@article{li2023forest, title={Forest Height Inversion by Convolutional Neural Networks Based on L-Band PolInSAR Data Without Prior Knowledge Dependency}, author={Li, Dandan and Lu, Hailiang and Li, Chao and Mohaisen, Linda and Jing, Weipeng}, journal={IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing}, year={2023}, publisher={IEEE} }
Dataset address (L-band lope site): https://uavsar.jpl.nasa.gov/
Important package:
pytorch2.0
kapok: for install it, please access to https://github.com/simard-landscape-lab/kapok/tree/main/docs
Before running the train-lope-PIDLF.py code, afrisar-lope.py (containing the four baseline methods) should be executed to generate the pseudo labels.
The test-lope-PIDLF.py script is compatible only with Windows systems. Running it on Linux will result in errors.
Please feel free to email any questions to 1229476319@qq.com.