Incorporating Negative Sample Training for Ship Detection Based on Deep Learning
While ship detection using high-resolution optical satellite images plays an important role in various civilian fields—including maritime traffic survey and maritime rescue—it is a difficult task due to influences of the complex background, especially when ships are near to land....
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-02-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/19/3/684 |
id |
doaj-9b502f93db744018a5a4c8c5415ddaf8 |
---|---|
record_format |
Article |
spelling |
doaj-9b502f93db744018a5a4c8c5415ddaf82020-11-24T21:16:11ZengMDPI AGSensors1424-82202019-02-0119368410.3390/s19030684s19030684Incorporating Negative Sample Training for Ship Detection Based on Deep LearningLianru Gao0Yiqun He1Xu Sun2Xiuping Jia3Bing Zhang4Key Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, ChinaSchool of Engineering and Information Technology, The University of New South Wales, Canberra Campus, Canberra, ACT 2006, AustraliaKey Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, ChinaWhile ship detection using high-resolution optical satellite images plays an important role in various civilian fields—including maritime traffic survey and maritime rescue—it is a difficult task due to influences of the complex background, especially when ships are near to land. In current literatures, land masking is generally required before ship detection to avoid many false alarms on land. However, sea⁻land segmentation not only has the risk of segmentation errors, but also requires expertise to adjust parameters. In this study, Faster Region-based Convolutional Neural Network (Faster R-CNN) is applied to detect ships without the need for land masking. We propose an effective training strategy for the Faster R-CNN by incorporating a large number of images containing only terrestrial regions as negative samples without any manual marking, which is different from the selection of negative samples by targeted way in other detection methods. The experiments using Gaofen-1 satellite (GF-1), Gaofen-2 satellite (GF-2), and Jilin-1 satellite (JL-1) images as testing datasets under different ship detection conditions were carried out to evaluate the effectiveness of the proposed strategy in the avoidance of false alarms on land. The results show that the method incorporating negative sample training can largely reduce false alarms in terrestrial areas, and is superior in detection performance, algorithm complexity, and time consumption. Compared with the method based on sea⁻land segmentation, the proposed method achieves the absolute increment of 70% of the F1-measure, when the image contains large land area such as the GF-1 image, and achieves the absolute increment of 42.5% for images with complex harbors and many coastal ships, such as the JL-1 images.https://www.mdpi.com/1424-8220/19/3/684ship detectiondeep learningnegative sample trainingsea–land segmentationhigh-resolution satellite images |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Lianru Gao Yiqun He Xu Sun Xiuping Jia Bing Zhang |
spellingShingle |
Lianru Gao Yiqun He Xu Sun Xiuping Jia Bing Zhang Incorporating Negative Sample Training for Ship Detection Based on Deep Learning Sensors ship detection deep learning negative sample training sea–land segmentation high-resolution satellite images |
author_facet |
Lianru Gao Yiqun He Xu Sun Xiuping Jia Bing Zhang |
author_sort |
Lianru Gao |
title |
Incorporating Negative Sample Training for Ship Detection Based on Deep Learning |
title_short |
Incorporating Negative Sample Training for Ship Detection Based on Deep Learning |
title_full |
Incorporating Negative Sample Training for Ship Detection Based on Deep Learning |
title_fullStr |
Incorporating Negative Sample Training for Ship Detection Based on Deep Learning |
title_full_unstemmed |
Incorporating Negative Sample Training for Ship Detection Based on Deep Learning |
title_sort |
incorporating negative sample training for ship detection based on deep learning |
publisher |
MDPI AG |
series |
Sensors |
issn |
1424-8220 |
publishDate |
2019-02-01 |
description |
While ship detection using high-resolution optical satellite images plays an important role in various civilian fields—including maritime traffic survey and maritime rescue—it is a difficult task due to influences of the complex background, especially when ships are near to land. In current literatures, land masking is generally required before ship detection to avoid many false alarms on land. However, sea⁻land segmentation not only has the risk of segmentation errors, but also requires expertise to adjust parameters. In this study, Faster Region-based Convolutional Neural Network (Faster R-CNN) is applied to detect ships without the need for land masking. We propose an effective training strategy for the Faster R-CNN by incorporating a large number of images containing only terrestrial regions as negative samples without any manual marking, which is different from the selection of negative samples by targeted way in other detection methods. The experiments using Gaofen-1 satellite (GF-1), Gaofen-2 satellite (GF-2), and Jilin-1 satellite (JL-1) images as testing datasets under different ship detection conditions were carried out to evaluate the effectiveness of the proposed strategy in the avoidance of false alarms on land. The results show that the method incorporating negative sample training can largely reduce false alarms in terrestrial areas, and is superior in detection performance, algorithm complexity, and time consumption. Compared with the method based on sea⁻land segmentation, the proposed method achieves the absolute increment of 70% of the F1-measure, when the image contains large land area such as the GF-1 image, and achieves the absolute increment of 42.5% for images with complex harbors and many coastal ships, such as the JL-1 images. |
topic |
ship detection deep learning negative sample training sea–land segmentation high-resolution satellite images |
url |
https://www.mdpi.com/1424-8220/19/3/684 |
work_keys_str_mv |
AT lianrugao incorporatingnegativesampletrainingforshipdetectionbasedondeeplearning AT yiqunhe incorporatingnegativesampletrainingforshipdetectionbasedondeeplearning AT xusun incorporatingnegativesampletrainingforshipdetectionbasedondeeplearning AT xiupingjia incorporatingnegativesampletrainingforshipdetectionbasedondeeplearning AT bingzhang incorporatingnegativesampletrainingforshipdetectionbasedondeeplearning |
_version_ |
1726016631495196672 |