Laser & Optoelectronics Progress, Volume. 59, Issue 16, 1611005(2022)
Hyperspectral Image Classification Based on Edge-Preserving Filter and Deep Residual Network
Fig. 1. Overall flow of the proposed method
Fig. 2. Structure diagram of residual element
Fig. 3. Model structure of depth residual network
Fig. 4. Indian Pines dataset. (a) False color image;(b) real ground data
Fig. 5. Pavia University dataset. (a) False color image;(b) real ground data
Fig. 6. Classification accuracy of different dropout values. (a) Indian Pines; (b) Pavia University
Fig. 7. Loss function and overall classification accuracy of different epoch values. (a) Indian Pines; (b) Pavia University
Fig. 8. Classification accuracy of different
Fig. 9. Overall classification accuracy of different
Fig. 10. Classification results of different algorithms in Indian Pines dataset
Fig. 11. Partial enlargement comparison of classification results of Indian Pines dataset
Fig. 12. Classification results of different algorithms in Pavia University dataset
Fig. 13. Partial enlargement comparison of classification results of Pavia University dataset
|
|
|
|
|
Get Citation
Copy Citation Text
Lü Huanhuan, Zhuolu Wang, Hui Zhang. Hyperspectral Image Classification Based on Edge-Preserving Filter and Deep Residual Network[J]. Laser & Optoelectronics Progress, 2022, 59(16): 1611005
Category: Imaging Systems
Received: Jul. 18, 2021
Accepted: Aug. 23, 2021
Published Online: Jul. 22, 2022
The Author Email: Zhang Hui (wangzl2019@126.com)