scsd-cdh/payload-ai-suite

AI Summary: Improve the security and sharing of environment variables used in the Space Concordia AI payload project by implementing a secure method (e.g., using a .env file with .gitignore) and updating the README.md to reflect the new workflow. This involves researching alternative solutions and documenting the chosen approach.

AI Summary: Implement a dynamic z-score normalization function in `preprocess.py` using OpenCV or simple Python logic. The function should normalize image channels to have a mean of 0 and a standard deviation of 1, excluding data-less pixels. This is a preprocessing step for a wildfire detection model, aiming for sensor and resolution agnostic image inputs.

AI Summary: Migrate the existing Python project's dependency management from pip to `uv`, incorporate `pyproject.toml` for standardized build configuration, and document the updated build process in a `BUILD.md` file. This involves updating the project's structure and potentially resolving any conflicts arising from the change in dependency management.

AI Summary: Integrate FIRMS data with Copernicus imagery. This involves querying the FIRMS database (or JSON output) to identify wildfire events, then using the Copernicus API to retrieve corresponding satellite imagery for those events. The imagery should be stored locally, following existing patterns in `fetch.py`. The sensor type in the Copernicus API call needs to be adjusted in the evaluation script. This is a step towards using both data sources for wildfire detection.