DEMO - Web Application


The goal of our research is to develop an accurate classification method for plastic waste detection to provide a viable platform for repeatable, cost-effective and large-scale monitoring. Such a robust waste monitoring solution would speed up the detection of illegal waste hot-spots close to water flows and floating waste islands on rivers, as well as support waste collection actions with an automatic monitoring system. This application automatically searches for newly recorded satellite images and downloads them on a daily basis. After this a Random Forest model classifies the pictures and displays the results in the web view. You can check out the extension of polluted areas on the set locations in the previous five days when the cloud cover over them was 0%.


  • Location: You can choose from four previosly set locations: Kisköre, Lake Călinești, Pusztazámor and Рахів
  • Date: It can be changed using the swipe. You can select from the last five most recent days when the cloud cover over the areas was 0%.
  • Colors:
    • Classified: Orange. All pixels that were classified as plastic waste.
    • Heatmap High: Red. Pixels that were classified as plastic waste with a confidence of 90% or higher.
    • Heatmap Medium: Yellow. Pixels that were classified as plastic waste with a confidence between 80% and 90%.
    • Heatmap Low: Green. Pixels that were classified as plastic waste with a confidence below 80%.


  1. Waste Detection and Change Analysis based on Multispectral Satellite Imagery
    Dávid Magyar, Máté Cserép, Zoltán Vincellér, Attila D. Molnár
    In Proceedings of KEPAF, art. 53., p. 18., 2023. DOI: 10.48550/arXiv.2303.14521
  2. Comparative Analysis of Riverine Plastic Pollution Combining Citizen Science, Remote Sensing and Water Quality Monitoring Techniques
    Attila D. Molnár, Kristóf Málnás, Sára Bőhm, Miklós Gyalai-Korpos, Máté Cserép, Tímea Kiss
    Sustainability, vol. 16 (12), art. 5040, 2024. DOI: 10.3390/su16125040