An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery
Lunar craters and rilles are significant topographic features on the lunar surface that will play an essential role in future research on space energy resources and geological evolution. However, previous studies have shown low efficiency in detecting lunar impact craters and poor accuracy in detect...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2022
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs14061391 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/14/6/1391/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/14/6/1391/ 2023-08-20T04:08:38+02:00 An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery Yutong Jia Lei Liu Siqing Peng Mingyang Feng Gang Wan agris 2022-03-13 application/pdf https://doi.org/10.3390/rs14061391 EN eng Multidisciplinary Digital Publishing Institute Remote Sensing Image Processing https://dx.doi.org/10.3390/rs14061391 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 14; Issue 6; Pages: 1391 crater detection rilles detection space energy resources GL-HRNet transfer learning density distribution Text 2022 ftmdpi https://doi.org/10.3390/rs14061391 2023-08-01T04:26:48Z Lunar craters and rilles are significant topographic features on the lunar surface that will play an essential role in future research on space energy resources and geological evolution. However, previous studies have shown low efficiency in detecting lunar impact craters and poor accuracy in detecting lunar rilles. There is no complete automated identification method for lunar features to explore space energy resources further. In this paper, we propose a new specific deep-learning method called high-resolution global–local networks (HR-GLNet) to explore craters and rilles and to discover space energy simultaneously. Based on the GLNet network, the ResNet structure in the global branch is replaced by HRNet, and the residual network and FPN are the local branches. Principal loss function and auxiliary loss function are used to aggregate global and local branches. In experiments, the model, combined with transfer learning methods, can accurately detect lunar craters, Mars craters, and lunar rilles. Compared with other networks, such as UNet, ERU-Net, HRNet, and GLNet, GL-HRNet has a higher accuracy (88.7 ± 8.9) and recall rate (80.1 ± 2.7) in lunar impact crater detection. In addition, the mean absolute error (MAE) of the GL-HRNet on global and local branches is 0.0612 and 0.0429, which are better than the GLNet in terms of segmentation accuracy and MAE. Finally, by analyzing the density distribution of lunar impact craters with a diameter of less than 5 km, it was found that: (i) small impact craters in a local area of the lunar north pole and highland (5°–85°E, 25°–50°S) show apparent high density, and (ii) the density of impact craters in the Orientale Basin is not significantly different from that in the surrounding areas, which is the direction for future geological research. Text North Pole MDPI Open Access Publishing North Pole Remote Sensing 14 6 1391 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
crater detection rilles detection space energy resources GL-HRNet transfer learning density distribution |
spellingShingle |
crater detection rilles detection space energy resources GL-HRNet transfer learning density distribution Yutong Jia Lei Liu Siqing Peng Mingyang Feng Gang Wan An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery |
topic_facet |
crater detection rilles detection space energy resources GL-HRNet transfer learning density distribution |
description |
Lunar craters and rilles are significant topographic features on the lunar surface that will play an essential role in future research on space energy resources and geological evolution. However, previous studies have shown low efficiency in detecting lunar impact craters and poor accuracy in detecting lunar rilles. There is no complete automated identification method for lunar features to explore space energy resources further. In this paper, we propose a new specific deep-learning method called high-resolution global–local networks (HR-GLNet) to explore craters and rilles and to discover space energy simultaneously. Based on the GLNet network, the ResNet structure in the global branch is replaced by HRNet, and the residual network and FPN are the local branches. Principal loss function and auxiliary loss function are used to aggregate global and local branches. In experiments, the model, combined with transfer learning methods, can accurately detect lunar craters, Mars craters, and lunar rilles. Compared with other networks, such as UNet, ERU-Net, HRNet, and GLNet, GL-HRNet has a higher accuracy (88.7 ± 8.9) and recall rate (80.1 ± 2.7) in lunar impact crater detection. In addition, the mean absolute error (MAE) of the GL-HRNet on global and local branches is 0.0612 and 0.0429, which are better than the GLNet in terms of segmentation accuracy and MAE. Finally, by analyzing the density distribution of lunar impact craters with a diameter of less than 5 km, it was found that: (i) small impact craters in a local area of the lunar north pole and highland (5°–85°E, 25°–50°S) show apparent high density, and (ii) the density of impact craters in the Orientale Basin is not significantly different from that in the surrounding areas, which is the direction for future geological research. |
format |
Text |
author |
Yutong Jia Lei Liu Siqing Peng Mingyang Feng Gang Wan |
author_facet |
Yutong Jia Lei Liu Siqing Peng Mingyang Feng Gang Wan |
author_sort |
Yutong Jia |
title |
An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery |
title_short |
An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery |
title_full |
An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery |
title_fullStr |
An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery |
title_full_unstemmed |
An Efficient High-Resolution Global–Local Network to Detect Lunar Features for Space Energy Discovery |
title_sort |
efficient high-resolution global–local network to detect lunar features for space energy discovery |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2022 |
url |
https://doi.org/10.3390/rs14061391 |
op_coverage |
agris |
geographic |
North Pole |
geographic_facet |
North Pole |
genre |
North Pole |
genre_facet |
North Pole |
op_source |
Remote Sensing; Volume 14; Issue 6; Pages: 1391 |
op_relation |
Remote Sensing Image Processing https://dx.doi.org/10.3390/rs14061391 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs14061391 |
container_title |
Remote Sensing |
container_volume |
14 |
container_issue |
6 |
container_start_page |
1391 |
_version_ |
1774721004188205056 |