Staff View: Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware

Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware

The advancement of unmanned aerial vehicles (UAVs) has drawn researchers to update object detection algorithms for better accuracy and computation performance. Previous works applying deep learning models for object detection applications required high graphics processing unit (GPU) computation powe...

Full description

Saved in:

Bibliographic Details
Main Authors:	Andika, Johan Lela, Khairuddin, Anis Salwa Mohd, Ramiah, Harikrishnan, Kanesan, Jeevan
Format:	Article
Published:	Springer 2024
Subjects:	QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering
Online Access:	http://eprints.um.edu.my/45256/ https://doi.org/10.1007/s11554-024-01457-1
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.um.eprints.45256
record_format	eprints
spelling	my.um.eprints.452562024-09-30T05:04:00Z http://eprints.um.edu.my/45256/ Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware Andika, Johan Lela Khairuddin, Anis Salwa Mohd Ramiah, Harikrishnan Kanesan, Jeevan QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering The advancement of unmanned aerial vehicles (UAVs) has drawn researchers to update object detection algorithms for better accuracy and computation performance. Previous works applying deep learning models for object detection applications required high graphics processing unit (GPU) computation power. Generally, object detection models suffer trade-off between accuracy and model size where the relationship is not always linear in deep learning models. Various factors such as architectural design, optimization techniques, and dataset characteristics can significantly influence the accuracy, model size, and computation cost in adopting object detection models for low-cost embedded devices. Hence, it is crucial to employ lightweight object detection models for real-time object identification for the solution to be sustainable. In this work, an improved feature extraction network is proposed by incorporating an efficient long-range aggregation network for vehicle detection (ELAN-VD) in the backbone layer. The architecture improvement in YOLOv7-tiny model is proposed to improve the accuracy of detecting small vehicles in the aerial image. Besides that, the image size output of the second and third prediction boxes is upscaled for better performance. This study showed that the proposed method yields a mean average precision (mAP) of 57.94%, which is higher than that of the conventional YOLOv7-tiny. In addition, the proposed model showed significant performance when compared to previous works, making it viable for application in low-cost embedded devices. Springer 2024-05 Article PeerReviewed Andika, Johan Lela and Khairuddin, Anis Salwa Mohd and Ramiah, Harikrishnan and Kanesan, Jeevan (2024) Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware. Journal of Real-Time Image Processing, 21 (3). p. 77. ISSN 1861-8200, DOI https://doi.org/10.1007/s11554-024-01457-1 <https://doi.org/10.1007/s11554-024-01457-1>. https://doi.org/10.1007/s11554-024-01457-1 10.1007/s11554-024-01457-1
institution	Universiti Malaya
building	UM Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Malaya
content_source	UM Research Repository
url_provider	http://eprints.um.edu.my/
topic	QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering
spellingShingle	QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering Andika, Johan Lela Khairuddin, Anis Salwa Mohd Ramiah, Harikrishnan Kanesan, Jeevan Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware
description	The advancement of unmanned aerial vehicles (UAVs) has drawn researchers to update object detection algorithms for better accuracy and computation performance. Previous works applying deep learning models for object detection applications required high graphics processing unit (GPU) computation power. Generally, object detection models suffer trade-off between accuracy and model size where the relationship is not always linear in deep learning models. Various factors such as architectural design, optimization techniques, and dataset characteristics can significantly influence the accuracy, model size, and computation cost in adopting object detection models for low-cost embedded devices. Hence, it is crucial to employ lightweight object detection models for real-time object identification for the solution to be sustainable. In this work, an improved feature extraction network is proposed by incorporating an efficient long-range aggregation network for vehicle detection (ELAN-VD) in the backbone layer. The architecture improvement in YOLOv7-tiny model is proposed to improve the accuracy of detecting small vehicles in the aerial image. Besides that, the image size output of the second and third prediction boxes is upscaled for better performance. This study showed that the proposed method yields a mean average precision (mAP) of 57.94%, which is higher than that of the conventional YOLOv7-tiny. In addition, the proposed model showed significant performance when compared to previous works, making it viable for application in low-cost embedded devices.
format	Article
author	Andika, Johan Lela Khairuddin, Anis Salwa Mohd Ramiah, Harikrishnan Kanesan, Jeevan
author_facet	Andika, Johan Lela Khairuddin, Anis Salwa Mohd Ramiah, Harikrishnan Kanesan, Jeevan
author_sort	Andika, Johan Lela
title	Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware
title_short	Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware
title_full	Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware
title_fullStr	Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware
title_full_unstemmed	Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware
title_sort	improved feature extraction network in lightweight yolov7 model for real-time vehicle detection on low-cost hardware
publisher	Springer
publishDate	2024
url	http://eprints.um.edu.my/45256/ https://doi.org/10.1007/s11554-024-01457-1
_version_	1811682109742383104
score	13.211869

Improved feature extraction network in lightweight YOLOv7 model for real-time vehicle detection on low-cost hardware

Similar Items