Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{189074,
        author = {Ashirvad pandey and Ahuti Singh and Nikhil Singh Rajawat},
        title = {Microplastic detection using Machine learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {4737-4742},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=189074},
        abstract = {Microplastic pollution has emerged as a critical environmental issue, with particles smaller than 5 mm infiltrating oceans, rivers, soils, and even the human food chain. These microplastics originate from industrial waste, cosmetic microbeads, textile fibers, tire wear, and the degradation of larger plastic debris. Once released, they persist in the environment due to their resistance to natural degradation, accumulate in aquatic organisms, and subsequently enter the human body through seafood, drinking water, and agricultural products. The ingestion of microplastics has been linked to numerous health risks including inflammation, oxidative stress, and endocrine disruption. Traditional detection methods, such as Fourier Transform Infrared Spectroscopy (FTIR), Raman Spectroscopy, and Scanning Electron Microscopy (SEM), provide accurate polymer identification but are expensive, time-consuming, and require expert handling. These constraints limit their application in large-scale monitoring and real-time field detection.
To address these challenges, this study proposes a novel, low-cost, and portable microplastic detection system that combines sensor technology with machine learning (ML). The system integrates turbidity sensors, infrared (IR) sensors, and high-resolution imaging sensors connected to an ESP32 microcontroller for real-time data acquisition. Collected sensor data undergoes preprocessing, including noise reduction, normalization, and feature extraction, before being analyzed using multiple machine learning algorithms, such as Random Forest (RF), Support Vector Machine (SVM), and Convolution Neural Networks (CNN). The CNN-based model demonstrated the highest accuracy, achieving up to 95% classification efficiency in distinguishing polyethylene (PE), polypropylene (PP), polystyrene (PS), and other microplastic types. Sensor fusion further enhanced detection reliability, allowing accurate recognition even in turbid water samples. The proposed system offers real-time monitoring capabilities, scalability, and portability, making it suitable for environmental applications, wastewater analysis, and field-based research.
Overall, the integration of sensors and machine learning presents a practical, cost-effective, and automated solution for microplastic detection, bridging the gap between laboratory-level analysis and field-deployable systems. This approach not only advances environmental monitoring practices but also supports data-driven decision-making for pollution management and public health protection. The proposed methodology sets a foundation for future enhancements, including edge AI implementation, IoT-based data logging, and autonomous environmental surveillance.},
        keywords = {},
        month = {December},
        }