Image-Clustering: Unsupervised Satellite Image Segmentation Via Silhouette Score Optimized K Means Clustering

This project segments satellite images at high levels of detail using an optimized K means Clustering Algorithm. The Clustering Notebook is the clustering model and edge detection model. The color spaces notebook segments images based on the color distributions of an image. The Mapbox script generates satellite images based on the latitude and longitude coordinates. The 3 folders titled San Francisco, Santa Cruz, and sat images contain some sample data that can be directly used. Each file has directions on how to run each model. For more details on this project, check out the report!