UAV-Driven Maritime Object Detection and Classification: Literature Review of AI Models, Data, and Performance

Laurinne Lorenna Cristina Toscano Lira de Oliveira; Gabriel Batista Reis; Ana Beatriz Alves Hougaz; Giselly Alves Reis; Lucas Amparo Barbosa

doi:10.34178/jbth.v9i7.665

Laurinne Lorenna Cristina Toscano Lira de Oliveira
Gabriel Batista Reis
Ana Beatriz Alves Hougaz
Giselly Alves Reis
Lucas Amparo Barbosa

DOI: https://doi.org/10.34178/jbth.v9i7.665

Keywords: UAV, Maritime Environment, Object Detection, Model Architecture, Machine Learning

Abstract

This literature review explores the integration of UAVs and AI in maritime surveillance, analyzing 59 peer-reviewed studies from 2020–2025. It highlights trends in object detection and classification, common methodologies, model architecture (notably YOLO), and evaluation metrics like mAP. Detection targets include boats, ships, and swimmers, with varying research focus. The review identifies technological approaches, research gaps, and growing interest in UAV-based maritime monitoring.