Plastic debris in marine and freshwater ecosystems has become a growing global concern, negatively impacting aquatic life and human health. The worldwide challenge of persistent macro-litter, which originates from various origins and is conveyed by rivers to the oceans, has significant biological, chemical, and ecological impacts on oceanic environments. Plastic contamination is a critical ecological issue that can cause a variety of adverse effects on ecosystems. Its presence affects organisms in both water and soil, significantly threatening the stability of these natural systems. This work aims to advance unmanned aerial vehicles (UAVs), remote sensing (RS), and machine learning (ML) methods for litter tracking and cleanup support, promoting ecological protection and waste reduction. AI-powered tools, such as camera- and sensor-equipped UAVs (drones), enable real-time monitoring of microplastics (MPs) contamination across ecosystems. AI approaches (e.g., ML) have greatly improved the performance and accuracy of observing MP sources and forecasting contamination levels. These solutions reduce the requirement for human engagement and simplify the decision-making procedure. Significantly, more advanced algorithms (e.g., YOLOv12) are needed to assess the migration trends of plastics into the ecosystem. This advancement indicates promoting AI's ability to control current pollution and mitigate additional environmental degradation. The protocol facilitates large-scale comparative studies, promoting integrated coastal management efforts at both international and national levels. • Marine plastic contamination is a critical environmental issue that harms aquatic ecosystems. • Monitoring plastic presence is necessary to develop strategies for controlling contamination. • Powered tools like UAVs, RS, and ML efficiently track plastic sources and predict pollution. • Integrating advanced monitoring technologies, such as UAVs, RS, and ML, is crucial. • Future developments should focus on advanced technologies to enhance remote and autonomous monitoring with high accuracy.
Nguyen et al. (Fri,) studied this question.