1811.08883."},{"key":"e_1_2_8_52_2","unstructured":"PaddlePaddle Parallel distributed deep learning: machine learning framework from industrial practice https:\/\/www.paddlepaddle.org\/."},{"key":"e_1_2_8_53_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-014-0733-5"},{"key":"e_1_2_8_54_2","unstructured":"ZhangH. CisseM. DauphinY. N. andLopez-PazD. Mixup: beyond empirical risk minimization 2017 https:\/\/arxiv.org\/abs\/1710.09412."},{"key":"e_1_2_8_55_2","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_2_8_56_2","unstructured":"KeskarN.andSocherR. Improving generalization performance by switching from Adam to SGD 2017 https:\/\/arxiv.org\/abs\/1712.07628."},{"key":"e_1_2_8_57_2","doi-asserted-by":"crossref","unstructured":"LinT. MaireM. BelongieS. HaysJ. PeronaP. RamananD. andZitnickC. Microsoft coco: common objects in context Proc. of the 13th Europ. Conf. 2014 Zurich Switzerland 740\u2013755.","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"e_1_2_8_58_2","unstructured":"ESA Copernicus open access hub 2020 https:\/\/scihub.copernicus.eu\/."},{"key":"e_1_2_8_59_2","first-page":"852","article-title":"Air-SARSHIP\u20131.0: high resolution SAR ship detection dataset","volume":"8","author":"Sun X.","year":"2019","journal-title":"Journal of Radars"},{"key":"e_1_2_8_60_2","unstructured":"Marine Traffic Historical AIS data 2020 https:\/\/www.marinetraffic.com\/zh\/p\/ais-historical-data."}],"container-title":["Journal of Sensors"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/downloads.hindawi.com\/journals\/js\/2021\/8893182.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/js\/2021\/8893182.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1155\/2021\/8893182","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,6]],"date-time":"2024-08-06T00:25:20Z","timestamp":1722903920000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1155\/2021\/8893182"}},"subtitle":[],"editor":[{"given":"Jaime","family":"Lloret","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2021,1]]},"references-count":60,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2021,1]]}},"alternative-id":["10.1155\/2021\/8893182"],"URL":"https:\/\/doi.org\/10.1155\/2021\/8893182","archive":["Portico"],"relation":{},"ISSN":["1687-725X","1687-7268"],"issn-type":[{"value":"1687-725X","type":"print"},{"value":"1687-7268","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,1]]},"assertion":[{"value":"2020-09-11","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-02-12","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-03-23","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"8893182"}}