Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene Classification
Scene classification is one of the fundamental techniques shared by many basic remote sensing tasks with a wide range of applications. As the demands of catering with situations under high variance in the data urgent conditions are rising, a research topic called few-shot scene classification is rec...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/3/485 |
| _version_ | 1797485140840022016 |
|---|---|
| author | Feimo Li Shuaibo Li Xinxin Fan Xiong Li Hongxing Chang |
| author_facet | Feimo Li Shuaibo Li Xinxin Fan Xiong Li Hongxing Chang |
| author_sort | Feimo Li |
| collection | DOAJ |
| description | Scene classification is one of the fundamental techniques shared by many basic remote sensing tasks with a wide range of applications. As the demands of catering with situations under high variance in the data urgent conditions are rising, a research topic called few-shot scene classification is receiving more interest with a focus on building classification model from few training samples. Currently, methods using the meta-learning principle or graphical models are achieving state-of-art performances. However, there are still significant gaps in between the few-shot methods and the traditionally trained ones, as there are implicit data isolations in standard meta-learning procedure and less-flexibility in the static graph neural network modeling technique, which largely limit the data-to-knowledge transition efficiency. To address these issues, this paper proposed an novel few-shot scene classification algorithm based on a different meta-learning principle called continual meta-learning, which enhances the inter-task correlation by fusing more historical prior knowledge from a sequence of tasks within sections of meta-training or meta-testing periods. Moreover, as to increase the discriminative power between classes, a graph transformer is introduced to produce the structural attention, which can optimize the distribution of sample features in the embedded space and promotes the overall classification capability of the model. The advantages of our proposed algorithm are verified by comparing with nine state-of-art meta-learning based on few-shot scene classification on three popular datasets, where a minimum of a 9% increase in accuracy can be observed. Furthermore, the efficiency of the newly added modular modifications have also be verified by comparing to the continual meta-learning baseline. |
| first_indexed | 2024-03-09T23:14:32Z |
| format | Article |
| id | doaj.art-4ef8e4325ea0415a9891767bd11b9464 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T23:14:32Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-4ef8e4325ea0415a9891767bd11b94642023-11-23T17:38:28ZengMDPI AGRemote Sensing2072-42922022-01-0114348510.3390/rs14030485Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene ClassificationFeimo Li0Shuaibo Li1Xinxin Fan2Xiong Li3Hongxing Chang4Institute of Automation Chinese Academy of Sciences, 95 Zhongguancun East Road, Haidian District, Beijing 100190, ChinaSchool of Information, Central University of Finance and Economics, Shunsha Road, Shahe Higher Education Park, Changping District, Beijing 102206, ChinaCollege of Software, Beihang University, 37 Xueyuan Road, Haidian District, Beijing 100190, ChinaSchool of Mechanical Electronic and Information Engineering, China University of Mining and Technology, Ding No.11 Xueyuan Road, Haidian District, Beijing 100083, ChinaInstitute of Automation Chinese Academy of Sciences, 95 Zhongguancun East Road, Haidian District, Beijing 100190, ChinaScene classification is one of the fundamental techniques shared by many basic remote sensing tasks with a wide range of applications. As the demands of catering with situations under high variance in the data urgent conditions are rising, a research topic called few-shot scene classification is receiving more interest with a focus on building classification model from few training samples. Currently, methods using the meta-learning principle or graphical models are achieving state-of-art performances. However, there are still significant gaps in between the few-shot methods and the traditionally trained ones, as there are implicit data isolations in standard meta-learning procedure and less-flexibility in the static graph neural network modeling technique, which largely limit the data-to-knowledge transition efficiency. To address these issues, this paper proposed an novel few-shot scene classification algorithm based on a different meta-learning principle called continual meta-learning, which enhances the inter-task correlation by fusing more historical prior knowledge from a sequence of tasks within sections of meta-training or meta-testing periods. Moreover, as to increase the discriminative power between classes, a graph transformer is introduced to produce the structural attention, which can optimize the distribution of sample features in the embedded space and promotes the overall classification capability of the model. The advantages of our proposed algorithm are verified by comparing with nine state-of-art meta-learning based on few-shot scene classification on three popular datasets, where a minimum of a 9% increase in accuracy can be observed. Furthermore, the efficiency of the newly added modular modifications have also be verified by comparing to the continual meta-learning baseline.https://www.mdpi.com/2072-4292/14/3/485remote sensing scene classificationfew shot learningcontinual meta-learninggraph transformer |
| spellingShingle | Feimo Li Shuaibo Li Xinxin Fan Xiong Li Hongxing Chang Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene Classification Remote Sensing remote sensing scene classification few shot learning continual meta-learning graph transformer |
| title | Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene Classification |
| title_full | Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene Classification |
| title_fullStr | Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene Classification |
| title_full_unstemmed | Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene Classification |
| title_short | Structural Attention Enhanced Continual Meta-Learning for Graph Edge Labeling Based Few-Shot Remote Sensing Scene Classification |
| title_sort | structural attention enhanced continual meta learning for graph edge labeling based few shot remote sensing scene classification |
| topic | remote sensing scene classification few shot learning continual meta-learning graph transformer |
| url | https://www.mdpi.com/2072-4292/14/3/485 |
| work_keys_str_mv | AT feimoli structuralattentionenhancedcontinualmetalearningforgraphedgelabelingbasedfewshotremotesensingsceneclassification AT shuaiboli structuralattentionenhancedcontinualmetalearningforgraphedgelabelingbasedfewshotremotesensingsceneclassification AT xinxinfan structuralattentionenhancedcontinualmetalearningforgraphedgelabelingbasedfewshotremotesensingsceneclassification AT xiongli structuralattentionenhancedcontinualmetalearningforgraphedgelabelingbasedfewshotremotesensingsceneclassification AT hongxingchang structuralattentionenhancedcontinualmetalearningforgraphedgelabelingbasedfewshotremotesensingsceneclassification |