Underwater Inspection Platform for Vision-Based Biodiversity Identification
Manduca, Gianluca Santaera, Gaspare Natoli, Ada De Masi, Giulia Stefanini, Cesare Romano, Donato
Manduca, Gianluca
Santaera, Gaspare
Natoli, Ada
De Masi, Giulia
Stefanini, Cesare
Romano, Donato
Supervisor
Department
Robotics
Embargo End Date
Type
Conference proceeding
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Marine biodiversity monitoring is crucial for understanding and mitigating the impacts of environmental change, overfishing, and habitat degradation. In this work, we present an autonomous, fish-inspired robotic platform designed for in situ identification of marine species. The system allows for realtime perception and classification without reliance on external computation. A lightweight convolutional neural network (CNN) was trained on a custom dataset comprising over 2,000 annotated images of marine organisms, including dolphins, belugas, jellyfish, sharks, and rays, collected from public datasets, fieldwork, and major aquariums worldwide. The combination of agile underwater navigation and embedded neural processing offers a scalable and non-invasive solution for marine biodiversity inspection. The synergy between robotic platforms and deep learning represents a fundamental advancement toward autonomous, in situ ecological monitoring in dynamic and unstructured aquatic habitats.
Citation
G. Manduca, G. Santaera, A. Natoli, G. De Masi, C. Stefanini and D. Romano, "Underwater Inspection Platform for Vision-Based Biodiversity Identification," 2025 IEEE International Workshop on Metrology for the Sea; Learning to Measure Sea Health Parameters (MetroSea), Genoa, Italy, 2025, pp. 45-49, doi: 10.1109/MetroSea66681.2025.11245779.
Source
2025 IEEE International Workshop on Metrology for the Sea; Learning to Measure Sea Health Parameters (MetroSea)
Conference
IEEE International Workshop on Metrology for the Sea; Learning to Measure Sea Health Parameters (MetroSea)
Keywords
Autonomous Underwater Vehicles, Deep Learning, Marine Biodiversity Monitoring, Embedded Vision, Biomimetic Robotics
Subjects
Source
IEEE International Workshop on Metrology for the Sea; Learning to Measure Sea Health Parameters (MetroSea)
Publisher
IEEE