AI-based vision system for sorting and recycling plastic waste
A CEMLab research project led by Prof. Dorra Sellami is developing an artificial-intelligence vision system to sort waste and support plastic recycling.
CEMLab is developing an artificial-intelligence-based vision system for sorting waste, applied to plastic recycling (Développement d'un système de vision basé sur l'intelligence artificielle en vue de tri de déchets pour le recyclage des déchets en plastique). The project is coordinated by Prof. Dorra Sellami.
Why it matters
Plastics come in many polymer types, colours and shapes, and they often reach sorting lines dirty or deformed. Manual sorting is slow and error-prone, and poorly sorted plastic cannot be recycled well. Camera-based systems driven by deep learning can recognise and classify items on the conveyor in real time. This improves the purity of the recovered material and cuts the amount of waste sent to landfill.
The project
The project applies the laboratory's artificial vision and pattern recognition expertise to an environmental and industrial problem. The goal is to recognise plastic waste automatically from images so it can be sorted for recycling.
One result of this line of research is the article Distilling the experts: A new path for plastic waste sorting by S. Daou, M. Zouari Mehdi, J. Frikha Elleuch, D. Sellami, S. Sahnoun, A. Fakhfakh and K. Elleuch. It was published in Expert Systems with Applications (vol. 324, 2026) with support from the Tunisian Ministry of Higher Education and Scientific Research.
Sources: CEMLab activity report 2025 (LR11ES47), section III-3; Expert Systems with Applications.
Recent Posts
