We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Empowering Sustainability Through AI-Driven Monitoring: The DEEP-PLAST Approach to Marine Plastic Detection and Trajectory Prediction for the Black Sea
AI summary Read the abstract
This study introduced DEEP-PLAST, an AI-based framework integrating YOLOv5 object detection, U-Net++/U-Net semantic segmentation on Sentinel-2 satellite imagery, and Lagrangian drift modeling to detect floating marine plastic debris and predict its trajectory in the Black Sea. U-Net++ achieved the best segmentation performance (F1 = 0.84, false positive rate 5.2%), and detected debris locations were linked to ocean current and wind models to identify accumulation zones supporting targeted cleanup.
Marine plastic pollution represents a critical ecological challenge, exerting long-lasting impacts on ecosystems, biodiversity, and human well-being. This study introduces the DEEP-PLAST project, an integrated AI-based framework designed for the detection and trajectory prediction of floating marine plastic waste using open-access Sentinel-2 satellite imagery and environmental models of ocean currents and wind. The DEEP-PLAST methodology integrates object detection (YOLOv5 on UAV data), semantic segmentation (U-Net/U-Net++ on Sentinel-2), and drift simulation using Copernicus and NOAA datasets. U-Net++ achieved the best performance (F1 = 0.84, false positive rate 5.2%), outperforming other models. Detected debris locations were linked to Lagrangian drift models to identify accumulation zones in the Black Sea, supporting targeted cleanup efforts. While promising, drift validation remains qualitative due to limited ground truth, to be addressed in future work with in situ and NGO data. This approach supports EU Mission Ocean, the Marine Strategy Framework Directive, and UN SDGs, demonstrating the potential of AI and remote sensing for marine protection. Future efforts will expand datasets, apply the platform to other seas, and launch a web tool for NGOs and policymakers.
More Papers Like This
Leveraging Artificial Intelligence for Plastic Waste Management in Coastal Waters: A Drone-Based Approach for Identification and Hotspot Detection
AI summary Read the abstract
A machine learning model trained on drone footage achieved 89.4% accuracy in classifying plastic debris types in coastal waters and identifying hotspot areas for targeted cleanup operations. The AI-driven approach offers faster and more scalable monitoring than traditional methods, though detection of small microplastics and environmental variability remain challenges.
A novel deep learning model for predicting marine pollution for sustainable ocean management
AI summary Read the abstract
This study developed a deep learning model combining two-step clustering and fully convolutional networks (FCN) to detect and categorize plastic waste in marine and undersea environments, achieving 98.38% accuracy using Kaggle's plastic location dataset. The trained model was applied to an automated floating plastic detection system, demonstrating the potential of AI-driven tools to enhance marine pollution monitoring.
Advanced Classification of Marine Pollutants Using Sentinel-2 Multispectral Thermal Imaging and Vision Transformer for Enhanced Water Quality Assessment
AI summary Read the abstract
This study used satellite multispectral imaging from the Sentinel-2 platform combined with a Vision Transformer machine learning model to automatically classify different types of marine pollutants — including plastics, algae, and oil — from aerial imagery. The AI-based approach significantly outperformed traditional classification methods and could detect plastic debris patches across large ocean areas. Automated large-scale detection of marine plastic pollution from satellites could transform the way we monitor and respond to ocean plastic contamination.
Deep Learning Approaches for Ocean Waste Detection: A Comparative Study of YOLOv8 and Faster R-CNN
AI summary Read the abstract
Deep learning models (YOLOv8 and Faster R-CNN) were compared for detecting ocean waste, with Faster R-CNN achieving better precision for meso- and macroplastic detection, though both systems struggled with microplastics due to their tiny size and visual similarity to natural marine particles. This highlights the urgent need for specialized AI detection methods capable of identifying microplastics in ocean monitoring applications.
Smart Ocean Cleanup: An AI-Integrated Autonomous System for Marine Waste Management
AI summary Read the abstract
This paper presents an AI-powered autonomous boat system designed to detect and collect marine pollution — including plastics, oil spills, and microplastics — using deep learning image classification, IoT sensors, and robotic collection mechanisms. The system demonstrated over 94% accuracy for pollutant detection and classification across several AI models. While focused more broadly on ocean cleanup technology than on microplastic science specifically, it demonstrates how AI-integrated robotics could help address the practical challenge of removing plastic waste from ocean surfaces before it breaks down further.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.