Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images
Classification tasks on land cover (LC) mapping are challenging due to the complex and heterogeneous characteristics of remote sensing images(RSIs). Current LC classifications are mainly based on deep convolutional neural networks (DCNNs), and previous works have been proven that spatial context can...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-03-01
|
| Series: | International Journal of Applied Earth Observations and Geoinformation |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0303243422000320 |
| _version_ | 1828071645340762112 |
|---|---|
| author | Xijie Cheng Xiaohui He Mengjia Qiao Panle Li Shaokai Hu Peng Chang Zhihui Tian |
| author_facet | Xijie Cheng Xiaohui He Mengjia Qiao Panle Li Shaokai Hu Peng Chang Zhihui Tian |
| author_sort | Xijie Cheng |
| collection | DOAJ |
| description | Classification tasks on land cover (LC) mapping are challenging due to the complex and heterogeneous characteristics of remote sensing images(RSIs). Current LC classifications are mainly based on deep convolutional neural networks (DCNNs), and previous works have been proven that spatial context can offer essential cues for performance improvement. However, they still have some drawbacks that limit context capture ability: the ambiguity of global context and lack of efficient context combination strategy. To address these issues, we develop a multilevel LC contextual (MLCC) framework that can adaptively integrate the effective global context with the local context for LC classification. The MLCC framework comprises two modules: a DCNN-based LC classification network (DLCN) and a multilevel context integration module (MCIM). By a well-defined deep network, DLCN could enhance the effective global context feature while weakening the ambiguous representation. Besides, MCIM enables adaptive combinate the global and local context under the guidance of uncertainty map in an efficient way. This collaboratively global–local contextual information further boosts the discriminate feature representation for effective and efficient LC classification. The experiments on LC datasets demonstrate that the proposed MLCC has superior capability in capturing contextual features and thus outperforms the existing methods. |
| first_indexed | 2024-04-11T00:57:09Z |
| format | Article |
| id | doaj.art-f70da8f53f8e4eb881017391c82edf8a |
| institution | Directory Open Access Journal |
| issn | 1569-8432 |
| language | English |
| last_indexed | 2024-04-11T00:57:09Z |
| publishDate | 2022-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Applied Earth Observations and Geoinformation |
| spelling | doaj.art-f70da8f53f8e4eb881017391c82edf8a2023-01-05T04:31:08ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322022-03-01107102706Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing imagesXijie Cheng0Xiaohui He1Mengjia Qiao2Panle Li3Shaokai Hu4Peng Chang5Zhihui Tian6School of Information Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Geoscience and Technology, Zhengzhou University, 450001, China; Ecometeorology Joint Laboratory of Zhengzhou University and Chinese Academy of Meteorological Science, Zhengzhou 450001, China; Corresponding author at: School of Geoscience and Technology, Zhengzhou University, 450001, China.School of Information Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Information Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Information Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Geoscience and Technology, Zhengzhou University, 450001, China; Ecometeorology Joint Laboratory of Zhengzhou University and Chinese Academy of Meteorological Science, Zhengzhou 450001, ChinaSchool of Geoscience and Technology, Zhengzhou University, 450001, China; Ecometeorology Joint Laboratory of Zhengzhou University and Chinese Academy of Meteorological Science, Zhengzhou 450001, ChinaClassification tasks on land cover (LC) mapping are challenging due to the complex and heterogeneous characteristics of remote sensing images(RSIs). Current LC classifications are mainly based on deep convolutional neural networks (DCNNs), and previous works have been proven that spatial context can offer essential cues for performance improvement. However, they still have some drawbacks that limit context capture ability: the ambiguity of global context and lack of efficient context combination strategy. To address these issues, we develop a multilevel LC contextual (MLCC) framework that can adaptively integrate the effective global context with the local context for LC classification. The MLCC framework comprises two modules: a DCNN-based LC classification network (DLCN) and a multilevel context integration module (MCIM). By a well-defined deep network, DLCN could enhance the effective global context feature while weakening the ambiguous representation. Besides, MCIM enables adaptive combinate the global and local context under the guidance of uncertainty map in an efficient way. This collaboratively global–local contextual information further boosts the discriminate feature representation for effective and efficient LC classification. The experiments on LC datasets demonstrate that the proposed MLCC has superior capability in capturing contextual features and thus outperforms the existing methods.http://www.sciencedirect.com/science/article/pii/S0303243422000320Land cover(LC) classificationDeep convolutional neural networks(DCNNs)Contextual informationFeature fusion |
| spellingShingle | Xijie Cheng Xiaohui He Mengjia Qiao Panle Li Shaokai Hu Peng Chang Zhihui Tian Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images International Journal of Applied Earth Observations and Geoinformation Land cover(LC) classification Deep convolutional neural networks(DCNNs) Contextual information Feature fusion |
| title | Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images |
| title_full | Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images |
| title_fullStr | Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images |
| title_full_unstemmed | Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images |
| title_short | Enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images |
| title_sort | enhanced contextual representation with deep neural networks for land cover classification based on remote sensing images |
| topic | Land cover(LC) classification Deep convolutional neural networks(DCNNs) Contextual information Feature fusion |
| url | http://www.sciencedirect.com/science/article/pii/S0303243422000320 |
| work_keys_str_mv | AT xijiecheng enhancedcontextualrepresentationwithdeepneuralnetworksforlandcoverclassificationbasedonremotesensingimages AT xiaohuihe enhancedcontextualrepresentationwithdeepneuralnetworksforlandcoverclassificationbasedonremotesensingimages AT mengjiaqiao enhancedcontextualrepresentationwithdeepneuralnetworksforlandcoverclassificationbasedonremotesensingimages AT panleli enhancedcontextualrepresentationwithdeepneuralnetworksforlandcoverclassificationbasedonremotesensingimages AT shaokaihu enhancedcontextualrepresentationwithdeepneuralnetworksforlandcoverclassificationbasedonremotesensingimages AT pengchang enhancedcontextualrepresentationwithdeepneuralnetworksforlandcoverclassificationbasedonremotesensingimages AT zhihuitian enhancedcontextualrepresentationwithdeepneuralnetworksforlandcoverclassificationbasedonremotesensingimages |