Detect waste in Pomerania

Helping environment with Machine Learning methods

Photo from

Detect waste with AI

This is a non-profit educational project from the AI for Good series in which I focus on the problem of waste in the environment with strong emphasis on plastics. During the project I learn together with the Women in Machine Learning and Data Science group from Tri-City and work on a model for trash detection. The project was attended by 9 participants supported by mentors (employees of StyleSage, Voicelab, BrainScan and NASA).

What and why?

Today, more than 300 million tons of plastic are produced annually. Plastic is everywhere and we constantly use it in our daily life. The idea of Plastic object detection project is to use Artificial Intelligence to detect plastic waste in the environment. Our solution will be applicable for video and photography. Our goal is to use AI for Good.


Main tasks:

  • development of waste object detection methods algorithms based on neural networks,
  • verifying possibility of using domain adaptation methods for training object detectors in case of detection made on photos in different styles (different illumination condition, or scenery).

Technologies used: Python, Pytorch

Methods used: Deep Neural Networks, Litter Detection and Recognition, Weak-Supervised Object Detection, Semi-Supervised Instance Segmentation


Main works:

Github code:

Medium posts: Creating an AI to Combat Environmental Pollution | by Magdalena Kortas and Sylwia Majchrowska | Towards Data Science

Learn more about the project

Project’s website: