RSTS-YOLOv5: An Improved Object Detector for Drone-Captured Images

Juan Xiu Liu, Jiachen Li, Ruqian Hao, Yanlong Yang, Jing Ming Zhang, Xiangzhou Wang, Guoming Lu, Ping Zhang, Jing zhang, Yong Liu, Lin Liu, Xingguo Wang, Hao Deng, Dongdong Wang, Xiaohui Du*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Despite the tremendous progress in object detection in recent years, object detection on drone-captured images is still a great challenge because of the large number of small objects that appear densely and obscure each other in drone-captured images. In order to solve the problem of difficult object detection on drone-captured images, we propose a robust and efficient deep learning network RSTS-YOLOv5. We constructed Res swin transformer stage (RSTS) based on Swin-Transformer stage to extract global and contextual information and embedded it in YOLOv5x to explore the position of the transformer-based structure added in the detection network. In addition, we propose a multi-scale data augmentation for object detection on drone-captured images, which can enhance the robustness of the model for different scale objects without introducing additional computations. Experimental results show that our proposed RSTS-YOLOv5 achieves a mAP of 34.72% on the VisDrone test-dev subset and 34.84% on the validation-dev subset. Specifically, RSTS-YOLOv5 generalizes well on various drone-captured scenes, and is extremely competitive in object detection tasks on drone-captured images.

Original languageEnglish
Title of host publicationProceedings of 2023 11th China Conference on Command and Control
PublisherSpringer Science and Business Media Deutschland GmbH
Pages355-366
Number of pages12
ISBN (Print)9789819990207
DOIs
Publication statusPublished - 2024
Externally publishedYes
Event11th China Conference on Command and Control, C2 2023 - Beijing, China
Duration: 24 Oct 202325 Oct 2023

Publication series

NameLecture Notes in Electrical Engineering
Volume1124 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

Conference11th China Conference on Command and Control, C2 2023
Country/TerritoryChina
CityBeijing
Period24/10/2325/10/23

Keywords

  • Deep learning
  • Drone-captured images
  • Multi-scale data augmentation
  • Tiny object detection

Fingerprint

Dive into the research topics of 'RSTS-YOLOv5: An Improved Object Detector for Drone-Captured Images'. Together they form a unique fingerprint.

Cite this