Global Mask R-CNN for marine ship instance segmentation

doi:10.1016/j.neucom.2022.01.017

	Global Mask R-CNN for marine ship instance segmentation
	Sun, Yuxin 1; Su, Li 1; Luo, Yongkang 2; Meng, Hao 1; Li, Wanyi 2; Zhang, Zhi 1; Wang, Peng 2; Zhang, Wen 1
刊名	NEUROCOMPUTING
	2022-04-01
卷号	480 页码:257-270
关键词	Instance segmentation Ship dataset Ship instance segmentation Precise RoI Pooling Global Mask Head
ISSN号	0925-2312
DOI	10.1016/j.neucom.2022.01.017
通讯作者	Su, Li(suli406@hrbeu.edu.cn)
英文摘要	Instance segmentation technology can provide accurate and efficient segmentation methods for visual perception of marine scenes, especially in the development of unmanned ships. However, the community lacks suitable open-source datasets. In order to address the problem of insufficient datasets, an instance segmentation dataset for the marine ship was collected and labeled. Our dataset, named MariShipInsSeg, consists of 4k high-quality visible light marine ship images with 8,413 instances. Due to marine ships being photographed far away, which causes ship objects with less detail information. Therefore, a global method is adopted to make full use of global location and semantic information, which is helpful for ship instance segmentation. We proposed a new method called Global Mask R-CNN (GM R-CNN), which utilized Precise RoI Pooling and Global Mask Head aiming to preserve global information of instances for improving the performance of ship instance segmentation. Experiments on the challenging MS COCO dataset and MariShipInsSeg dataset show that Global Mask R-CNN achieves state-of-the-art performance. Without any bells and whistles, the proposed GM R-CNN achieves 38.7% mask AP on MS COCO test-dev and 48.6% mask AP on MariShipInsSeg testing sets, which are gain of 1.6% and 1.9% compared with Mask R-CNN. (c) 2022 Elsevier B.V. All rights reserved.
资助项目	National Key R&D Program of China[2018YFB1601502] ; Project of Intelligent Situation Awareness System for Smart Ship[MC-201920-X01] ; National Natural Science Foundation of China[U1613213] ; National Natural Science Foundation of China[91748131] ; National Natural Science Foundation of China[61771471]
WOS关键词	SAR IMAGES ; DATASET ; SYSTEM
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000761803900005
资助机构	National Key R&D Program of China ; Project of Intelligent Situation Awareness System for Smart Ship ; National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/47987]
专题	智能机器人系统研究
通讯作者	Su, Li
作者单位	1.Harbin Engn Univ, Coll Intelligent Syst Sci & Engn, Room 4138,Bldg 61,145 Nantong Ave, Harbin 150001, Heilongjiang, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Sun, Yuxin,Su, Li,Luo, Yongkang,et al. Global Mask R-CNN for marine ship instance segmentation[J]. NEUROCOMPUTING,2022,480:257-270.
APA	Sun, Yuxin.,Su, Li.,Luo, Yongkang.,Meng, Hao.,Li, Wanyi.,...&Zhang, Wen.(2022).Global Mask R-CNN for marine ship instance segmentation.NEUROCOMPUTING,480,257-270.
MLA	Sun, Yuxin,et al."Global Mask R-CNN for marine ship instance segmentation".NEUROCOMPUTING 480(2022):257-270.