CloudViT: A Lightweight Vision Transformer Network for Remote Sensing Cloud Detection
- 29 December 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 20 (1545598X) , 1-5
Abstract
Clouds inevitably exist in satellite images, which limits the processing and application of satellite images to a certain extent. Therefore, cloud detection is a preprocessing task in satellite image extraction and analysis processing. However, the existing methods are difficult to mine robust features, and the number of parameters and computation is large, which is not conducive to the deployment of the model. In this letter, CloudViT, a lightweight vision transformer network for cloud detection from satellite imagery, is proposed. In detail, to utilize dark channel priors in multispectral imagery to guide the network to learn features, a multi-scale dark channel extractor is used to first predict dark channels, and then the dark channel features and image features are input to the attention mechanism-based dark channel-guided context aggregation module to enhance image features, which in turn makes cloud detection results more accurate. At the same time, to enhance the transfer ability of the network between different satellite sensors, a plug-and-play channel adaptive module is proposed to deal with the inconsistency of the number of different satellite sensor bands. Experimental results on the Landsat7 dataset show that our network CloudViT outperforms state-of-the-art methods while keeping the number of parameters and computation small. At the same time, the experimental results on transfer to three other datasets show that using the channel adaptation module can greatly improve the transfer ability of the model.Keywords
Funding Information
- National Natural Science Foundation of China (42030102)
- National Natural Science Foundation of China (42192583)
- Fund for Innovative Research Groups of the Hubei Natural Science Foundation (2020CFA003)
- Major Special Project of Guizhou ([2022]001)
This publication has 15 references indexed in Scilit:
- Fmask 4.0: Improved cloud and cloud shadow detection in Landsats 4–8 and Sentinel-2 imageryRemote Sensing of Environment, 2019
- Selective Kernel Networks2019
- Dual Attention Network for Scene Segmentation2019
- CDnet: CNN-Based Cloud Detection for Remote Sensing ImageryIEEE Transactions on Geoscience and Remote Sensing, 2019
- Deep learning based cloud detection for medium and high resolution remote sensing images of different sensorsISPRS Journal of Photogrammetry and Remote Sensing, 2019
- Cloud Detection in Remote Sensing Images Based on Multiscale Features-Convolutional Neural NetworkIEEE Transactions on Geoscience and Remote Sensing, 2019
- Squeeze-and-Excitation Networks2018
- Non-local Neural Networks2018
- Pyramid Scene Parsing Network2017
- Single Image Haze Removal Using Dark Channel PriorIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010