Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection
As an active microwave imaging sensor for the high-resolution earth observation, synthetic aperture radar (SAR) has been extensively applied in military, agriculture, geology, ecology, oceanography, etc., due to its prominent advantages of all-weather and all-time working capacity. Especially, in th...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2019
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs11212483 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/11/21/2483/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/11/21/2483/ 2023-08-20T04:09:46+02:00 Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection Zhang Shi Wei agris 2019-10-24 application/pdf https://doi.org/10.3390/rs11212483 EN eng Multidisciplinary Digital Publishing Institute Remote Sensing Image Processing https://dx.doi.org/10.3390/rs11212483 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 11; Issue 21; Pages: 2483 synthetic aperture radar (SAR) ship detection high-speed convolution neural network (CNN) depthwise separable convolution neural network (DS-CNN) depthwise convolution (D-Conv2D) pointwise convolution (P-Conv2D) Text 2019 ftmdpi https://doi.org/10.3390/rs11212483 2023-07-31T22:43:44Z As an active microwave imaging sensor for the high-resolution earth observation, synthetic aperture radar (SAR) has been extensively applied in military, agriculture, geology, ecology, oceanography, etc., due to its prominent advantages of all-weather and all-time working capacity. Especially, in the marine field, SAR can provide numerous high-quality services for fishery management, traffic control, sea-ice monitoring, marine environmental protection, etc. Among them, ship detection in SAR images has attracted more and more attention on account of the urgent requirements of maritime rescue and military strategy formulation. Nowadays, most researches are focusing on improving the ship detection accuracy, while the detection speed is frequently neglected, regardless of traditional feature extraction methods or modern deep learning (DL) methods. However, the high-speed SAR ship detection is of great practical value, because it can provide real-time maritime disaster rescue and emergency military planning. Therefore, in order to address this problem, we proposed a novel high-speed SAR ship detection approach by mainly using depthwise separable convolution neural network (DS-CNN). In this approach, we integrated multi-scale detection mechanism, concatenation mechanism and anchor box mechanism to establish a brand-new light-weight network architecture for the high-speed SAR ship detection. We used DS-CNN, which consists of a depthwise convolution (D-Conv2D) and a pointwise convolution (P-Conv2D), to substitute for the conventional convolution neural network (C-CNN). In this way, the number of network parameters gets obviously decreased, and the ship detection speed gets dramatically improved. We experimented on an open SAR ship detection dataset (SSDD) to validate the correctness and feasibility of the proposed method. To verify the strong migration capacity of our method, we also carried out actual ship detection on a wide-region large-size Sentinel-1 SAR image. Ultimately, under the same hardware platform with ... Text Sea ice MDPI Open Access Publishing Remote Sensing 11 21 2483 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
synthetic aperture radar (SAR) ship detection high-speed convolution neural network (CNN) depthwise separable convolution neural network (DS-CNN) depthwise convolution (D-Conv2D) pointwise convolution (P-Conv2D) |
spellingShingle |
synthetic aperture radar (SAR) ship detection high-speed convolution neural network (CNN) depthwise separable convolution neural network (DS-CNN) depthwise convolution (D-Conv2D) pointwise convolution (P-Conv2D) Zhang Shi Wei Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection |
topic_facet |
synthetic aperture radar (SAR) ship detection high-speed convolution neural network (CNN) depthwise separable convolution neural network (DS-CNN) depthwise convolution (D-Conv2D) pointwise convolution (P-Conv2D) |
description |
As an active microwave imaging sensor for the high-resolution earth observation, synthetic aperture radar (SAR) has been extensively applied in military, agriculture, geology, ecology, oceanography, etc., due to its prominent advantages of all-weather and all-time working capacity. Especially, in the marine field, SAR can provide numerous high-quality services for fishery management, traffic control, sea-ice monitoring, marine environmental protection, etc. Among them, ship detection in SAR images has attracted more and more attention on account of the urgent requirements of maritime rescue and military strategy formulation. Nowadays, most researches are focusing on improving the ship detection accuracy, while the detection speed is frequently neglected, regardless of traditional feature extraction methods or modern deep learning (DL) methods. However, the high-speed SAR ship detection is of great practical value, because it can provide real-time maritime disaster rescue and emergency military planning. Therefore, in order to address this problem, we proposed a novel high-speed SAR ship detection approach by mainly using depthwise separable convolution neural network (DS-CNN). In this approach, we integrated multi-scale detection mechanism, concatenation mechanism and anchor box mechanism to establish a brand-new light-weight network architecture for the high-speed SAR ship detection. We used DS-CNN, which consists of a depthwise convolution (D-Conv2D) and a pointwise convolution (P-Conv2D), to substitute for the conventional convolution neural network (C-CNN). In this way, the number of network parameters gets obviously decreased, and the ship detection speed gets dramatically improved. We experimented on an open SAR ship detection dataset (SSDD) to validate the correctness and feasibility of the proposed method. To verify the strong migration capacity of our method, we also carried out actual ship detection on a wide-region large-size Sentinel-1 SAR image. Ultimately, under the same hardware platform with ... |
format |
Text |
author |
Zhang Shi Wei |
author_facet |
Zhang Shi Wei |
author_sort |
Zhang |
title |
Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection |
title_short |
Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection |
title_full |
Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection |
title_fullStr |
Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection |
title_full_unstemmed |
Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection |
title_sort |
depthwise separable convolution neural network for high-speed sar ship detection |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2019 |
url |
https://doi.org/10.3390/rs11212483 |
op_coverage |
agris |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
Remote Sensing; Volume 11; Issue 21; Pages: 2483 |
op_relation |
Remote Sensing Image Processing https://dx.doi.org/10.3390/rs11212483 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs11212483 |
container_title |
Remote Sensing |
container_volume |
11 |
container_issue |
21 |
container_start_page |
2483 |
_version_ |
1774723447916593152 |