TensorFlow implementation of semantic template matching.
Please use Python 3.7, install NumPy, OpenCV (3.4.2) and TensorFlow (2.0.0).
| data | |
|---|---|
| google earth image | https://drive.google.com/drive/folders/1LV8n80daRKCySmCCQ1nZP6lB4aRsN3CM?usp=sharing |
| landsat-8 | |
| GF-2 | |
| SAR-optical | https://drive.google.com/drive/folders/12x2m2temb5IdUjUfhWuEzCK1sXXT2ZME?usp=sharing |
reference image----------------------,---------------------template image,---------------------------,label
First download the training data, place it under the project, and then generate the .tfrecord file using the code in generate data.
Training semantic template matching models using model files
Li L, Han L, Ding M, et al. A deep learning semantic template matching framework for remote sensing image registration[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 181: 205-217.