pith. sign in

arxiv: 2605.25942 · v1 · pith:KEWULPQMnew · submitted 2026-05-25 · 💻 cs.CV · cs.RO

LRDDv3: High-Resolution Long-Range Drone Detection Dataset with Range Information and Thermal Data

Knut Peterson , Zaid Mayers , Azmain Yousuf , Priontu Chowdhury , Asher Zaczepinski , Solmaz Arezoomandan , Reihaneh Maarefdoust , David Han This is my paper

Pith reviewed 2026-06-29 22:31 UTC · model grok-4.3

classification 💻 cs.CV cs.RO
keywords drone detectionUAV datasetlong-range detectionthermal imaginghigh-resolution datasetrange informationIR imagescomputer vision
0
0 comments X

The pith

A new dataset supplies 102,532 4K long-range drone RGB images paired with thermal data and range measurements collected across eight months.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces LRDDv3 to fill gaps in drone detection data by supplying high-resolution long-range imagery. It consists of RGB frames at 4K resolution along with paired thermal images and explicit range values for each drone. Collection spanned 128 video clips over 17 days in eight months to include varied lighting, locations, and backgrounds. This setup is meant to support training of detectors that can identify small drones at distance where prior datasets fall short. The result is positioned as a resource for safer shared airspace operations between manned aircraft and UAVs.

Core claim

The authors release LRDDv3 containing 102,532 long-range RGB images at 4K resolution sampled at 5 FPS from 128 distinct mid-flight video clips, together with range annotations for every frame and 29,630 paired 640x512 IR images, gathered across 17 collection days distributed over eight months to capture diversity in lighting scenarios, flight locations, and background elements.

What carries the argument

The LRDDv3 dataset, built around 4K RGB frames paired with 640x512 thermal images and per-frame range values, collected via a multi-month sampling strategy to provide the training material for long-range drone detectors.

If this is right

  • Detection models can locate drones at greater distances using the 4K resolution than lower-resolution datasets allow.
  • Paired thermal images enable detection under low-light or visually obscured conditions.
  • Explicit range labels support distance-aware or scale-aware detection algorithms.
  • The multi-month collection supports training of models that generalize across different environments and times of day.
  • The dataset can serve as a benchmark for comparing new long-range UAV detection methods.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The range annotations could be used to train models that output both detection and estimated distance in a single forward pass.
  • Combined RGB-IR pairs open the door to testing whether thermal data improves performance specifically when RGB contrast is low.
  • The dataset size and pairing structure could support experiments on whether multi-modal training reduces false positives at long range compared to RGB-only training.
  • Release of the full set at the provided link allows direct replication checks on whether the claimed diversity actually produces measurable gains in cross-day generalization.

Load-bearing premise

The sampling across 17 days and eight months supplies enough variation in lighting, locations, and backgrounds for models to generalize to unseen long-range drone sightings.

What would settle it

A detector trained only on LRDDv3 performs no better than chance on a new long-range test set recorded under lighting or background conditions absent from the original 17 collection days.

Figures

Figures reproduced from arXiv: 2605.25942 by Asher Zaczepinski, Azmain Yousuf, David Han, Knut Peterson, Priontu Chowdhury, Reihaneh Maarefdoust, Solmaz Arezoomandan, Zaid Mayers.

Figure 1
Figure 1. Figure 1: Examples of images in the dataset, featuring different challenges such as drones occluded by objects and foliage, drones blending into noisy [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Example images from the dataset, featuring a wide variety of backgrounds, camera angles, drone distances, and lighting conditions. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: An overview of dataset splits and distributions. (a) Shows the number of images in the training, validation, and test sets for both RGB and Thermal [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Examples of thermal image pairs from the dataset. The RGB images [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: (a) Distribution of bounding box locations for the different classes across the dataset. (b) Distribution of bounding box sizes for the different [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
read the original abstract

Unmanned Aerial Vehicles (UAVs) have quickly become common in various airspaces, representing a wide range of applications from recreation flying to commercial photography and package delivery. With the increasing prevalence of UAVs, it becomes critical that both manned and unmanned aircraft can detect UAVs and other flying objects from long range to effectively track movement and ensure safe operation in shared spaces. While several datasets have been introduced for drone detection, the need for expanded high-quality data persists, especially in the area of high-resolution long-range drone data. To address this, we introduce a high-resolution dataset of 102,532 long-range RGB images of drones, sampled at 5 FPS from 128 distinct video clips taken mid flight during 17 different data collection days spread over 8 months to ensure a wide variety of lighting scenarios, flight locations, and background elements. The dataset boasts comprehensive drone range information across the dataset, as well as 29,630 IR images, all paired with RGB counterparts from the base dataset. As one of the first drone detection datasets to leverage 4K image resolution and paired 640x512 IR images, our work represents a significant advancement to enable the detection of drones at long range. For access to the complete dataset, please visit https://research.coe.drexel.edu/ece/imaple/lrddv3/

