LRDDv2: Enhanced Long-Range Drone Detection Dataset with Range Information and Comprehensive Real-World Challenges

2025-08-05

AI Summary

Detecting and ranging small, distant drones—especially in complex urban environments—remains challenging due to limited visual cues, occlusion, and the absence of reliable distance supervision. Method: We introduce LRDDv2, the largest long-range drone detection dataset to date, comprising 39,516 high-resolution images with meticulous manual annotations enabling detection of targets under 50 pixels. Crucially, we incorporate over 8,000 samples with precise monocular distance labels—enabling, for the first time, joint detection and single-view depth estimation via structured annotation. Data collection spans diverse weather conditions, illumination levels, viewing angles, and operational ranges (up to >1.5 km). Contribution/Results: LRDDv2 significantly enhances environmental diversity and practicality, establishing a new benchmark and paradigm for vision-based long-range drone perception, airspace security monitoring, and autonomous collision avoidance research.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Application Domains: Mobility, Driving & Flight

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Large-scale security measurements

Abstract

The exponential growth in Unmanned Aerial Vehicles (UAVs) usage underscores the critical need of detecting them at extended distances to ensure safe operations, especially in densely populated areas. Despite the tremendous advances made in computer vision through deep learning, the detection of these small airborne objects remains a formidable challenge. While several datasets have been developed specifically for drone detection, the need for a more extensive and diverse collection of drone image data persists, particularly for long-range detection under varying environmental conditions. We introduce here the Long Range Drone Detection (LRDD) Version 2 dataset, comprising 39,516 meticulously annotated images, as a second release of the LRDD dataset released previously. The LRDDv2 dataset enhances the LRDDv1 by incorporating a greater variety of images, providing a more diverse and comprehensive resource for drone detection research. What sets LRDDv2 apart is its inclusion of target range information for over 8,000 images, making it possible to develop algorithms for drone range estimation. Tailored for long-range aerial object detection, the majority of LRDDv2's dataset consists of images capturing drones with 50 or fewer pixels in 1080p resolution. For access to the complete Long-Range Drone Detection Dataset (LRDD)v2, please visit https://research.coe.drexel.edu/ece/imaple/lrddv2/ .

Problem

• Detecting small drones at long ranges remains challenging
• Existing drone detection datasets lack diversity and range information
• Need for improved datasets to enhance drone detection algorithms

Innovation

• Enhanced dataset with 39,516 annotated images
• Includes target range information for 8,000 images
• Focuses on long-range detection with small drones