Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed Images
Land cover change detection (LCCD) with remote-sensed images plays an important role in observing Earth’s surface changes. In recent years, the use of a spatial-spectral channel attention mechanism in information processing has gained interest. In this study, aiming to improve the performance of LCC...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/1/87 |
| _version_ | 1797431257775210496 |
|---|---|
| author | Xu Yang Zhiyong Lv Jón Atli Benediktsson Fengrui Chen |
| author_facet | Xu Yang Zhiyong Lv Jón Atli Benediktsson Fengrui Chen |
| author_sort | Xu Yang |
| collection | DOAJ |
| description | Land cover change detection (LCCD) with remote-sensed images plays an important role in observing Earth’s surface changes. In recent years, the use of a spatial-spectral channel attention mechanism in information processing has gained interest. In this study, aiming to improve the performance of LCCD with remote-sensed images, a novel spatial-spectral channel attention neural network (SSCAN) is proposed. In the proposed SSCAN, the spatial channel attention module and convolution block attention module are employed to process pre- and post-event images, respectively. In contrast to the scheme of traditional methods, the motivation of the proposed operation lies in amplifying the change magnitude among the changed areas and minimizing the change magnitude among the unchanged areas. Moreover, a simple but effective batch-size dynamic adjustment strategy is promoted to train the proposed SSCAN, thus guaranteeing convergence to the global optima of the objective function. Results from comparative experiments of seven cognate and state-of-the-art methods effectively demonstrate the superiority of the proposed network in accelerating the network convergence speed, reinforcing the learning efficiency, and improving the performance of LCCD. For example, the proposed SSCAN can achieve an improvement of approximately 0.17–23.84% in OA on Dataset-A. |
| first_indexed | 2024-03-09T09:41:19Z |
| format | Article |
| id | doaj.art-022f594f04734afa8a183faa8936511a |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T09:41:19Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-022f594f04734afa8a183faa8936511a2023-12-02T00:50:52ZengMDPI AGRemote Sensing2072-42922022-12-011518710.3390/rs15010087Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed ImagesXu Yang0Zhiyong Lv1Jón Atli Benediktsson2Fengrui Chen3College of Geological Engineering and Geomatics, Chang’an University, Xi’an 710054, ChinaSchool of Computer Science and Engineering, Xi’an University of Technology, Xi’an 710048, ChinaFaculty of Electrical and Computer Engineering, University of Iceland, IS 107 Reykjavik, IcelandKey Laboratory of Geospatial Technology for Middle and Lower Yellow River Regions, Henan University, Ministry of Education, Kaifeng 475001, ChinaLand cover change detection (LCCD) with remote-sensed images plays an important role in observing Earth’s surface changes. In recent years, the use of a spatial-spectral channel attention mechanism in information processing has gained interest. In this study, aiming to improve the performance of LCCD with remote-sensed images, a novel spatial-spectral channel attention neural network (SSCAN) is proposed. In the proposed SSCAN, the spatial channel attention module and convolution block attention module are employed to process pre- and post-event images, respectively. In contrast to the scheme of traditional methods, the motivation of the proposed operation lies in amplifying the change magnitude among the changed areas and minimizing the change magnitude among the unchanged areas. Moreover, a simple but effective batch-size dynamic adjustment strategy is promoted to train the proposed SSCAN, thus guaranteeing convergence to the global optima of the objective function. Results from comparative experiments of seven cognate and state-of-the-art methods effectively demonstrate the superiority of the proposed network in accelerating the network convergence speed, reinforcing the learning efficiency, and improving the performance of LCCD. For example, the proposed SSCAN can achieve an improvement of approximately 0.17–23.84% in OA on Dataset-A.https://www.mdpi.com/2072-4292/15/1/87change detectiondeep learningattention moduleremote-sensed images |
| spellingShingle | Xu Yang Zhiyong Lv Jón Atli Benediktsson Fengrui Chen Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed Images Remote Sensing change detection deep learning attention module remote-sensed images |
| title | Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed Images |
| title_full | Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed Images |
| title_fullStr | Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed Images |
| title_full_unstemmed | Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed Images |
| title_short | Novel Spatial–Spectral Channel Attention Neural Network for Land Cover Change Detection with Remote Sensed Images |
| title_sort | novel spatial spectral channel attention neural network for land cover change detection with remote sensed images |
| topic | change detection deep learning attention module remote-sensed images |
| url | https://www.mdpi.com/2072-4292/15/1/87 |
| work_keys_str_mv | AT xuyang novelspatialspectralchannelattentionneuralnetworkforlandcoverchangedetectionwithremotesensedimages AT zhiyonglv novelspatialspectralchannelattentionneuralnetworkforlandcoverchangedetectionwithremotesensedimages AT jonatlibenediktsson novelspatialspectralchannelattentionneuralnetworkforlandcoverchangedetectionwithremotesensedimages AT fengruichen novelspatialspectralchannelattentionneuralnetworkforlandcoverchangedetectionwithremotesensedimages |