STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite Data
Satellite data is of high importance for ocean environment monitoring and protection. However, due to the missing values in satellite data, caused by various force majeure factors such as cloud cover, bad weather and sensor failure, the quality of satellite data is reduced greatly, which hinders the...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/1/88 |
| _version_ | 1797439855230189568 |
|---|---|
| author | Shuyu Wang Wengen Li Siyun Hou Jihong Guan Jiamin Yao |
| author_facet | Shuyu Wang Wengen Li Siyun Hou Jihong Guan Jiamin Yao |
| author_sort | Shuyu Wang |
| collection | DOAJ |
| description | Satellite data is of high importance for ocean environment monitoring and protection. However, due to the missing values in satellite data, caused by various force majeure factors such as cloud cover, bad weather and sensor failure, the quality of satellite data is reduced greatly, which hinders the applications of satellite data in practice. Therefore, a variety of methods have been proposed to conduct missing data imputation for satellite data to improve its quality. However, these methods cannot well learn the short-term temporal dependence and dynamic spatial dependence in satellite data, resulting in bad imputation performance when the data missing rate is large. To address this issue, we propose the Spatio-Temporal Attention Generative Adversarial Network (STA-GAN) for missing value imputation in satellite data. First, we develop the Spatio-Temporal Attention (STA) mechanism based on Graph Attention Network (GAT) to learn features for capturing both short-term temporal dependence and dynamic spatial dependence in satellite data. Then, the learned features from STA are fused to enrich the spatio-temporal information for training the generator and discriminator of STA-GAN. Finally, we use the generated imputation data by the trained generator of STA-GAN to fill the missing values in satellite data. Experimental results on real datasets show that STA-GAN largely outperforms the baseline data imputation methods, especially for filling satellite data with large missing rates. |
| first_indexed | 2024-03-09T11:59:45Z |
| format | Article |
| id | doaj.art-938965bd96e44fbc8c2b9f6807afad18 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T11:59:45Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-938965bd96e44fbc8c2b9f6807afad182023-11-30T23:05:28ZengMDPI AGRemote Sensing2072-42922022-12-011518810.3390/rs15010088STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite DataShuyu Wang0Wengen Li1Siyun Hou2Jihong Guan3Jiamin Yao4Department of Computer Science and Technology, Tongji University, Shanghai 200082, ChinaDepartment of Computer Science and Technology, Tongji University, Shanghai 200082, ChinaDepartment of Computer Science and Technology, Tongji University, Shanghai 200082, ChinaDepartment of Computer Science and Technology, Tongji University, Shanghai 200082, ChinaDepartment of Computer Science and Technology, Tongji University, Shanghai 200082, ChinaSatellite data is of high importance for ocean environment monitoring and protection. However, due to the missing values in satellite data, caused by various force majeure factors such as cloud cover, bad weather and sensor failure, the quality of satellite data is reduced greatly, which hinders the applications of satellite data in practice. Therefore, a variety of methods have been proposed to conduct missing data imputation for satellite data to improve its quality. However, these methods cannot well learn the short-term temporal dependence and dynamic spatial dependence in satellite data, resulting in bad imputation performance when the data missing rate is large. To address this issue, we propose the Spatio-Temporal Attention Generative Adversarial Network (STA-GAN) for missing value imputation in satellite data. First, we develop the Spatio-Temporal Attention (STA) mechanism based on Graph Attention Network (GAT) to learn features for capturing both short-term temporal dependence and dynamic spatial dependence in satellite data. Then, the learned features from STA are fused to enrich the spatio-temporal information for training the generator and discriminator of STA-GAN. Finally, we use the generated imputation data by the trained generator of STA-GAN to fill the missing values in satellite data. Experimental results on real datasets show that STA-GAN largely outperforms the baseline data imputation methods, especially for filling satellite data with large missing rates.https://www.mdpi.com/2072-4292/15/1/88satellite datadata imputationspatio-temporal analyticsgenerative adversarial network |
| spellingShingle | Shuyu Wang Wengen Li Siyun Hou Jihong Guan Jiamin Yao STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite Data Remote Sensing satellite data data imputation spatio-temporal analytics generative adversarial network |
| title | STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite Data |
| title_full | STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite Data |
| title_fullStr | STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite Data |
| title_full_unstemmed | STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite Data |
| title_short | STA-GAN: A Spatio-Temporal Attention Generative Adversarial Network for Missing Value Imputation in Satellite Data |
| title_sort | sta gan a spatio temporal attention generative adversarial network for missing value imputation in satellite data |
| topic | satellite data data imputation spatio-temporal analytics generative adversarial network |
| url | https://www.mdpi.com/2072-4292/15/1/88 |
| work_keys_str_mv | AT shuyuwang staganaspatiotemporalattentiongenerativeadversarialnetworkformissingvalueimputationinsatellitedata AT wengenli staganaspatiotemporalattentiongenerativeadversarialnetworkformissingvalueimputationinsatellitedata AT siyunhou staganaspatiotemporalattentiongenerativeadversarialnetworkformissingvalueimputationinsatellitedata AT jihongguan staganaspatiotemporalattentiongenerativeadversarialnetworkformissingvalueimputationinsatellitedata AT jiaminyao staganaspatiotemporalattentiongenerativeadversarialnetworkformissingvalueimputationinsatellitedata |