Laser & Optoelectronics Progress, Volume. 59, Issue 18, 1810015(2022)
Cloud-Type Recognition Based on Multiscale Features and Gradient Information
Fig. 1. Framework of DGNet based on gradient algorithm
Fig. 2. Procedure of extracting gradient information from feature maps
Fig. 3. Patterns of 10 cloud types
Fig. 4. Comparison between DGNet121 and classical classification networks. (a) Training loss; (b) testing accuracy
Fig. 5. Double-threaded structure with gradient information
Fig. 6. Classification accuracy of DGNet network with different position gradient features
Fig. 7. Training and testing loss with various data ratio
Fig. 8. Visualization of characteristic graph of ResNet50 network
Fig. 9. Visualization of characteristic graph of DenseNet121 network
Fig. 10. Visualization of characteristic graph of DGNet121 network
|
|
|
|
|
|
Get Citation
Copy Citation Text
Lingjie Jin, Zhiwei Lin, Yu Hong. Cloud-Type Recognition Based on Multiscale Features and Gradient Information[J]. Laser & Optoelectronics Progress, 2022, 59(18): 1810015
Category: Image Processing
Received: Jul. 8, 2021
Accepted: Aug. 10, 2021
Published Online: Aug. 29, 2022
The Author Email: Lin Zhiwei (cwlin@fafu.edu.cn)