Global Satellite Flood Detection - U-Net Ready Patches (sf-p-256)

This repository contains a highly optimized, machine-learning-ready dataset compiled from various globally collected satellite images and diverse flood events.

🛠️ Data Processing Pipeline

1. Patching: Images cropped into 256x256 patches.
2. NaN Filtering: Patches with more than 30% missing data removed.
3. Imputation: Remaining NaN/Inf values set to 0.0.
4. Split: train (70%), val (15%), test (15%).

📂 Dataset Structure

• images/: Sentinel-1 SAR (VV, VH). Shape (256, 256, 2).
• ndwi/: NDWI from Sentinel-2. Shape (256, 256, 1).
• masks/: Binary ground truth. Shape (256, 256, 1).

📊 Global Normalization Stats

stats.npz contains train-only mean/std for normalization.

🖼️ Visual Samples

🚀 How to Load

import numpy as np
s1_patch = np.load('train/images/train_00001.npy')
ndwi_patch = np.load('train/ndwi/train_00001.npy')
mask_patch = np.load('train/masks/train_00001.npy')
stats = np.load('stats.npz')
s1_mean, s1_std = stats['s1_mean'], stats['s1_std']
s1_normalized = (s1_patch - s1_mean) / (s1_std + 1e-8)