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 0 minor

Summary. The manuscript introduces LRDDv3, a high-resolution long-range drone detection dataset containing 102,532 4K RGB frames sampled at 5 FPS from 128 video clips collected over 17 days spanning 8 months, together with 29,630 paired 640x512 IR images and range metadata for the drones.

Significance. If released as described, the dataset supplies a substantial new resource of high-resolution RGB imagery paired with thermal data and range information, addressing a documented gap in existing drone detection collections and supporting development of long-range detection systems.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their thorough review and positive recommendation to accept the manuscript. We appreciate the recognition of LRDDv3 as a valuable resource addressing gaps in long-range drone detection datasets.

Circularity Check

0 steps flagged

No significant circularity; dataset release with no derivations

full rationale

The paper is a data release describing collection of 102532 4K RGB frames, 29630 paired IR frames, and range metadata across 128 clips over 17 days. No equations, models, predictions, or fitted parameters are present. The central claim is descriptive (dataset size, resolution, pairing, and collection protocol), with no load-bearing derivation that could reduce to self-definition or self-citation. Diversity of sampling is asserted as a collection fact, not derived from any fitted result.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a dataset release paper. No free parameters are fitted, no mathematical axioms are invoked, and no new physical or conceptual entities are postulated.

pith-pipeline@v0.9.1-grok · 5814 in / 1137 out tokens · 32601 ms · 2026-06-29T22:31:21.991573+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

34 extracted references · 1 canonical work pages

  1. [1]

    Unmanned aerial vehicles (uavs): Practical aspects, applications, open challenges, security issues, and future trends,

    S. A. H. Mohsan, N. Q. H. Othman, Y . Li, M. H. Alsharif, and M. A. Khan, “Unmanned aerial vehicles (uavs): Practical aspects, applications, open challenges, security issues, and future trends,” Intelligent service robotics, vol. 16, no. 1, pp. 109–137, 2023

    2023

  2. [2]

    Optimizing uav location awareness telemetry data for low power wide area network,

    T. Karachalios, C. Xouris, and T. Orphanoudakis, “Optimizing uav location awareness telemetry data for low power wide area network,” in2022 25th Euromicro Conference on Digital System Design (DSD), pp. 885–888, 2022

    2022

  3. [3]

    Long-range drone detection dataset,

    A. Rouhi, H. Umare, S. Patal, R. Kapoor, N. Deshpande, S. Are- zoomandan, P. Shah, and D. Han, “Long-range drone detection dataset,” in2024 IEEE International Conference on Consumer Elec- tronics (ICCE), pp. 1–6, 2024

    2024

  4. [4]

    Lrddv2: Enhanced long-range drone detection dataset with range information and comprehensive real-world challenges,

    A. Rouhi, S. Patel, N. McCarthy, S. Khan, H. Khorsand, K. Lefkowitz, and D. Han, “Lrddv2: Enhanced long-range drone detection dataset with range information and comprehensive real-world challenges,” in 2024 International Symposium of Robotics Research (ISRR), pp. 1–6, 2024

    2024

  5. [5]

    Comparison of midwave versus longwave intensity signatures for infrared search and track of small rotorcraft unmanned aerial vehicles,

    N. Fudala, R. Short, J. Hewitt, C. Halford, T. Pace, D. Manville, M. Theisen, C. Fearing, A. Dapore, C. Olson, G. Tener, D. Gaudiosi, K. Richardson, and R. Driggers, “Comparison of midwave versus longwave intensity signatures for infrared search and track of small rotorcraft unmanned aerial vehicles,”Optical Engineering, vol. 58, no. 4, p. 043107, 2019

    2019

  6. [6]

    Investigation of uav de- tection in images with complex backgrounds and rainy artifacts,

    A. Munir, A. J. Siddiqui, and S. Anwar, “Investigation of uav de- tection in images with complex backgrounds and rainy artifacts,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops, pp. 221–230, January 2024

    2024

  7. [7]

