SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image Registration
Accurate registration between synthetic aperture radar (SAR) images is the basis for high-precision geometric correction of SAR images. The feature points extracted by conventional feature extraction methods are unsatisfactory, which are affected by imaging geometric characteristics and speckle nois...
| 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/9850384/ |
| _version_ | 1818486405315166208 |
|---|---|
| author | Xin Li Taoyang Wang Hao Cui Guo Zhang Qian Cheng Tiancheng Dong Boyang Jiang |
| author_facet | Xin Li Taoyang Wang Hao Cui Guo Zhang Qian Cheng Tiancheng Dong Boyang Jiang |
| author_sort | Xin Li |
| collection | DOAJ |
| description | Accurate registration between synthetic aperture radar (SAR) images is the basis for high-precision geometric correction of SAR images. The feature points extracted by conventional feature extraction methods are unsatisfactory, which are affected by imaging geometric characteristics and speckle noise of SAR images. This article innovatively proposes a spaceborne SAR image feature learning framework to realize automatic sample generation and model training. It mainly includes two modules: The feature sample generation module based on the initial geometric information of spaceborne SAR. The initial rational polynomial coefficient (RPC) parameters of the spaceborne SAR are adopted to realize the initial positioning of the SAR image, and a variety of feature extraction operators are used to match the overlapping areas to obtain high-precision matching points, which are employed as training samples for image pairs; pseudo-Siamese feature learning network SARPointNet for SAR image feature learning. The pseudo-Siamese network is used to extract the feature points and descriptors of the sample image pairs. The feature optimization process is realized through the descriptor constraints between image pairs, which promotes the network to improve the accuracy of feature extraction. The proposed method has been tested in mountain, hilly, flatland, and urban scenarios, respectively. The results demonstrate that the correspondence points extracted by SARPointNet are evenly distributed, are in large quantities (at least ten times that of other methods), and achieve high precision (the root mean square error is less than 1 pixels), which shows great advantages over other methods. |
| first_indexed | 2024-12-10T16:22:31Z |
| format | Article |
| id | doaj.art-55dab022c484480babfdd50fec40f88d |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-10T16:22:31Z |
| 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-55dab022c484480babfdd50fec40f88d2022-12-22T01:41:45ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352022-01-01156371638110.1109/JSTARS.2022.31963839850384SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image RegistrationXin Li0https://orcid.org/0000-0002-2569-7380Taoyang Wang1https://orcid.org/0000-0002-6014-5354Hao Cui2https://orcid.org/0000-0001-7111-8901Guo Zhang3https://orcid.org/0000-0002-3987-5336Qian Cheng4https://orcid.org/0000-0003-2902-7322Tiancheng Dong5Boyang Jiang6State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaAccurate registration between synthetic aperture radar (SAR) images is the basis for high-precision geometric correction of SAR images. The feature points extracted by conventional feature extraction methods are unsatisfactory, which are affected by imaging geometric characteristics and speckle noise of SAR images. This article innovatively proposes a spaceborne SAR image feature learning framework to realize automatic sample generation and model training. It mainly includes two modules: The feature sample generation module based on the initial geometric information of spaceborne SAR. The initial rational polynomial coefficient (RPC) parameters of the spaceborne SAR are adopted to realize the initial positioning of the SAR image, and a variety of feature extraction operators are used to match the overlapping areas to obtain high-precision matching points, which are employed as training samples for image pairs; pseudo-Siamese feature learning network SARPointNet for SAR image feature learning. The pseudo-Siamese network is used to extract the feature points and descriptors of the sample image pairs. The feature optimization process is realized through the descriptor constraints between image pairs, which promotes the network to improve the accuracy of feature extraction. The proposed method has been tested in mountain, hilly, flatland, and urban scenarios, respectively. The results demonstrate that the correspondence points extracted by SARPointNet are evenly distributed, are in large quantities (at least ten times that of other methods), and achieve high precision (the root mean square error is less than 1 pixels), which shows great advantages over other methods.https://ieeexplore.ieee.org/document/9850384/Convolutional neural network (CNN)rational polynomial coefficients (RPCs)registrationsynthetic aperture radar (SAR) image |
| spellingShingle | Xin Li Taoyang Wang Hao Cui Guo Zhang Qian Cheng Tiancheng Dong Boyang Jiang SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image Registration IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural network (CNN) rational polynomial coefficients (RPCs) registration synthetic aperture radar (SAR) image |
| title | SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image Registration |
| title_full | SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image Registration |
| title_fullStr | SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image Registration |
| title_full_unstemmed | SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image Registration |
| title_short | SARPointNet: An Automated Feature Learning Framework for Spaceborne SAR Image Registration |
| title_sort | sarpointnet an automated feature learning framework for spaceborne sar image registration |
| topic | Convolutional neural network (CNN) rational polynomial coefficients (RPCs) registration synthetic aperture radar (SAR) image |
| url | https://ieeexplore.ieee.org/document/9850384/ |
| work_keys_str_mv | AT xinli sarpointnetanautomatedfeaturelearningframeworkforspacebornesarimageregistration AT taoyangwang sarpointnetanautomatedfeaturelearningframeworkforspacebornesarimageregistration AT haocui sarpointnetanautomatedfeaturelearningframeworkforspacebornesarimageregistration AT guozhang sarpointnetanautomatedfeaturelearningframeworkforspacebornesarimageregistration AT qiancheng sarpointnetanautomatedfeaturelearningframeworkforspacebornesarimageregistration AT tianchengdong sarpointnetanautomatedfeaturelearningframeworkforspacebornesarimageregistration AT boyangjiang sarpointnetanautomatedfeaturelearningframeworkforspacebornesarimageregistration |