SatellightSight

Waste Detection in Satellite Images using Machine Learning

A Machine Learning approach to detect waste dumps using satellite images.
It uses texture analysis techniques like GLCM and LBP.
A state-of-the-art Sliding Window Algorithm to analyses sub-images. It is adapted to the training data used for the model.
A binary classifier model is used to classify each sub-image of a given satellite image into 2 categories - waste or non-waste.

Click here to see it in action!

Dataset:

https://www.kaggle.com/datasets/shenhaibb/swad-dataset

References:

https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9935119
https://github.com/jrieke/streamlit-image-select/tree/main