Maritime Small Object Detection from UAVs using Deep Learning with Altitude-Aware Dynamic Tiling

TL;DR

This paper introduces an altitude-aware dynamic tiling method for UAV-based small object detection in maritime environments, significantly improving accuracy and speed by adaptively processing images based on altitude.

Contribution

The novel altitude-aware adaptive tiling approach enhances small object detection performance and efficiency in UAV maritime imagery, outperforming static tiling methods.

Findings

38% increase in mAP for small objects

More than double inference speed

Effective under diverse maritime conditions

Abstract

Unmanned Aerial Vehicles (UAVs) are crucial in Search and Rescue (SAR) missions due to their ability to monitor vast maritime areas. However, small objects often remain difficult to detect from high altitudes due to low object-to-background pixel ratios. We propose an altitude-aware dynamic tiling method that scales and adaptively subdivides the image into tiles for enhanced small object detection. By integrating altitude-dependent scaling with an adaptive tiling factor, we reduce unnecessary computation while maintaining detection performance. Tested on the SeaDronesSee dataset [1] with YOLOv5 [2] and Slicing Aided Hyper Inference (SAHI) framework [3], our approach improves Mean Average Precision (mAP) for small objects by 38% compared to a baseline and achieves more than double the inference speed compared to static tiling. This approach enables more efficient and accurate UAV-based SAR operations under diverse conditions.