    Visual signature reduction of unmanned aerial vehicles,

    Z. W. Zhong, Z. X. Ma, Jayawijayaningtiyas, and J. H. H. Ngoh, “Visual signature reduction of unmanned aerial vehicles,” inTarget and Background Signatures II(K. U. Stein and R. H. M. A. Schleijpen, eds.), vol. 9997, p. 999707, International Society for Optics and Photonics, SPIE, 2016

    2016

  8. [8]

    Vision- based uav detection in complex backgrounds and rainy conditions,

    P. Singh, K. Gupta, A. K. Jain, Vishakha, A. Jain, and A. Jain, “Vision- based uav detection in complex backgrounds and rainy conditions,” in2024 2nd International Conference on Disruptive Technologies (ICDT), pp. 1097–1102, 2024

    2024

  9. [9]

    Anti-occlusion uav tracking algorithm with a low-altitude complex background by integrating attention mechanism,

    C. Wang, Z. Shi, L. Meng, J. Wang, T. Wang, Q. Gao, and E. Wang, “Anti-occlusion uav tracking algorithm with a low-altitude complex background by integrating attention mechanism,”Drones, vol. 6, no. 6, 2022

    2022

  10. [10]

    Enhanced long-range uav detection: Leveraging slicing aided hyper inference with yolov8,

    H. Khorsand, S. Arezoomandan, and D. K. Han, “Enhanced long-range uav detection: Leveraging slicing aided hyper inference with yolov8,” in2025 IEEE International Conference on Consumer Electronics (ICCE), pp. 1–6, 2025

    2025

  11. [11]

    A dataset for multi-sensor drone detection,

    F. Svanstr ¨om, F. Alonso-Fernandez, and C. Englund, “A dataset for multi-sensor drone detection,”Data in Brief, vol. 39, p. 107521, 2021

    2021

  12. [12]

    Birdrone: A novel dataset for enhanced drone and bird detection using yolov9,

    F. N. M. Zamri, T. S. Gunawan, S. H. Yusoff, Y . M. Mustafah, M. Kar- tiwi, and N. M. Yusoff, “Birdrone: A novel dataset for enhanced drone and bird detection using yolov9,” in2024 IEEE 10th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA), pp. 1–6, 2024

    2024

  13. [13]

    Visiodect: a novel approach to drone detection using cbam-integrated yolo and gelan-e models,

    S. B. Islam, M. E. Chowdhury, M. Hasan-Zia, S. B. A. Kashem, M. E. Majid, A. K. Ansaruddin Kunju, A. Khandakar, A. Ashraf, M. Nashbat,et al., “Visiodect: a novel approach to drone detection using cbam-integrated yolo and gelan-e models,”Neural Computing and Applications, pp. 1–24, 2025

    2025

  14. [14]

    Real world object detection dataset for quadcopter unmanned aerial vehicle detection,

    M. L. Pawelczyk and M. Wojtyra, “Real world object detection dataset for quadcopter unmanned aerial vehicle detection,”IEEE Access, vol. 8, pp. 174394–174409, 2020

    2020

  15. [15]

    Syndronevi- sion: A synthetic dataset for image-based drone detection,

    T. R. Lenhard, A. Weinmann, K. Franke, and T. Koch, “Syndronevi- sion: A synthetic dataset for image-based drone detection,” in2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 7637–7647, 2025

    2025

  16. [16]

    Vision-based anti-uav detec- tion and tracking,

    J. Zhao, J. Zhang, D. Li, and D. Wang, “Vision-based anti-uav detec- tion and tracking,”IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 12, pp. 25323–25334, 2022

    2022

  17. [17]

    Autonomous drone hunter operating by deep learning and all-onboard computations in gps-denied environments,

    P. M. Wyder, Y .-S. Chen, A. J. Lasrado, R. J. Pelles, R. Kwiatkowski, E. O. A. Comas, R. Kennedy, A. Mangla, Z. Huang, X. Hu, Z. Xiong, T. Aharoni, T.-C. Chuang, and H. Lipson, “Autonomous drone hunter operating by deep learning and all-onboard computations in gps-denied environments,”PLOS ONE, vol. 14, pp. 1–18, 11 2019

    2019

  18. [18]

    Fred: The florence rgb-event drone dataset,

    G. Magrini, N. Marini, F. Becattini, L. Berlincioni, N. Biondi, P. Pala, and A. Del Bimbo, “Fred: The florence rgb-event drone dataset,”arXiv preprint arXiv:2506.05163, 2025

    work page arXiv 2025

  19. [19]

    Global-local mav detection under challenging conditions based on appearance and motion,

