UAV image matching of mountainous terrain using the LoFTR deep learning model
In natural terrain scene UAV image matching, traditional feature point-based methods often have problems such as an unstable number of extracted feature points, difficulty in detecting feature points in weak texture areas, uneven distribution, and low robustness. Deep learning-based image matching m...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-09-01
|
| Series: | Frontiers in Earth Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/feart.2023.1203078/full |
| _version_ | 1797690061675823104 |
|---|---|
| author | Huilin Zong Huilin Zong Xiping Yuan Xiping Yuan Shu Gan Shu Gan Xiaolun Zhang Minglong Yang Jie Lv |
| author_facet | Huilin Zong Huilin Zong Xiping Yuan Xiping Yuan Shu Gan Shu Gan Xiaolun Zhang Minglong Yang Jie Lv |
| author_sort | Huilin Zong |
| collection | DOAJ |
| description | In natural terrain scene UAV image matching, traditional feature point-based methods often have problems such as an unstable number of extracted feature points, difficulty in detecting feature points in weak texture areas, uneven distribution, and low robustness. Deep learning-based image matching methods can produce larger and more reasonably distributed matching pairs, so this research paper tries to perform UAV image matching based on a deep learning LoFTR algorithm for natural terrain scenes. The critical technical process was: first, the LoFTR algorithm was used to generate dense feature matching, and then the epipolar line constraints were used to purify the interior points, specifically, this study used the MAGSAC++ method to estimate the fundamental matrix, eliminate the wrong matching pairs, and finally get reliable matching results. In this research paper, six sets of visible images taken by different UAVs equipped with different sensors in the field were selected as experimental data to test the method and were compared and analyzed with the traditional classical SIFT, ASIFT, and AKAZE algorithms and the KeyNet-AdaLAM deep learning method. The experimental results show that the method in this study obtains a dense number of robust matching pairs with uniform spatial distribution in the UAV image matching of natural scenes mainly in mountainous areas, and the comprehensive performance is higher and more advantageous than the comparison methods. |
| first_indexed | 2024-03-12T01:54:07Z |
| format | Article |
| id | doaj.art-5127e884a0b84e0eaa7d7036c9308b1a |
| institution | Directory Open Access Journal |
| issn | 2296-6463 |
| language | English |
| last_indexed | 2024-03-12T01:54:07Z |
| publishDate | 2023-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Earth Science |
| spelling | doaj.art-5127e884a0b84e0eaa7d7036c9308b1a2023-09-08T07:24:44ZengFrontiers Media S.A.Frontiers in Earth Science2296-64632023-09-011110.3389/feart.2023.12030781203078UAV image matching of mountainous terrain using the LoFTR deep learning modelHuilin Zong0Huilin Zong1Xiping Yuan2Xiping Yuan3Shu Gan4Shu Gan5Xiaolun Zhang6Minglong Yang7Jie Lv8Faculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming, Yunnan, ChinaPlateau Mountain Spatial Information Survey Technique Application Engineering Research Center at Yunnan Provinces Universities, Kunming, Yunnan, ChinaPlateau Mountain Spatial Information Survey Technique Application Engineering Research Center at Yunnan Provinces Universities, Kunming, Yunnan, ChinaCollege of Geosciences and Engineering, West Yunnan University of Applied Sciences, Dali, Yunnan, ChinaFaculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming, Yunnan, ChinaPlateau Mountain Spatial Information Survey Technique Application Engineering Research Center at Yunnan Provinces Universities, Kunming, Yunnan, ChinaFaculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming, Yunnan, ChinaFaculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming, Yunnan, ChinaFaculty of Land Resource Engineering, Kunming University of Science and Technology, Kunming, Yunnan, ChinaIn natural terrain scene UAV image matching, traditional feature point-based methods often have problems such as an unstable number of extracted feature points, difficulty in detecting feature points in weak texture areas, uneven distribution, and low robustness. Deep learning-based image matching methods can produce larger and more reasonably distributed matching pairs, so this research paper tries to perform UAV image matching based on a deep learning LoFTR algorithm for natural terrain scenes. The critical technical process was: first, the LoFTR algorithm was used to generate dense feature matching, and then the epipolar line constraints were used to purify the interior points, specifically, this study used the MAGSAC++ method to estimate the fundamental matrix, eliminate the wrong matching pairs, and finally get reliable matching results. In this research paper, six sets of visible images taken by different UAVs equipped with different sensors in the field were selected as experimental data to test the method and were compared and analyzed with the traditional classical SIFT, ASIFT, and AKAZE algorithms and the KeyNet-AdaLAM deep learning method. The experimental results show that the method in this study obtains a dense number of robust matching pairs with uniform spatial distribution in the UAV image matching of natural scenes mainly in mountainous areas, and the comprehensive performance is higher and more advantageous than the comparison methods.https://www.frontiersin.org/articles/10.3389/feart.2023.1203078/fulldeep learningUAV remote sensing imageimage matchingLoFTR algorithmattention mechanisminline point purification |
| spellingShingle | Huilin Zong Huilin Zong Xiping Yuan Xiping Yuan Shu Gan Shu Gan Xiaolun Zhang Minglong Yang Jie Lv UAV image matching of mountainous terrain using the LoFTR deep learning model Frontiers in Earth Science deep learning UAV remote sensing image image matching LoFTR algorithm attention mechanism inline point purification |
| title | UAV image matching of mountainous terrain using the LoFTR deep learning model |
| title_full | UAV image matching of mountainous terrain using the LoFTR deep learning model |
| title_fullStr | UAV image matching of mountainous terrain using the LoFTR deep learning model |
| title_full_unstemmed | UAV image matching of mountainous terrain using the LoFTR deep learning model |
| title_short | UAV image matching of mountainous terrain using the LoFTR deep learning model |
| title_sort | uav image matching of mountainous terrain using the loftr deep learning model |
| topic | deep learning UAV remote sensing image image matching LoFTR algorithm attention mechanism inline point purification |
| url | https://www.frontiersin.org/articles/10.3389/feart.2023.1203078/full |
| work_keys_str_mv | AT huilinzong uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT huilinzong uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT xipingyuan uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT xipingyuan uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT shugan uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT shugan uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT xiaolunzhang uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT minglongyang uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel AT jielv uavimagematchingofmountainousterrainusingtheloftrdeeplearningmodel |