Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR Images
With the wide application of synthetic aperture radar in maritime surveillance, a ship detection method has been rapidly developed. However, there is still a key problem common in most methods, i.e., how to select positive and negative samples. The mainstream MaxIoUAssign has inherent problems, such...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2022-01-01
|
Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9852278/ |
_version_ | 1811281290116202496 |
---|---|
author | Yilong Lv Min Li Yujie He Yu Song Weidong Du |
author_facet | Yilong Lv Min Li Yujie He Yu Song Weidong Du |
author_sort | Yilong Lv |
collection | DOAJ |
description | With the wide application of synthetic aperture radar in maritime surveillance, a ship detection method has been rapidly developed. However, there is still a key problem common in most methods, i.e., how to select positive and negative samples. The mainstream MaxIoUAssign has inherent problems, such as a fixed threshold and rough classification, resulting in the low quality of the positive samples. To solve these problems, we propose a new sample selection method called adaptively center-shape sensitive sample selection. The proposed method introduces shape similarity between proposal boxes and ground truth as one of the evaluation criteria and collaborates with intersection of union (IoU) to measure the quality of the proposal boxes. Meanwhile, the center distance between proposal boxes and ground truth is used to control the influence degree of IoU and shape similarity. In this way, the quality score of the proposal boxes can be determined through IoU, shape similarity, and center position, making sample selection more comprehensive. Additionally, to avoid a fixed threshold, the standard deviation of the quality score is used as a variable to form the adaptive threshold. Finally, we conducted extensive experiments on the benchmark SAR ship detection dataset (SSDD) and high-resolution SAR images datasets (HRSID) datasets. The experimental results demonstrated the superiority of our method. |
first_indexed | 2024-04-13T01:29:56Z |
format | Article |
id | doaj.art-6e602f3399d44ef4a87355a21311690d |
institution | Directory Open Access Journal |
issn | 2151-1535 |
language | English |
last_indexed | 2024-04-13T01:29:56Z |
publishDate | 2022-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
spelling | doaj.art-6e602f3399d44ef4a87355a21311690d2022-12-22T03:08:32ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352022-01-01156752676510.1109/JSTARS.2022.31971849852278Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR ImagesYilong Lv0https://orcid.org/0000-0001-9611-4464Min Li1Yujie He2https://orcid.org/0000-0002-2299-4945Yu Song3Weidong Du4Xi'an Institute of High Technology, Xi'an, ChinaXi'an Institute of High Technology, Xi'an, ChinaXi'an Institute of High Technology, Xi'an, ChinaXi'an Institute of High Technology, Xi'an, ChinaXi'an Institute of High Technology, Xi'an, ChinaWith the wide application of synthetic aperture radar in maritime surveillance, a ship detection method has been rapidly developed. However, there is still a key problem common in most methods, i.e., how to select positive and negative samples. The mainstream MaxIoUAssign has inherent problems, such as a fixed threshold and rough classification, resulting in the low quality of the positive samples. To solve these problems, we propose a new sample selection method called adaptively center-shape sensitive sample selection. The proposed method introduces shape similarity between proposal boxes and ground truth as one of the evaluation criteria and collaborates with intersection of union (IoU) to measure the quality of the proposal boxes. Meanwhile, the center distance between proposal boxes and ground truth is used to control the influence degree of IoU and shape similarity. In this way, the quality score of the proposal boxes can be determined through IoU, shape similarity, and center position, making sample selection more comprehensive. Additionally, to avoid a fixed threshold, the standard deviation of the quality score is used as a variable to form the adaptive threshold. Finally, we conducted extensive experiments on the benchmark SAR ship detection dataset (SSDD) and high-resolution SAR images datasets (HRSID) datasets. The experimental results demonstrated the superiority of our method.https://ieeexplore.ieee.org/document/9852278/Adaptive thresholdcenter distancesamples selectionshape similarityship detectionsynthetic aperture radar (SAR) |
spellingShingle | Yilong Lv Min Li Yujie He Yu Song Weidong Du Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Adaptive threshold center distance samples selection shape similarity ship detection synthetic aperture radar (SAR) |
title | Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR Images |
title_full | Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR Images |
title_fullStr | Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR Images |
title_full_unstemmed | Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR Images |
title_short | Adaptively Center-Shape Sensitive Sample Selection for Ship Detection in SAR Images |
title_sort | adaptively center shape sensitive sample selection for ship detection in sar images |
topic | Adaptive threshold center distance samples selection shape similarity ship detection synthetic aperture radar (SAR) |
url | https://ieeexplore.ieee.org/document/9852278/ |
work_keys_str_mv | AT yilonglv adaptivelycentershapesensitivesampleselectionforshipdetectioninsarimages AT minli adaptivelycentershapesensitivesampleselectionforshipdetectioninsarimages AT yujiehe adaptivelycentershapesensitivesampleselectionforshipdetectioninsarimages AT yusong adaptivelycentershapesensitivesampleselectionforshipdetectioninsarimages AT weidongdu adaptivelycentershapesensitivesampleselectionforshipdetectioninsarimages |