D2Mamba: A Mamba-based Method for Floodway Obstructions Segmentation

This repository contains the official implementation and dataset samples for the paper:
"D2Mamba: A Mamba-based method for floodway obstructions segmentation from multispectral satellite imagery".

🛰️ Dataset

We constructed the Multispectral Floodway Obstructions Dataset (MFOD), containing 8,301 high-resolution images.

Transparency Note: We provide supplementary sample data via the Baidu Netdisk folder MFOD_samples (link, extraction code: tgvk).

🛠️ Environment

To reproduce the results, please set up the environment using the following dependencies.

# Basic requirements
python
torch
torchvision
torchaudio

# Mamba specific requirements (please refer to official mamba installation)
pip install "causal-conv1d==1.0.0"
pip install "mamba-ssm==1.0.1"

# Other dependencies
pip install -r requirements.txt

📧 Contact

For questions regarding the dataset or code, please contact the corresponding author.

---