This project implements water body segmentation using the DeepLabV3+ model. Mainly focus on the importance of ASPP in Deeplab V3+ for extracting the water bodies. It includes code for training the model, evaluating its performance, and metrics calculations.
deeplabv3Ex.py
: DeepLabV3+ modeltrain2.py
: Training the model codeeval.py
: Model evaluation codemetrics.py
: Evaluation metrics code
git clone https://github.com/sunandhini96/Water_body_segmentation-DeepLabV3plus.git
cd Water_body_segmentation-DeepLabV3plus
python train2.py
python eval.py
The project uses RGB satellite images and corresponding masks from Sentinel-2 A/B satellite. You can obtain the dataset https://www.kaggle.com/datasets/franciscoescobar/satellite-images-of-water-bodies
If you use this code in your research, please cite our paper for more details.
For a detailed explanation of the project and results, refer to our paper.