Plastic waste pollution has become a critical environmental challenge that requires innovative monitoring approaches to support effective environmental management. This systematic review synthesizes recent advancements in remote sensing (RS) technologies for plastic waste detection, analyzing 84 studies published between 2018 and 2024 following PRISMA guidelines. The review evaluates RS platforms, sensor types, spectral ranges, classification methods, and polymer identification across diverse environmental settings. Satellite platforms dominate large-scale marine monitoring (45% of studies), while unmanned aerial vehicles (UAVs) excelled in high-resolution coastal applications (23%). Correspondence analysis identified four distinct research clusters optimized for specific platform-environment combinations. Supervised learning was most prevalent (50%), though deep learning approaches and hybrid models show emerging promise. Polyethylene was most frequently detected across platforms. Limitation of the research field includes geographic bias towards European sites (> 50%), focus on controlled conditions rather than operational deployment, inability to detect microplastics, and lack of standardized protocols. The review also highlights emerging developments in RS technologies, including spectral mechanisms that support polymer discrimination and ongoing gaps in plastic monitoring. An integrated framework is proposed that combines multi-platform Earth Observation (EO), machine learning, and citizen science to enable scalable plastic waste monitoring. The findings provide theoretical and practical insights to guide future sensor design, algorithm development, and global monitoring strategies.
Potiracha et al. (Sat,) studied this question.