fairfolio
Saved Projects

MARLIN: An ROV-based AI System for Detection and Density Mapping of Invasive Sargassum horneri in Monterey Bay

CSEF · 2026 Environmental Engineering (Senior Division)

Cynthia Zhang, Cassia Seem

Overview

Rapidly emerging blooms of Sargassum horneri threaten kelp forest ecosystems and shoreline biodiversity in Monterey Bay, California. This study introduces MARLIN, an AI-robotics pipeline operating in real time for scalable autonomous underwater surveys of benthic organisms such as invasive macroalgae. The system integrates a custom-built remotely operated vehicle (ROV) equipped with a 1920 × 1080 low-light video camera, subsea LED lighting, and onboard deep-learning inference to detect S. horneri in situ. Approximately 700 field images were manually annotated with bounding boxes and class labels to fine-tune a YOLOv8 object detection model. Model performance was evaluated using precision, recall, F1-score, and mean average precision ([email protected]). Detection under degraded underwater conditions was compared between a baseline model and an enhanced pipeline incorporating image restoration and data augmentation. During semi-autonomous field deployment at McAbee Beach under variable visibility conditions, MARLIN achieved 87.4% precision, 96.3% recall, and an F1 score of 91.6%. Surveying a 20 m × 10 m transect in 3 minutes, the system autonomously detected 104 true positives and generated a spatial density map revealing clumped invasion patterns. A second density-mapping analysis yielded 208 true positives, 16 false positives, and 7 false negatives, corresponding to 91.7% precision, 96.2% recall, and a 93.9% F1 score. MARLIN’s ability to navigate transects, perform onboard inference, and reconstruct spatial density metrics enables rapid, repeatable monitoring while reducing dependence on diver surveys.

Competition history

  • CSEF 2026 Environmental Engineering (Senior Division) · Entry S-11-17

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

Source: California Science & Engineering Fair public projects

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google