Study on lightweight strategies for L-YOLO algorithm in road object detection

Abstract With the increasing complexity of urban traffic, object detection has become critical in autonomous driving and intelligent traffic management. The demand for real-time, efficient object detection systems is growing. However, traditional algorithms often suffer from large parameter sizes an...

Full description

Saved in:
Bibliographic Details
Main Authors: Ji Hong, Kuntao Ye, Shubin Qiu
Format: Article
Language:English
Published: Nature Portfolio 2025-03-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-025-92148-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850029796973608960
author Ji Hong
Kuntao Ye
Shubin Qiu
author_facet Ji Hong
Kuntao Ye
Shubin Qiu
author_sort Ji Hong
collection DOAJ
description Abstract With the increasing complexity of urban traffic, object detection has become critical in autonomous driving and intelligent traffic management. The demand for real-time, efficient object detection systems is growing. However, traditional algorithms often suffer from large parameter sizes and high computational costs, limiting their applicability in resource-constrained environments. To address this issue, we propose L-YOLO, an improved lightweight road object detection algorithm based on YOLOv8s. First, L-HGNetV2 replaces the backbone network of YOLOv8s to enhance feature extraction and fusion efficiency. Second, a small object detection layer is introduced into the feature fusion network, replacing the original C2f modules with the new CStar modules. This modification improves the capture of features and contextual information for small vehicle targets without significantly increasing computational demands. Third, the CIoU loss function is replaced by the FPIoU2 loss function, enhancing the model’s robustness. Finally, the layer adaptive magnitude-based model pruning (LAMP) method is applied to prune the convolutional layer channels, significantly reducing the computational burden and parameter count while maintaining accuracy, thus improving operational efficiency. On the KITTI public dataset, L-YOLO achieves a mAP50 of 93.8%, a 2.5% improvement over YOLOv8s. The number of parameters decreases from 11.12 M to 3.58 M, and the computational load is reduced from 28.4 GFLOPs to 14.2 GFLOPs.
format Article
id doaj-art-c45f5d4a9eca47f1b28598f145b1ef9e
institution DOAJ
issn 2045-2322
language English
publishDate 2025-03-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-c45f5d4a9eca47f1b28598f145b1ef9e2025-08-20T02:59:24ZengNature PortfolioScientific Reports2045-23222025-03-0115111910.1038/s41598-025-92148-9Study on lightweight strategies for L-YOLO algorithm in road object detectionJi Hong0Kuntao Ye1Shubin Qiu2School of Science, Jiangxi University of Science and TechnologySchool of Science, Jiangxi University of Science and TechnologySchool of Science, Jiangxi University of Science and TechnologyAbstract With the increasing complexity of urban traffic, object detection has become critical in autonomous driving and intelligent traffic management. The demand for real-time, efficient object detection systems is growing. However, traditional algorithms often suffer from large parameter sizes and high computational costs, limiting their applicability in resource-constrained environments. To address this issue, we propose L-YOLO, an improved lightweight road object detection algorithm based on YOLOv8s. First, L-HGNetV2 replaces the backbone network of YOLOv8s to enhance feature extraction and fusion efficiency. Second, a small object detection layer is introduced into the feature fusion network, replacing the original C2f modules with the new CStar modules. This modification improves the capture of features and contextual information for small vehicle targets without significantly increasing computational demands. Third, the CIoU loss function is replaced by the FPIoU2 loss function, enhancing the model’s robustness. Finally, the layer adaptive magnitude-based model pruning (LAMP) method is applied to prune the convolutional layer channels, significantly reducing the computational burden and parameter count while maintaining accuracy, thus improving operational efficiency. On the KITTI public dataset, L-YOLO achieves a mAP50 of 93.8%, a 2.5% improvement over YOLOv8s. The number of parameters decreases from 11.12 M to 3.58 M, and the computational load is reduced from 28.4 GFLOPs to 14.2 GFLOPs.https://doi.org/10.1038/s41598-025-92148-9
spellingShingle Ji Hong
Kuntao Ye
Shubin Qiu
Study on lightweight strategies for L-YOLO algorithm in road object detection
Scientific Reports
title Study on lightweight strategies for L-YOLO algorithm in road object detection
title_full Study on lightweight strategies for L-YOLO algorithm in road object detection
title_fullStr Study on lightweight strategies for L-YOLO algorithm in road object detection
title_full_unstemmed Study on lightweight strategies for L-YOLO algorithm in road object detection
title_short Study on lightweight strategies for L-YOLO algorithm in road object detection
title_sort study on lightweight strategies for l yolo algorithm in road object detection
url https://doi.org/10.1038/s41598-025-92148-9
work_keys_str_mv AT jihong studyonlightweightstrategiesforlyoloalgorithminroadobjectdetection
AT kuntaoye studyonlightweightstrategiesforlyoloalgorithminroadobjectdetection
AT shubinqiu studyonlightweightstrategiesforlyoloalgorithminroadobjectdetection