An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 Imagery
Crop-type mapping is the foundation of grain security and digital agricultural management. Accuracy, efficiency and large-scale scene consistency are required to perform crop classification from remote sensing images. Many current remote-sensing crop extraction methods based on deep learning cannot...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/8/1990 |
| _version_ | 1797603649111719936 |
|---|---|
| author | Xiangyu Tian Yongqing Bai Guoqing Li Xuan Yang Jianxi Huang Zhengchao Chen |
| author_facet | Xiangyu Tian Yongqing Bai Guoqing Li Xuan Yang Jianxi Huang Zhengchao Chen |
| author_sort | Xiangyu Tian |
| collection | DOAJ |
| description | Crop-type mapping is the foundation of grain security and digital agricultural management. Accuracy, efficiency and large-scale scene consistency are required to perform crop classification from remote sensing images. Many current remote-sensing crop extraction methods based on deep learning cannot account for adaptation effects in large-scale, complex scenes. Therefore, this study proposes a novel adaptive feature-fusion network for crop classification using single-temporal Sentinel-2 images. The selective patch module implemented in the network can adaptively integrate the features of different patch sizes to assess complex scenes better. TabNet was used simultaneously to extract spectral information from the center pixels of the patches. Multitask learning was used to supervise the extraction process to improve the weight of the spectral characteristics while mitigating the negative impact of a small sample size. In the network, superpixel optimization was applied to post-process the classification results to improve the crop edges. By conducting the crop classification of peanut, rice, and corn based on Sentinel-2 images in 2022 in Henan Province, China, the novel method proposed in this paper was more accurate, indicated by an F1 score of 96.53%, than other mainstream methods. This indicates our model’s potential for application in crop classification in large scenes. |
| first_indexed | 2024-03-11T04:34:59Z |
| format | Article |
| id | doaj.art-63f113c165504fd789f3ee43b3957ed9 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T04:34:59Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-63f113c165504fd789f3ee43b3957ed92023-11-17T21:10:28ZengMDPI AGRemote Sensing2072-42922023-04-01158199010.3390/rs15081990An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 ImageryXiangyu Tian0Yongqing Bai1Guoqing Li2Xuan Yang3Jianxi Huang4Zhengchao Chen5Airborne Remote Sensing Center, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAirborne Remote Sensing Center, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaHenan Institute of Remote Sensing and Geomatics, Zhengzhou 450003, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaCollege of Land Science and Technology, China Agricultural University, Beijing 100083, ChinaAirborne Remote Sensing Center, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaCrop-type mapping is the foundation of grain security and digital agricultural management. Accuracy, efficiency and large-scale scene consistency are required to perform crop classification from remote sensing images. Many current remote-sensing crop extraction methods based on deep learning cannot account for adaptation effects in large-scale, complex scenes. Therefore, this study proposes a novel adaptive feature-fusion network for crop classification using single-temporal Sentinel-2 images. The selective patch module implemented in the network can adaptively integrate the features of different patch sizes to assess complex scenes better. TabNet was used simultaneously to extract spectral information from the center pixels of the patches. Multitask learning was used to supervise the extraction process to improve the weight of the spectral characteristics while mitigating the negative impact of a small sample size. In the network, superpixel optimization was applied to post-process the classification results to improve the crop edges. By conducting the crop classification of peanut, rice, and corn based on Sentinel-2 images in 2022 in Henan Province, China, the novel method proposed in this paper was more accurate, indicated by an F1 score of 96.53%, than other mainstream methods. This indicates our model’s potential for application in crop classification in large scenes.https://www.mdpi.com/2072-4292/15/8/1990crop mappingdeep learningfeature fusionmultitask learningSentinel-2 |
| spellingShingle | Xiangyu Tian Yongqing Bai Guoqing Li Xuan Yang Jianxi Huang Zhengchao Chen An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 Imagery Remote Sensing crop mapping deep learning feature fusion multitask learning Sentinel-2 |
| title | An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 Imagery |
| title_full | An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 Imagery |
| title_fullStr | An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 Imagery |
| title_full_unstemmed | An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 Imagery |
| title_short | An Adaptive Feature Fusion Network with Superpixel Optimization for Crop Classification Using Sentinel-2 Imagery |
| title_sort | adaptive feature fusion network with superpixel optimization for crop classification using sentinel 2 imagery |
| topic | crop mapping deep learning feature fusion multitask learning Sentinel-2 |
| url | https://www.mdpi.com/2072-4292/15/8/1990 |
| work_keys_str_mv | AT xiangyutian anadaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT yongqingbai anadaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT guoqingli anadaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT xuanyang anadaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT jianxihuang anadaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT zhengchaochen anadaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT xiangyutian adaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT yongqingbai adaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT guoqingli adaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT xuanyang adaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT jianxihuang adaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery AT zhengchaochen adaptivefeaturefusionnetworkwithsuperpixeloptimizationforcropclassificationusingsentinel2imagery |