Semantic segmentation of clouds in satellite images based on U-Net++ architecture and attention mechanism

•Proposed U-Net++ based network with attention mechanism for cloud segmentation.•Proposed SEUNet++ achieves an IoU score of 91.8 %.•Transfer learning helps to improve the segmentation results.•SEUNet++ performs better than the original U-Net++ architecture.•SEUNet++ improves the state-of-the-art by...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Expert systems with applications Jg. 209; S. 118380
Hauptverfasser: Kaur Buttar, Preetpal, Sachan, Manoj Kumar
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 15.12.2022
Schlagworte:
95-Cloud
Cloud segmentation
Multispectral satellite data
ResNet
Semantic segmentation
U-Net
Cloud segmentation
Semantic segmentation
U-Net
95-Cloud
ResNet
Multispectral satellite data
ISSN:0957-4174, 1873-6793
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Proposed U-Net++ based network with attention mechanism for cloud segmentation.•Proposed SEUNet++ achieves an IoU score of 91.8 %.•Transfer learning helps to improve the segmentation results.•SEUNet++ performs better than the original U-Net++ architecture.•SEUNet++ improves the state-of-the-art by lifting the IoU score by 0.23 %. The presence of clouds in satellite imagery may pose hindrances to the accurate and reliable analysis of the objects present on the land. Therefore, automatic cloud detection is a vital pre-processing step before lending the satellite images to any further analysis. This is a challenging task due to the varying thickness and densities of clouds. It is also very difficult to distinguish clouds from certain terrains such as snow and white sandy beaches. This paper proposes a deep learning based algorithm to solve the problem of cloud segmentation on the Landsat 8 multispectral dataset, 95-Cloud: SEUNet++. Specifically, the proposed model consists of a U-Net++ semantic segmentation model with a lightweight channel attention mechanism. We also experimented with using different encoder backbones in the U-Net++ encoder-decoder architecture such as ResNet (Residual Neural Networks) variants including ResNet-18, ResNet-34, ResNet-50, and ResNet-101, DenseNet-264, CSPNet (Cross Stage Partial Network), and EfficientNet-B8 and compared their performance. The experimental results show that the proposed architecture achieves an IoU (Intersection over Union) score of 91.8 %, improving the state-of-the-art on the task by 0.23 %. It also boosts the accuracy, precision, and recall values creating crisp cloud boundaries and detecting even thin layers of clouds. We also experimented using transfer learning and found that it has a positive impact on the cloud segmentation task. The proposed model also beats the original U-Net++ architecture in terms of various evaluation metrics such as the IoU score, accuracy, precision, and recall. The experimental results thereby demonstrate that our model is computationally efficient and achieves precise segmentation results.
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2022.118380