Laser & Optoelectronics Progress, Volume. 58, Issue 14, 1401002(2021)
Cloud Detection of Landsat Image Based on MS-UNet
Figures & Tables(11)
![Schematic diagrams of activation function. (a) ReLU function: max(x,0); (b) FReLU function: max[x,T(x)]](/Images/icon/loading.gif){kind=icon}
Fig. 4. Schematic diagrams of activation function. (a) ReLU function: max(x,0); (b) FReLU function: max[x,T(x)]
Fig. 8. Comparison of results of improved modules. (a) (b) Broken clouds; (c) (d) thin clouds
Fig. 9. Comparison of cloud detection results of different methods. (a) (b) Broken clouds; (c) (d) thin clouds
Table 1. Index comparison results between improved algorithm and original algorithms
View tableTable 1. Index comparison results between improved algorithm and original algorithms
Method Precision Accuracy Recall F1-Score IoU MIoU UNet 0.890 0.868 0.787 0.835 0.717 0.760 MS-UNet 0.881 0.936 0.941 0.912 0.839 0.871 MS-UNet+FReLU 0.901 0.943 0.945 0.921 0.855 0.884
Table 2. Comparison of indicators of different algorithms
View tableTable 2. Comparison of indicators of different algorithms
Method Precision Accuracy Recall F1-Score IoU MIoU MF-CNN[ 19 ]0.893 0.878 0.886 0.914 0.795 0.833 SegNet[ 20 ]0.828 0.873 0.833 0.831 0.711 0.763 DeepLabV3_ResNet50[ 21 ]0.931 0.930 0.877 0.909 0.833 0.863 DeepLabV3_ResNet101[ 21 ]0.912 0.938 0.923 0.918 0.848 0.877 MS-UNet 0.881 0.936 0.941 0.912 0.839 0.871 MS-UNet+FReLU 0.901 0.943 0.945 0.921 0.855 0.884
Tools
Get Citation
Copy Citation Text
Haitao Wang, Yichen Wang, Yongqiang Wang, Yurong Qian. Cloud Detection of Landsat Image Based on MS-UNet[J]. Laser & Optoelectronics Progress, 2021, 58(14): 1401002
Paper Information
Category: Atmospheric Optics and Oceanic Optics
Received: Oct. 12, 2020
Accepted: Nov. 19, 2020
Published Online: Jun. 30, 2021
The Author Email: Yurong Qian (qyr@xju.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20