Tyndall-effect-based optical monitoring of micro-plastic pollution in aquatic environments

Abstract

The paper presents the development and validation of an experimental system for monitoring microplastic pollution in aquatic environments, based on the Tyndall effect and artificial intelligence techniques. The proposed system uses a laser light source and a video acquisition module to detect light scattered by microplastic particles, with data processed in real time on the NVIDIA Jetson Orin Nano Developer Kit (SBC) platform. Mathematical models and discrete algorithms are developed for correlating the optical signal with the concentration of microplastics, and the performance of the method is improved by integrating artificial neural networks such as CNN and LSTM. The experimental results, presented in the form of images and quantitative graphs, demonstrate the system's ability to identify the presence of microplastics and dynamically monitor concentration variations. The proposed solution offers a fast, non-invasive, and scalable approach with high potential for continuous monitoring of the aquatic environment.