Swin‐YOLOX for autonomous and accurate drone visual landing

Abstract As UAVs are more and more widely used in military and civilian fields, their intelligent applications have also been developed rapidly. However, high‐precision autonomous landing is still an industry challenge. GPS‐based methods will not work in places where GPS signals are not available; m...

Full description

Saved in:
Bibliographic Details
Main Authors: Rongbin Chen, Ying Xu, Mohamad Sabri bin Sinal, Dongsheng Zhong, Xinru Li, Bo Li, Yadong Guo, Qingjia Luo
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:IET Image Processing
Subjects:
Online Access:https://doi.org/10.1049/ipr2.13282
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850118441422290944
author Rongbin Chen
Ying Xu
Mohamad Sabri bin Sinal
Dongsheng Zhong
Xinru Li
Bo Li
Yadong Guo
Qingjia Luo
author_facet Rongbin Chen
Ying Xu
Mohamad Sabri bin Sinal
Dongsheng Zhong
Xinru Li
Bo Li
Yadong Guo
Qingjia Luo
author_sort Rongbin Chen
collection DOAJ
description Abstract As UAVs are more and more widely used in military and civilian fields, their intelligent applications have also been developed rapidly. However, high‐precision autonomous landing is still an industry challenge. GPS‐based methods will not work in places where GPS signals are not available; multi‐sensor combination navigation is difficult to be widely used because of the high equipment requirements; traditional vision‐based methods are sensitive to scale transformation, background complexity and occlusion, which affect the detection performance. In this paper, we address these problems and apply deep learning methods to target detection in the UAV landing phase. Firstly, we optimize the backbone network of YOLOX and propose the Swin Transformer based YOLOX (Swin‐YOLOX) UAV landing visual positioning algorithm. Secondly, based on the UAV‐VPD database, a batch of actual acquisition data is added to build the UAV‐VPDV2 database by AI annotation method. And finally, the RBN data batch normalization method is used to improve the performance of the model in extracting effective features from the data. Extensive experiments have shown that the AP50 of the proposed method can reach 98.7%, which is superior to other detection models, with a detection speed of 38.4 frames/second, and can meet the requirements of real‐time detection.
format Article
id doaj-art-9732be50acfa4e489a970099160d05c3
institution OA Journals
issn 1751-9659
1751-9667
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series IET Image Processing
spelling doaj-art-9732be50acfa4e489a970099160d05c32025-08-20T02:35:53ZengWileyIET Image Processing1751-96591751-96672024-12-0118144731474410.1049/ipr2.13282Swin‐YOLOX for autonomous and accurate drone visual landingRongbin Chen0Ying Xu1Mohamad Sabri bin Sinal2Dongsheng Zhong3Xinru Li4Bo Li5Yadong Guo6Qingjia Luo7College of Information Engineering, Jiangmen Polytechnic Jiangmen Guangdong ChinaDepartment of Intelligent Manufacturing Wuyi University Jiangmen ChinaSchool of Computing Universiti Utara Malaysia Kedah MalaysiaDepartment of Intelligent Manufacturing Wuyi University Jiangmen ChinaDepartment of Intelligent Manufacturing Wuyi University Jiangmen ChinaDepartment of Intelligent Manufacturing Wuyi University Jiangmen ChinaCollege of Information Engineering, Jiangmen Polytechnic Jiangmen Guangdong ChinaCollege of Information Engineering, Jiangmen Polytechnic Jiangmen Guangdong ChinaAbstract As UAVs are more and more widely used in military and civilian fields, their intelligent applications have also been developed rapidly. However, high‐precision autonomous landing is still an industry challenge. GPS‐based methods will not work in places where GPS signals are not available; multi‐sensor combination navigation is difficult to be widely used because of the high equipment requirements; traditional vision‐based methods are sensitive to scale transformation, background complexity and occlusion, which affect the detection performance. In this paper, we address these problems and apply deep learning methods to target detection in the UAV landing phase. Firstly, we optimize the backbone network of YOLOX and propose the Swin Transformer based YOLOX (Swin‐YOLOX) UAV landing visual positioning algorithm. Secondly, based on the UAV‐VPD database, a batch of actual acquisition data is added to build the UAV‐VPDV2 database by AI annotation method. And finally, the RBN data batch normalization method is used to improve the performance of the model in extracting effective features from the data. Extensive experiments have shown that the AP50 of the proposed method can reach 98.7%, which is superior to other detection models, with a detection speed of 38.4 frames/second, and can meet the requirements of real‐time detection.https://doi.org/10.1049/ipr2.13282computer visionimage processingimage recognitionremote sensing
spellingShingle Rongbin Chen
Ying Xu
Mohamad Sabri bin Sinal
Dongsheng Zhong
Xinru Li
Bo Li
Yadong Guo
Qingjia Luo
Swin‐YOLOX for autonomous and accurate drone visual landing
IET Image Processing
computer vision
image processing
image recognition
remote sensing
title Swin‐YOLOX for autonomous and accurate drone visual landing
title_full Swin‐YOLOX for autonomous and accurate drone visual landing
title_fullStr Swin‐YOLOX for autonomous and accurate drone visual landing
title_full_unstemmed Swin‐YOLOX for autonomous and accurate drone visual landing
title_short Swin‐YOLOX for autonomous and accurate drone visual landing
title_sort swin yolox for autonomous and accurate drone visual landing
topic computer vision
image processing
image recognition
remote sensing
url https://doi.org/10.1049/ipr2.13282
work_keys_str_mv AT rongbinchen swinyoloxforautonomousandaccuratedronevisuallanding
AT yingxu swinyoloxforautonomousandaccuratedronevisuallanding
AT mohamadsabribinsinal swinyoloxforautonomousandaccuratedronevisuallanding
AT dongshengzhong swinyoloxforautonomousandaccuratedronevisuallanding
AT xinruli swinyoloxforautonomousandaccuratedronevisuallanding
AT boli swinyoloxforautonomousandaccuratedronevisuallanding
AT yadongguo swinyoloxforautonomousandaccuratedronevisuallanding
AT qingjialuo swinyoloxforautonomousandaccuratedronevisuallanding