FluoTrack: Detection Of Microplastics
ML-Powered Detection Of Microplastics
Created on 18th January 2026
•
FluoTrack: Detection Of Microplastics
ML-Powered Detection Of Microplastics
The problem FluoTrack: Detection Of Microplastics solves
Microplastic pollution has emerged as a serious environmental and public health concern, yet its detection and monitoring remain largely manual, time-consuming, and inconsistent. Existing methods rely heavily on expert-driven microscopic analysis, which is prone to human error, limited scalability, and high operational cost. As a result, real-time and large-scale monitoring of microplastics in environmental samples is extremely difficult. FluoTrack addresses this gap by automating the detection and tracking of microplastics using fluorescence imaging and AI, enabling faster, more accurate, and reproducible identification of microplastic particles that are otherwise invisible to the naked eye.
Challenges we ran into
One of the major challenges was accurately distinguishing microplastics from background noise and organic particles under varying fluorescence intensities and lighting conditions. Training robust AI models required careful preprocessing, annotation, and augmentation of microscopy images, which was both time-intensive and technically demanding. Additionally, optimizing model performance while maintaining real-time inference posed constraints on computational efficiency. Integrating image processing techniques with deep learning models in a way that remained interpretable and reliable for environmental monitoring was another key challenge we had to overcome.
Technologies used