Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study introduces EcoVision, an AI-driven framework for monitoring salt marsh vegetation using high-resolution UAV imagery processed through a modular pipeline that includes transformer-based semantic segmentation and ConvNeXt for species classification. The system effectively identifies two key halophytic grasses, Spartina maritima and Puccinellia maritima, achieving a mean Intersection over Union (IoU) of 0.56 and an object-level F1 score of 0.99, while also providing dominance estimates that align closely with traditional field surveys. By translating pixel-level predictions into ecologically meaningful metrics, EcoVision lays the groundwork for scalable and precise monitoring of salt marsh ecosystems.
AI-powered drone imaging can achieve near-perfect species classification and dominance mapping in salt marshes, revolutionizing ecological monitoring.
High-resolution RGB imagery acquired from low-altitude UAV surveys was processed through a modular pipeline incorporating transformer-based semantic segmentation, connected-component vegetation extraction, fine-grained species classification using a ConvNeXt architecture, and grid-based dominance scoring at 2x2m resolution. The framework targeted two ecologically significant halophytic grasses, Spartina maritima and Puccinellia maritima, and was trained using a curated and manually annotated UAV imagery, along with biodiversity imagery sourced from publicly accessible datasets. In order to identify these plants from the imagery, our segmentation yielded reliable species masks (mean IoU = 0.56; pixel-level accuracy = 0.96), while object-level classification achieved very good discrimination (F1 = 0.99). Dominance estimates closely matched quadrat-based field surveys, with mean absolute differences below 8%, preserving fine-scale spatial structure under realistic survey conditions. The developed system, named EcoVision, establishes a practical foundation for scalable, high-resolution salt marsh monitoring, demonstrating how AI-driven workflows can translate pixel-level predictions into ecologically interpretable metrics.