A Lightweight Direction-Aware Network for Vehicle Detection

Vehicle detection algorithms, which are essential to intelligent traffic management and control systems, have attracted growing attention. However, most high-precision vehicle detection algorithms suffer from high computational effort and slow detection speeds, resulting in the challenging task of d...

Full description

Saved in:
Bibliographic Details
Main Authors: Luxia Yang, Yilin Hou, Hongrui Zhang, Chuanghui Zhang
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10877820/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850189942943121408
author Luxia Yang
Yilin Hou
Hongrui Zhang
Chuanghui Zhang
author_facet Luxia Yang
Yilin Hou
Hongrui Zhang
Chuanghui Zhang
author_sort Luxia Yang
collection DOAJ
description Vehicle detection algorithms, which are essential to intelligent traffic management and control systems, have attracted growing attention. However, most high-precision vehicle detection algorithms suffer from high computational effort and slow detection speeds, resulting in the challenging task of deploying these algorithms on mobile devices. In this paper, we propose a lightweight direction-aware network (LDAN) based on the YOLOv8 for vehicle detection on mobile devices. First, a lightweight C2f-GSP module is proposed to optimize the backbone network, which enhances the interaction of local features and fully extracts vehicle information. Then, a triple efficient coordinate attention mechanism (TECA) is designed. The mechanism can fully perceive the details and salient information of input features in multiple directions, thus improving the ability of the model to capture critical features. Moreover, to further reduce model parameters and computational requirements, a lightweight shared convolutional detection head (SCL-Head) is devised using a parameter-sharing mechanism. Finally, experimental results on the KITTI dataset show that the proposed method not only reduces resource consumption but also improves the accuracy of vehicle detection, which provides a novel technical path to realize real-time and accurate vehicle detection on mobile devices.
format Article
id doaj-art-0299524d27e44c37829e7c6ee08b206f
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-0299524d27e44c37829e7c6ee08b206f2025-08-20T02:15:25ZengIEEEIEEE Access2169-35362025-01-0113275162752610.1109/ACCESS.2025.353969210877820A Lightweight Direction-Aware Network for Vehicle DetectionLuxia Yang0https://orcid.org/0009-0009-5920-3340Yilin Hou1https://orcid.org/0009-0003-3071-1567Hongrui Zhang2https://orcid.org/0009-0007-7438-4819Chuanghui Zhang3https://orcid.org/0009-0004-4764-9576School of Computer Science and Technology, Taiyuan Normal University, Jinzhong, ChinaSchool of Computer Science and Technology, Taiyuan Normal University, Jinzhong, ChinaSchool of Computer Science and Technology, Taiyuan Normal University, Jinzhong, ChinaSchool of Computer Science and Technology, Taiyuan Normal University, Jinzhong, ChinaVehicle detection algorithms, which are essential to intelligent traffic management and control systems, have attracted growing attention. However, most high-precision vehicle detection algorithms suffer from high computational effort and slow detection speeds, resulting in the challenging task of deploying these algorithms on mobile devices. In this paper, we propose a lightweight direction-aware network (LDAN) based on the YOLOv8 for vehicle detection on mobile devices. First, a lightweight C2f-GSP module is proposed to optimize the backbone network, which enhances the interaction of local features and fully extracts vehicle information. Then, a triple efficient coordinate attention mechanism (TECA) is designed. The mechanism can fully perceive the details and salient information of input features in multiple directions, thus improving the ability of the model to capture critical features. Moreover, to further reduce model parameters and computational requirements, a lightweight shared convolutional detection head (SCL-Head) is devised using a parameter-sharing mechanism. Finally, experimental results on the KITTI dataset show that the proposed method not only reduces resource consumption but also improves the accuracy of vehicle detection, which provides a novel technical path to realize real-time and accurate vehicle detection on mobile devices.https://ieeexplore.ieee.org/document/10877820/Vehicle detectionYOLOv8lightweightattention mechanism
spellingShingle Luxia Yang
Yilin Hou
Hongrui Zhang
Chuanghui Zhang
A Lightweight Direction-Aware Network for Vehicle Detection
IEEE Access
Vehicle detection
YOLOv8
lightweight
attention mechanism
title A Lightweight Direction-Aware Network for Vehicle Detection
title_full A Lightweight Direction-Aware Network for Vehicle Detection
title_fullStr A Lightweight Direction-Aware Network for Vehicle Detection
title_full_unstemmed A Lightweight Direction-Aware Network for Vehicle Detection
title_short A Lightweight Direction-Aware Network for Vehicle Detection
title_sort lightweight direction aware network for vehicle detection
topic Vehicle detection
YOLOv8
lightweight
attention mechanism
url https://ieeexplore.ieee.org/document/10877820/
work_keys_str_mv AT luxiayang alightweightdirectionawarenetworkforvehicledetection
AT yilinhou alightweightdirectionawarenetworkforvehicledetection
AT hongruizhang alightweightdirectionawarenetworkforvehicledetection
AT chuanghuizhang alightweightdirectionawarenetworkforvehicledetection
AT luxiayang lightweightdirectionawarenetworkforvehicledetection
AT yilinhou lightweightdirectionawarenetworkforvehicledetection
AT hongruizhang lightweightdirectionawarenetworkforvehicledetection
AT chuanghuizhang lightweightdirectionawarenetworkforvehicledetection