    H. Guo, Y . Zheng, Y . Zhang, Z. Gao, and S. Zhao, “Global-local mav detection under challenging conditions based on appearance and motion,”IEEE Transactions on Intelligent Transportation Systems, 2024

    2024

  20. [20]

    Reconstruction of 3d flight trajectories from ad-hoc camera networks,

    J. Li, J. Murray, D. Ismaili, K. Schindler, and C. Albl, “Reconstruction of 3d flight trajectories from ad-hoc camera networks,” in2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 1621–1628, 2020

    2020

  21. [21]

    uav detect dataset

    GET, “uav detect dataset.”https://universe.roboflow.co m/get/uav-detect-pfiqs, jan 2023. visited on 2025-09-15

    2023

  22. [22]

    Air-to-air visual detection of micro-uavs: An experimental evaluation of deep learning,

    Y . Zheng, Z. Chen, D. Lv, Z. Li, Z. Lan, and S. Zhao, “Air-to-air visual detection of micro-uavs: An experimental evaluation of deep learning,” IEEE Robotics and Automation Letters, vol. 6, no. 2, pp. 1020–1027, 2021

    2021

  23. [23]

    The drone-vs-bird detection grand challenge at icassp 2023: A review of methods and results,

    A. Coluccia, A. Fascista, L. Sommer, A. Schumann, A. Dimou, and D. Zarpalas, “The drone-vs-bird detection grand challenge at icassp 2023: A review of methods and results,”IEEE Open Journal of Signal Processing, vol. 5, pp. 766–779, 2024

    2023

  24. [24]

    Airborne Object Tracking Dataset,

    Amazon, “Airborne Object Tracking Dataset,” July 2021

    2021

  25. [25]

    A. A. Bakar, H. Mujtaba, M. S. Awan, K. Hasnain, and W. Mumtaz, “Amd3ir,” 2025

    2025

  26. [26]

    Ev-flying: an event-based dataset for in-the-wild recognition of flying objects,

    G. Magrini, F. Becattini, G. Colombo, and P. Pala, “Ev-flying: an event-based dataset for in-the-wild recognition of flying objects,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pp. 4956–4964, June 2025

    2025

  27. [27]

    Fast and robust uav to uav detection and tracking from video,

    J. Li, D. H. Ye, M. Kolsch, J. P. Wachs, and C. A. Bouman, “Fast and robust uav to uav detection and tracking from video,” IEEE Transactions on Emerging Topics in Computing, vol. 10, no. 3, pp. 1519–1531, 2022

    2022

  28. [28]

    MMFW-UA V Dataset,

    Y . Liu, Z. Sun, L. Xi, L. Zhang, W. Dong, C. Chen, M. Lu, H. Fu, and F. Deng, “MMFW-UA V Dataset,” Nov. 2024

    2024

  29. [29]

    Dac- sdc low power object detection challenge for uav applications,

    X. Xu, X. Zhang, B. Yu, X. S. Hu, C. Rowen, J. Hu, and Y . Shi, “Dac- sdc low power object detection challenge for uav applications,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 2, pp. 392–403, 2021

    2021

  30. [30]

    Ugˆ 2: A video benchmark for assessing the impact of image restoration and enhancement on automatic visual recognition,

    R. G. Vidal, S. Banerjee, K. Grm, V . Struc, and W. J. Scheirer, “Ugˆ 2: A video benchmark for assessing the impact of image restoration and enhancement on automatic visual recognition,” in2018 IEEE Winter Conference on Applications of Computer Vision (WACV), pp. 1597– 1606, IEEE, 2018

    2018

  31. [31]

    Ultralytics yolo11,

    G. Jocher and J. Qiu, “Ultralytics yolo11,” 2024

    2024

  32. [32]

    Slicing aided hyper inference and fine-tuning for small object detection,

    F. C. Akyon, S. Onur Altinuc, and A. Temizel, “Slicing aided hyper inference and fine-tuning for small object detection,” in2022 IEEE International Conference on Image Processing (ICIP), IEEE, Oct. 2022

    2022

  33. [33]

    Labelimg

    Tzutalin and HumanSignal, “Labelimg.”https://github.com /HumanSignal/labelImg, 2024. Version available at GitHub

    2024

  34. [34]

    Hrformer: High-resolution transformer for dense predic- tion,

    Y . Yuan, R. Fu, L. Huang, W. Lin, C. Zhang, X. Chen, and J. Wang, “Hrformer: High-resolution transformer for dense predic- tion,”NeurIPS, 2021

    2021