KeyShip: Towards High-Precision Oriented SAR Ship Detection Using Key Points
Open Access
- 12 April 2023
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 15 (8) , 2035
Abstract
Synthetic Aperture Radar (SAR) is an all-weather sensing technology that has proven its effectiveness for ship detection. However, detecting ships accurately with oriented bounding boxes (OBB) on SAR images is challenging due to arbitrary ship orientations and misleading scattering. In this article, we propose a novel anchor-free key-point-based detection method, KeyShip, for detecting orientated SAR ships with high precision. Our approach uses a shape descriptor to model a ship as a combination of three types of key points located at the short-edge centers, long-edge centers, and the target center. These key points are detected separately and clustered based on predicted shape descriptors to construct the final OBB detection results. To address the boundary problem that arises with the shape descriptor representation, we propose a soft training target assignment strategy that facilitates successful shape descriptor training and implicitly learns the shape information of the targets. Our experimental results on three datasets (SSDD, RSDD, and HRSC2016) demonstrate our proposed method’s high performance and robustness.Keywords
Funding Information
- National Natural Science Foundation of China (61976167, U19B2030, 62101416, 2022JQ-708)
- Natural Science Basic Research Program of Shaanxi (61976167, U19B2030, 62101416, 2022JQ-708)
This publication has 56 references indexed in Scilit:
- Proposal of wet snowmapping with focus on incident angle influential to depolarization of surface scattering2016
- SSD: Single Shot MultiBox Detector2016
- Stacked Hourglass Networks for Human Pose Estimation2016
- You Only Look Once: Unified, Real-Time Object Detection2016
- Robust CFAR Detector Based on Truncated Statistics in Multiple-Target SituationsIEEE Transactions on Geoscience and Remote Sensing, 2015
- Quaternion Neural-Network-Based PolSAR Land Classification in Poincare-Sphere-Parameter SpaceIEEE Transactions on Geoscience and Remote Sensing, 2013
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009
- An Adaptive and Fast CFAR Algorithm Based on Automatic Censoring for Target Detection in High-Resolution SAR ImagesIEEE Transactions on Geoscience and Remote Sensing, 2008
- A CFAR adaptive matched filter detectorIEEE Transactions on Aerospace and Electronic Systems, 1992
- Adaptive cell-averaging CFAR detection in distributed sensor networksIEEE Transactions on Aerospace and Electronic Systems, 1991