The remote sensing image semantic segmentation repository based on tf.keras includes backbone networks such as resnet, densenet, mobilenet, and segmentation networks such as deeplabv3+, pspnet, panet, and segnet.

This repository has been used to participate in the remote sensing semantic image segmentation track of the 2020 National Artificial Intelligence Competition (NAIC).

• python 3.7
• tensorflow-gpu 2.3
• opencv-python
• tqdm
• numpy
• argparse
• matplotlib
• Pillow

data

python split_val_data_from_train.py

Modify the config.py file

python train.py --model DeepLabV3Plus --backBone ResNet152 --lr_scheduler cosine_decay --lr_warmup True

Link (TBD)

python predict.py

MIou, FWIou
The inference result image is in the results folder