Identification of the density of incorrectly parked vehicles from satellite images (e.g. on a lawn, garage driveway, etc.) - i.e. image analysis and classification. Objective: assessment of the difficulty of finding a parking space
Using GPS coordinates (latitude and longitude) define tuples in src/scraper/get_images.py and run the code. The screenshots from Google Maps and Openstreet Map will then appear in the data/ directory.
If the image data needs to be aligned, this should be done using the prepared application src/scraper/app.py.
The prepared method for calculating metrics uses masks that we generate on the aligned images using an src/preprocessing/parking_extractor.py script.