A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection

Uncrewed aerial vehicles (UAVs) have been widely adopted across various domains; however, their potential for malicious use cases, including espionage, illicit trafficking, and unauthorized surveillance, has escalated security threats. This growing risk underscores the urgent need for precise and ti...

Full description

Saved in:
Bibliographic Details
Main Authors: Sungkyu Jung, Jaeyeon Jang, Chang Ouk Kim
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11084805/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849417179239284736
author Sungkyu Jung
Jaeyeon Jang
Chang Ouk Kim
author_facet Sungkyu Jung
Jaeyeon Jang
Chang Ouk Kim
author_sort Sungkyu Jung
collection DOAJ
description Uncrewed aerial vehicles (UAVs) have been widely adopted across various domains; however, their potential for malicious use cases, including espionage, illicit trafficking, and unauthorized surveillance, has escalated security threats. This growing risk underscores the urgent need for precise and timely detection systems that can ensure robust countermeasures against UAV-related threats. In response to this demand, various studies have explored deep learning-based UAV detection approaches that can attain increased detection accuracy and efficiency. However, several major challenges remain in this domain, including the difficulty of detecting small UAVs and the increased computational cost associated with achieving performance improvements for deep learning models. To address the frequent misclassification of small UAVs in one-stage object detectors, this study proposes a two-step detection method, as illustrated in the graphical abstract, which summarizes the proposed approach. First, YOLO object detection model identifies candidate small UAVs. Then, a Vision Transformer classifier refines these results by filtering out false positives. This integrated approach significantly improves detection accuracy for small UAVs that are easily confused with other flying objects. Furthermore, we propose a compression strategy that integrates pruning, weight initialization, and knowledge distillation (KD) to reduce the computational costs of real-time applications. Experimental results demonstrate that the proposed two-step method outperforms a diverse set of state-of-the-art object detection and classification models. Specifically, in experiments where multiple types of aerial vehicles are present, the proposed method achieves a 5.6% improvement in mean average precision (mAP) compared to conventional object detection models. Additionally, our compression approach preserves critical UAV features by selectively removing low-importance parameters, significantly reducing the degree of redundancy while minimizing the induced detection performance loss.
format Article
id doaj-art-bc656bf108744c949b19d6ac151c6c6b
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-bc656bf108744c949b19d6ac151c6c6b2025-08-20T03:32:55ZengIEEEIEEE Access2169-35362025-01-011312877112878810.1109/ACCESS.2025.359047311084805A Lightweight Two-Step Detection Method for Real-Time Small UAV DetectionSungkyu Jung0https://orcid.org/0009-0005-2940-170XJaeyeon Jang1https://orcid.org/0000-0001-6255-2044Chang Ouk Kim2https://orcid.org/0000-0002-6936-5409Department of Industrial Engineering, Yonsei University, Seoul, South KoreaDepartment of Data Science, The Catholic University of Korea, Bucheon, South KoreaDepartment of Industrial Engineering, Yonsei University, Seoul, South KoreaUncrewed aerial vehicles (UAVs) have been widely adopted across various domains; however, their potential for malicious use cases, including espionage, illicit trafficking, and unauthorized surveillance, has escalated security threats. This growing risk underscores the urgent need for precise and timely detection systems that can ensure robust countermeasures against UAV-related threats. In response to this demand, various studies have explored deep learning-based UAV detection approaches that can attain increased detection accuracy and efficiency. However, several major challenges remain in this domain, including the difficulty of detecting small UAVs and the increased computational cost associated with achieving performance improvements for deep learning models. To address the frequent misclassification of small UAVs in one-stage object detectors, this study proposes a two-step detection method, as illustrated in the graphical abstract, which summarizes the proposed approach. First, YOLO object detection model identifies candidate small UAVs. Then, a Vision Transformer classifier refines these results by filtering out false positives. This integrated approach significantly improves detection accuracy for small UAVs that are easily confused with other flying objects. Furthermore, we propose a compression strategy that integrates pruning, weight initialization, and knowledge distillation (KD) to reduce the computational costs of real-time applications. Experimental results demonstrate that the proposed two-step method outperforms a diverse set of state-of-the-art object detection and classification models. Specifically, in experiments where multiple types of aerial vehicles are present, the proposed method achieves a 5.6% improvement in mean average precision (mAP) compared to conventional object detection models. Additionally, our compression approach preserves critical UAV features by selectively removing low-importance parameters, significantly reducing the degree of redundancy while minimizing the induced detection performance loss.https://ieeexplore.ieee.org/document/11084805/Deep learningknowledge distillationmodel compressionobject detectionpruningUAV
spellingShingle Sungkyu Jung
Jaeyeon Jang
Chang Ouk Kim
A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection
IEEE Access
Deep learning
knowledge distillation
model compression
object detection
pruning
UAV
title A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection
title_full A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection
title_fullStr A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection
title_full_unstemmed A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection
title_short A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection
title_sort lightweight two step detection method for real time small uav detection
topic Deep learning
knowledge distillation
model compression
object detection
pruning
UAV
url https://ieeexplore.ieee.org/document/11084805/
work_keys_str_mv AT sungkyujung alightweighttwostepdetectionmethodforrealtimesmalluavdetection
AT jaeyeonjang alightweighttwostepdetectionmethodforrealtimesmalluavdetection
AT changoukkim alightweighttwostepdetectionmethodforrealtimesmalluavdetection
AT sungkyujung lightweighttwostepdetectionmethodforrealtimesmalluavdetection
AT jaeyeonjang lightweighttwostepdetectionmethodforrealtimesmalluavdetection
AT changoukkim lightweighttwostepdetectionmethodforrealtimesmalluavdetection