Cargando…

Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm

Achieving a real-time and accurate detection of drones in natural environments is essential for the interception of drones intruding into high-security areas. However, a rapid and accurate detection of drones is difficult because of their small size and fast speed. In this paper a drone detection me...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Xuetao, Fan, Kuangang, Hou, Haonan, Liu, Chuankai
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9785146/
https://www.ncbi.nlm.nih.gov/pubmed/36557498
http://dx.doi.org/10.3390/mi13122199
_version_ 1784857977909411840
author Zhang, Xuetao
Fan, Kuangang
Hou, Haonan
Liu, Chuankai
author_facet Zhang, Xuetao
Fan, Kuangang
Hou, Haonan
Liu, Chuankai
author_sort Zhang, Xuetao
collection PubMed
description Achieving a real-time and accurate detection of drones in natural environments is essential for the interception of drones intruding into high-security areas. However, a rapid and accurate detection of drones is difficult because of their small size and fast speed. In this paper a drone detection method as proposed by pruning the convolutional channel and residual structures of YOLOv3-SPP3. First, the k-means algorithm was used to cluster label the boxes. Second, the channel and shortcut layer pruning algorithm was used to prune the model. Third, the model was fine tuned to achieve a real-time detection of drones. The experimental results obtained by using the Ubuntu server under the Python 3.6 environment show that the YOLOv3-SPP3 algorithm is better than YOLOV3, Tiny-YOLOv3, CenterNet, SSD300, and faster R-CNN. There is significant compression in the size, the maximum compression factor is 20.1 times, the maximum detection speed is increased by 10.2 times, the maximum map value is increased by 15.2%, and the maximum precision is increased by 16.54%. The proposed algorithm achieves the mAP score of 95.15% and the detection speed of 112 f/s, which can meet the requirements of the real-time detection of UAVs.
format Online
Article
Text
id pubmed-9785146
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-97851462022-12-24 Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm Zhang, Xuetao Fan, Kuangang Hou, Haonan Liu, Chuankai Micromachines (Basel) Article Achieving a real-time and accurate detection of drones in natural environments is essential for the interception of drones intruding into high-security areas. However, a rapid and accurate detection of drones is difficult because of their small size and fast speed. In this paper a drone detection method as proposed by pruning the convolutional channel and residual structures of YOLOv3-SPP3. First, the k-means algorithm was used to cluster label the boxes. Second, the channel and shortcut layer pruning algorithm was used to prune the model. Third, the model was fine tuned to achieve a real-time detection of drones. The experimental results obtained by using the Ubuntu server under the Python 3.6 environment show that the YOLOv3-SPP3 algorithm is better than YOLOV3, Tiny-YOLOv3, CenterNet, SSD300, and faster R-CNN. There is significant compression in the size, the maximum compression factor is 20.1 times, the maximum detection speed is increased by 10.2 times, the maximum map value is increased by 15.2%, and the maximum precision is increased by 16.54%. The proposed algorithm achieves the mAP score of 95.15% and the detection speed of 112 f/s, which can meet the requirements of the real-time detection of UAVs. MDPI 2022-12-11 /pmc/articles/PMC9785146/ /pubmed/36557498 http://dx.doi.org/10.3390/mi13122199 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Zhang, Xuetao
Fan, Kuangang
Hou, Haonan
Liu, Chuankai
Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm
title Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm
title_full Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm
title_fullStr Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm
title_full_unstemmed Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm
title_short Real-Time Detection of Drones Using Channel and Layer Pruning, Based on the YOLOv3-SPP3 Deep Learning Algorithm
title_sort real-time detection of drones using channel and layer pruning, based on the yolov3-spp3 deep learning algorithm
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9785146/
https://www.ncbi.nlm.nih.gov/pubmed/36557498
http://dx.doi.org/10.3390/mi13122199
work_keys_str_mv AT zhangxuetao realtimedetectionofdronesusingchannelandlayerpruningbasedontheyolov3spp3deeplearningalgorithm
AT fankuangang realtimedetectionofdronesusingchannelandlayerpruningbasedontheyolov3spp3deeplearningalgorithm
AT houhaonan realtimedetectionofdronesusingchannelandlayerpruningbasedontheyolov3spp3deeplearningalgorithm
AT liuchuankai realtimedetectionofdronesusingchannelandlayerpruningbasedontheyolov3spp3deeplearningalgorithm