Few-shot SAR target classification via meta-learning with hybrid models
Currently, in Synthetic Aperture Radar Automatic Target Recognition (SAR ATR), few-shot methods can save cost and resources while enhancing adaptability. However, due to the limitations of SAR imaging environments and observation conditions, obtaining a large amount of high-value target data is chal...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-11-01
|
| Series: | Frontiers in Earth Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/feart.2024.1469032/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846163470154727424 |
|---|---|
| author | Qingtian Geng Yaning Wang Qingliang Li |
| author_facet | Qingtian Geng Yaning Wang Qingliang Li |
| author_sort | Qingtian Geng |
| collection | DOAJ |
| description | Currently, in Synthetic Aperture Radar Automatic Target Recognition (SAR ATR), few-shot methods can save cost and resources while enhancing adaptability. However, due to the limitations of SAR imaging environments and observation conditions, obtaining a large amount of high-value target data is challenging, leading to a severe shortage of datasets. This paper proposes the use of an Adaptive Dynamic Weight Hybrid Model (ADW-HM) meta-learning framework to address the problem of poor recognition accuracy for unknown classes caused by sample constraints. By dynamically weighting and learning model parameters independently, the framework dynamically integrates model results to improve recognition accuracy for unknown classes. Experiments conducted on the TASK-MSTAR and OpenSARShip datasets demonstrate that the ADW-HM framework can obtain more comprehensive and integrated feature representations, reduce overfitting, and enhance generalization capability for unknown classes. The accuracy is improved in both 1-shot and 5-shot scenarios, indicating that ADW-HM is feasible for addressing few-shot problems. |
| format | Article |
| id | doaj-art-160754cd0f98444cb15dc77ee9c1a14d |
| institution | Kabale University |
| issn | 2296-6463 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Earth Science |
| spelling | doaj-art-160754cd0f98444cb15dc77ee9c1a14d2024-11-19T06:15:20ZengFrontiers Media S.A.Frontiers in Earth Science2296-64632024-11-011210.3389/feart.2024.14690321469032Few-shot SAR target classification via meta-learning with hybrid modelsQingtian GengYaning WangQingliang LiCurrently, in Synthetic Aperture Radar Automatic Target Recognition (SAR ATR), few-shot methods can save cost and resources while enhancing adaptability. However, due to the limitations of SAR imaging environments and observation conditions, obtaining a large amount of high-value target data is challenging, leading to a severe shortage of datasets. This paper proposes the use of an Adaptive Dynamic Weight Hybrid Model (ADW-HM) meta-learning framework to address the problem of poor recognition accuracy for unknown classes caused by sample constraints. By dynamically weighting and learning model parameters independently, the framework dynamically integrates model results to improve recognition accuracy for unknown classes. Experiments conducted on the TASK-MSTAR and OpenSARShip datasets demonstrate that the ADW-HM framework can obtain more comprehensive and integrated feature representations, reduce overfitting, and enhance generalization capability for unknown classes. The accuracy is improved in both 1-shot and 5-shot scenarios, indicating that ADW-HM is feasible for addressing few-shot problems.https://www.frontiersin.org/articles/10.3389/feart.2024.1469032/fullfew-shot learning (FSL)adaptive dynamic weight hybrid modelsynthetic aperture radarautomatic target recognitionmeta-learning |
| spellingShingle | Qingtian Geng Yaning Wang Qingliang Li Few-shot SAR target classification via meta-learning with hybrid models Frontiers in Earth Science few-shot learning (FSL) adaptive dynamic weight hybrid model synthetic aperture radar automatic target recognition meta-learning |
| title | Few-shot SAR target classification via meta-learning with hybrid models |
| title_full | Few-shot SAR target classification via meta-learning with hybrid models |
| title_fullStr | Few-shot SAR target classification via meta-learning with hybrid models |
| title_full_unstemmed | Few-shot SAR target classification via meta-learning with hybrid models |
| title_short | Few-shot SAR target classification via meta-learning with hybrid models |
| title_sort | few shot sar target classification via meta learning with hybrid models |
| topic | few-shot learning (FSL) adaptive dynamic weight hybrid model synthetic aperture radar automatic target recognition meta-learning |
| url | https://www.frontiersin.org/articles/10.3389/feart.2024.1469032/full |
| work_keys_str_mv | AT qingtiangeng fewshotsartargetclassificationviametalearningwithhybridmodels AT yaningwang fewshotsartargetclassificationviametalearningwithhybridmodels AT qingliangli fewshotsartargetclassificationviametalearningwithhybridmodels |