justjoshtings / satellite_image_segmentation

Experimented with a U-Net variant to perform pixel wise multi-class classification to segment high resolution satellite imagery into land use types with potential applications in deforestation and flood monitoring. Developed with TensorFlow framework.

Data Scientist and Data Engineer

Leveraging my background in tech, clean energy engineering, and data science/machine learning to drive solutions in climate and sustainability.

Learn more about me and my work at my personal website.

satellite_image_segmentation's Issues

Data preparation process

Would you please explain the data preparation process?

unet_model_v3_epoch6.h5 file missing

Thank you very much for sharing your implementation. I am learning a lot from your Land_Cover_Segmentation.ipynb file.
Unfortunately, I cannot find the file "Land_Cover_Segmentation.ipynb" to load the pretrained weights. Can you please help me?

Thanks

Dataset source

Hi, I saw you recently changed the dataset source from kaggle to the official website, however I note it is not actually available there and moreover there appear to be many unanswered requests for access on https://competitions.codalab.org/forums/15208/
You could suggest both locations with the above caveat.
Cheers