Microplastics are emerging environmental pollutants that pose significant threats to aquatic ecosystems and human health. Traditional methods of detecting microplastics involve manual inspection and spectroscopy, which are time-consuming and require expert intervention. This project presents a machine learning-based solution using the YOLO (You Only Look Once) object detection algorithm to automate and accelerate the detection of microplastics in environmental samples. High-resolution microscopic images are used to train and validate the YOLO model, enabling real-time and accurate identification of microplastic particles based on shape, size, and color. The system achieves fast processing speeds and maintains high accuracy in distinguishing microplastics from organic and inorganic debris. The proposed approach demonstrates the potential of deep learning in environmental monitoring and provides a scalable tool for researchers and policymakers in addressing plastic pollution. Data Preparation: Convert impedance measurement datasets into images, capturing unique features of micro plastics for analysis. Model Utilization: Employ YOLO pre-trained models for real-time detection and classification of micro plastics in aquatic environments. Deep Learning Integration: Use Python and Flask for backend processing, leveraging deep learning to enhance detection precision and scalability. Output and Application: Deliver results via a user-friendly HTML front-end, supporting efficient water quality monitoring and ecological protection efforts.
