NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attention

Due to the advantages of all-weather capability and high resolution, synthetic aperture radar (SAR) image ship detection has been widely applied in the military, civilian, and other domains. However, SAR-based ship detection suffers from limitations such as strong scattering of targets, multiple sca...

Full description

Saved in:
Bibliographic Details
Main Authors: Hao Wang, Dezhi Han, Mingming Cui, Chongqing Chen
Format: Article
Language:English
Published: Taylor & Francis Group 2023-12-01
Series:Connection Science
Subjects:
Online Access:http://dx.doi.org/10.1080/09540091.2023.2257399
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702994546786304
author Hao Wang
Dezhi Han
Mingming Cui
Chongqing Chen
author_facet Hao Wang
Dezhi Han
Mingming Cui
Chongqing Chen
author_sort Hao Wang
collection DOAJ
description Due to the advantages of all-weather capability and high resolution, synthetic aperture radar (SAR) image ship detection has been widely applied in the military, civilian, and other domains. However, SAR-based ship detection suffers from limitations such as strong scattering of targets, multiple scales, and background interference, leading to low detection accuracy. To address these limitations, this paper presents a novel SAR ship detection method, NAS-YOLOX, which leverages the efficient feature fusion of the neural architecture search feature pyramid network (NAS-FPN) and the effective feature extraction of the multi-scale attention mechanism. Specifically, NAS-FPN replaces the PAFPN in the baseline YOLOX, greatly enhances the fusion performance of the model’s multi-scale feature information, and a dilated convolution feature enhancement module (DFEM) is designed and integrated into the backbone network to improve the network’s receptive field and target information extraction capabilities. Furthermore, a multi-scale channel-spatial attention (MCSA) mechanism is conceptualised to enhance focus on target regions, improve small-scale target detection, and adapt to multi-scale targets. Additionally, extensive experiments conducted on benchmark datasets, HRSID and SSDD, demonstrate that NAS-YOLOX achieves comparable or superior performance compared to other state-of-the-art ship detection models and reaches best accuracies of 91.1% and 97.2% on AP0.5, respectively.
format Article
id doaj-art-0fd2339f9ecf4d90a2bf2bd7e32cbc78
institution DOAJ
issn 0954-0091
1360-0494
language English
publishDate 2023-12-01
publisher Taylor & Francis Group
record_format Article
series Connection Science
spelling doaj-art-0fd2339f9ecf4d90a2bf2bd7e32cbc782025-08-20T03:17:27ZengTaylor & Francis GroupConnection Science0954-00911360-04942023-12-0135113210.1080/09540091.2023.22573992257399NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attentionHao Wang0Dezhi Han1Mingming Cui2Chongqing Chen3Shanghai Maritime UniversityShanghai Maritime UniversityShanghai Maritime UniversityShanghai Maritime UniversityDue to the advantages of all-weather capability and high resolution, synthetic aperture radar (SAR) image ship detection has been widely applied in the military, civilian, and other domains. However, SAR-based ship detection suffers from limitations such as strong scattering of targets, multiple scales, and background interference, leading to low detection accuracy. To address these limitations, this paper presents a novel SAR ship detection method, NAS-YOLOX, which leverages the efficient feature fusion of the neural architecture search feature pyramid network (NAS-FPN) and the effective feature extraction of the multi-scale attention mechanism. Specifically, NAS-FPN replaces the PAFPN in the baseline YOLOX, greatly enhances the fusion performance of the model’s multi-scale feature information, and a dilated convolution feature enhancement module (DFEM) is designed and integrated into the backbone network to improve the network’s receptive field and target information extraction capabilities. Furthermore, a multi-scale channel-spatial attention (MCSA) mechanism is conceptualised to enhance focus on target regions, improve small-scale target detection, and adapt to multi-scale targets. Additionally, extensive experiments conducted on benchmark datasets, HRSID and SSDD, demonstrate that NAS-YOLOX achieves comparable or superior performance compared to other state-of-the-art ship detection models and reaches best accuracies of 91.1% and 97.2% on AP0.5, respectively.http://dx.doi.org/10.1080/09540091.2023.2257399synthetic aperture radar (sar)ship detectionyou only look once version x (yolox)neural architecture search-feature pyramid network (nas-fpn)
spellingShingle Hao Wang
Dezhi Han
Mingming Cui
Chongqing Chen
NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attention
Connection Science
synthetic aperture radar (sar)
ship detection
you only look once version x (yolox)
neural architecture search-feature pyramid network (nas-fpn)
title NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attention
title_full NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attention
title_fullStr NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attention
title_full_unstemmed NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attention
title_short NAS-YOLOX: a SAR ship detection using neural architecture search and multi-scale attention
title_sort nas yolox a sar ship detection using neural architecture search and multi scale attention
topic synthetic aperture radar (sar)
ship detection
you only look once version x (yolox)
neural architecture search-feature pyramid network (nas-fpn)
url http://dx.doi.org/10.1080/09540091.2023.2257399
work_keys_str_mv AT haowang nasyoloxasarshipdetectionusingneuralarchitecturesearchandmultiscaleattention
AT dezhihan nasyoloxasarshipdetectionusingneuralarchitecturesearchandmultiscaleattention
AT mingmingcui nasyoloxasarshipdetectionusingneuralarchitecturesearchandmultiscaleattention
AT chongqingchen nasyoloxasarshipdetectionusingneuralarchitecturesearchandmultiscaleattention