Low-resolution Infrared Image Target Detection based on Improved YOLOv5

Xueming Zhang; Danning Wang; Ping Tang; Heng Liu

doi:10.1109/CBASE60015.2023.10439144

Low-resolution Infrared Image Target Detection based on Improved YOLOv5

Xueming Zhang
, Danning Wang
, Ping Tang
, Heng Liu^*

^*Corresponding author for this work

School of Information and Electronics

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In the context of border defense and peacekeeping operations carried out in challenging environments, conventional visible light target detection methods prove to be poor in performance. Therefore, the utilization of infrared images detection is predominantly employed. In this study, we focus on the improvement of target detection performance of YOLOv5 model in infrared images. To enhance the original YOLOv5 network structure, we draw inspiration from BoTNet and introduce the BoT3 block and Multi Head Self Attention mechanism. The convolutional blocks in the neck were substituted with GSConv modules. Additionally, the Focal Loss was replaced with VariFocal Loss to achieve a more balanced weighting of samples. The experimental results display that the enhanced YOLOv5 model outperforms the original model and achieve better detection outcomes when applied to low-resolution infrared images.

Original language	English
Title of host publication	2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	196-200
Number of pages	5
ISBN (Electronic)	9798350331448
DOIs	https://doi.org/10.1109/CBASE60015.2023.10439144
Publication status	Published - 2023
Event	2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023 - Hybrid, Chengdu, China Duration: 3 Nov 2023 → 5 Nov 2023

Publication series

Name	2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023

Conference

Conference	2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023
Country/Territory	China
City	Hybrid, Chengdu
Period	3/11/23 → 5/11/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 16 Peace, Justice and Strong Institutions

Keywords

Infrared image
Object detection
YOLO

Access to Document

10.1109/CBASE60015.2023.10439144

Cite this

Zhang, X., Wang, D., Tang, P., & Liu, H. (2023). Low-resolution Infrared Image Target Detection based on Improved YOLOv5. In 2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023 (pp. 196-200). (2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CBASE60015.2023.10439144

Zhang, Xueming ; Wang, Danning ; Tang, Ping et al. / Low-resolution Infrared Image Target Detection based on Improved YOLOv5. 2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 196-200 (2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023).

@inproceedings{547049bfbf7b495fb12248a424e214fb,

title = "Low-resolution Infrared Image Target Detection based on Improved YOLOv5",

abstract = "In the context of border defense and peacekeeping operations carried out in challenging environments, conventional visible light target detection methods prove to be poor in performance. Therefore, the utilization of infrared images detection is predominantly employed. In this study, we focus on the improvement of target detection performance of YOLOv5 model in infrared images. To enhance the original YOLOv5 network structure, we draw inspiration from BoTNet and introduce the BoT3 block and Multi Head Self Attention mechanism. The convolutional blocks in the neck were substituted with GSConv modules. Additionally, the Focal Loss was replaced with VariFocal Loss to achieve a more balanced weighting of samples. The experimental results display that the enhanced YOLOv5 model outperforms the original model and achieve better detection outcomes when applied to low-resolution infrared images.",

keywords = "Infrared image, Object detection, YOLO",

author = "Xueming Zhang and Danning Wang and Ping Tang and Heng Liu",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023 ; Conference date: 03-11-2023 Through 05-11-2023",

year = "2023",

doi = "10.1109/CBASE60015.2023.10439144",

language = "English",

series = "2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "196--200",

booktitle = "2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023",

address = "United States",

}

Zhang, X, Wang, D, Tang, P & Liu, H 2023, Low-resolution Infrared Image Target Detection based on Improved YOLOv5. in 2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023. 2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023, Institute of Electrical and Electronics Engineers Inc., pp. 196-200, 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023, Hybrid, Chengdu, China, 3/11/23. https://doi.org/10.1109/CBASE60015.2023.10439144

Low-resolution Infrared Image Target Detection based on Improved YOLOv5. / Zhang, Xueming; Wang, Danning; Tang, Ping et al.
2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 196-200 (2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Low-resolution Infrared Image Target Detection based on Improved YOLOv5

AU - Zhang, Xueming

AU - Wang, Danning

AU - Tang, Ping

AU - Liu, Heng

PY - 2023

Y1 - 2023

N2 - In the context of border defense and peacekeeping operations carried out in challenging environments, conventional visible light target detection methods prove to be poor in performance. Therefore, the utilization of infrared images detection is predominantly employed. In this study, we focus on the improvement of target detection performance of YOLOv5 model in infrared images. To enhance the original YOLOv5 network structure, we draw inspiration from BoTNet and introduce the BoT3 block and Multi Head Self Attention mechanism. The convolutional blocks in the neck were substituted with GSConv modules. Additionally, the Focal Loss was replaced with VariFocal Loss to achieve a more balanced weighting of samples. The experimental results display that the enhanced YOLOv5 model outperforms the original model and achieve better detection outcomes when applied to low-resolution infrared images.

AB - In the context of border defense and peacekeeping operations carried out in challenging environments, conventional visible light target detection methods prove to be poor in performance. Therefore, the utilization of infrared images detection is predominantly employed. In this study, we focus on the improvement of target detection performance of YOLOv5 model in infrared images. To enhance the original YOLOv5 network structure, we draw inspiration from BoTNet and introduce the BoT3 block and Multi Head Self Attention mechanism. The convolutional blocks in the neck were substituted with GSConv modules. Additionally, the Focal Loss was replaced with VariFocal Loss to achieve a more balanced weighting of samples. The experimental results display that the enhanced YOLOv5 model outperforms the original model and achieve better detection outcomes when applied to low-resolution infrared images.

KW - Infrared image

KW - Object detection

KW - YOLO

UR - https://www.scopus.com/pages/publications/85186764299

U2 - 10.1109/CBASE60015.2023.10439144

DO - 10.1109/CBASE60015.2023.10439144

M3 - Conference contribution

AN - SCOPUS:85186764299

T3 - 2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023

SP - 196

EP - 200

BT - 2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023

Y2 - 3 November 2023 through 5 November 2023

ER -

Zhang X, Wang D, Tang P, Liu H. Low-resolution Infrared Image Target Detection based on Improved YOLOv5. In 2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 196-200. (2023 2nd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2023). doi: 10.1109/CBASE60015.2023.10439144

Low-resolution Infrared Image Target Detection based on Improved YOLOv5

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this