Satellite Imagery Segmentation on Image with 3 layers RGB using U-Net Deep learing model This project aims to segment satellite imagery using a U-Net deep learning model. The model is trained on annotated satellite images to identify specific features of interest,

such as :

• Building
• Land
• Road
• Vegetation
• Water
• Unlabeled

and to segment them from the rest of the image.

• Satellite_Seg_Prediction_GradioUI.ipynb : for Use the trained model and predict rusltes from Code
• Satellite-Imagery-Segmentation-with-U-Net-Deep-le.ipynb : for trained the Orignal model from scratch

• https://huggingface.co/spaces/AH-MD/Satellite-Imagery-Segmentation

• Dubai Segmentation Dataset Kaggle: https://www.kaggle.com/datasets/humansintheloop/semantic-segmentation-of-aerial-imageryx