Improved method for a pedestrian detection model based on YOLO

To address the dual challenges of excessive energy consumption and operational inefficiency inherent in the reliance of current agricultural machinery on direct supervision, this study developed an enhanced YOLOv8n-SS pedestrian detection algorithm through architectural modifications to the baseline...

Full description

Saved in:
Bibliographic Details
Main Authors: Yanfei LI, Chengyi DONG
Format: Article
Language:English
Published: Higher Education Press 2025-06-01
Series:Frontiers of Agricultural Science and Engineering
Subjects:
Online Access:https://journal.hep.com.cn/fase/EN/PDF/10.15302/J-FASE-2025613
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850149094122586112
author Yanfei LI
Chengyi DONG
author_facet Yanfei LI
Chengyi DONG
author_sort Yanfei LI
collection DOAJ
description To address the dual challenges of excessive energy consumption and operational inefficiency inherent in the reliance of current agricultural machinery on direct supervision, this study developed an enhanced YOLOv8n-SS pedestrian detection algorithm through architectural modifications to the baseline YOLOv8n framework. The proposed method had superior performance in dense agricultural contexts while improving detection capabilities for pedestrian distribution patterns under complex farmland conditions, including variable lighting and mechanical occlusions. The main innovations were: (1) integration of spatial pyramid dilated (SPD) operations with conventional convolution layers to construct SPD-Conv modules, which effectively mitigated feature information loss while enhancing small-target detection accuracy; (2) incorporation of selective kernel attention mechanisms to enable context-aware feature selection and adaptive feature extraction. Experimental validation revealed significant performance improvements over the original YOLOv8n model. This enhanced architecture achieved 7.2% and 9.2% increases in mAP0.5 and mAP0.5:0.95 metrics respectively for dense pedestrian detection, with corresponding improvements of 7.6% and 8.7% observed in actual farmland working environments. The proposed method ultimately provides a computationally efficient and robust intelligent monitoring solution for agricultural mechanization, facilitating the transition from conventional agricultural practices toward sustainable, low-carbon production paradigms through algorithmic optimization.
format Article
id doaj-art-ffe1989a58c44542b479efb2b1e3d160
institution OA Journals
issn 2095-7505
language English
publishDate 2025-06-01
publisher Higher Education Press
record_format Article
series Frontiers of Agricultural Science and Engineering
spelling doaj-art-ffe1989a58c44542b479efb2b1e3d1602025-08-20T02:27:01ZengHigher Education PressFrontiers of Agricultural Science and Engineering2095-75052025-06-0112224526010.15302/J-FASE-2025613Improved method for a pedestrian detection model based on YOLOYanfei LIChengyi DONGTo address the dual challenges of excessive energy consumption and operational inefficiency inherent in the reliance of current agricultural machinery on direct supervision, this study developed an enhanced YOLOv8n-SS pedestrian detection algorithm through architectural modifications to the baseline YOLOv8n framework. The proposed method had superior performance in dense agricultural contexts while improving detection capabilities for pedestrian distribution patterns under complex farmland conditions, including variable lighting and mechanical occlusions. The main innovations were: (1) integration of spatial pyramid dilated (SPD) operations with conventional convolution layers to construct SPD-Conv modules, which effectively mitigated feature information loss while enhancing small-target detection accuracy; (2) incorporation of selective kernel attention mechanisms to enable context-aware feature selection and adaptive feature extraction. Experimental validation revealed significant performance improvements over the original YOLOv8n model. This enhanced architecture achieved 7.2% and 9.2% increases in mAP0.5 and mAP0.5:0.95 metrics respectively for dense pedestrian detection, with corresponding improvements of 7.6% and 8.7% observed in actual farmland working environments. The proposed method ultimately provides a computationally efficient and robust intelligent monitoring solution for agricultural mechanization, facilitating the transition from conventional agricultural practices toward sustainable, low-carbon production paradigms through algorithmic optimization.https://journal.hep.com.cn/fase/EN/PDF/10.15302/J-FASE-2025613YOLOv8ndense pedestrian detectionSPD-ConvSK attention mechanismadaptive extraction
spellingShingle Yanfei LI
Chengyi DONG
Improved method for a pedestrian detection model based on YOLO
Frontiers of Agricultural Science and Engineering
YOLOv8n
dense pedestrian detection
SPD-Conv
SK attention mechanism
adaptive extraction
title Improved method for a pedestrian detection model based on YOLO
title_full Improved method for a pedestrian detection model based on YOLO
title_fullStr Improved method for a pedestrian detection model based on YOLO
title_full_unstemmed Improved method for a pedestrian detection model based on YOLO
title_short Improved method for a pedestrian detection model based on YOLO
title_sort improved method for a pedestrian detection model based on yolo
topic YOLOv8n
dense pedestrian detection
SPD-Conv
SK attention mechanism
adaptive extraction
url https://journal.hep.com.cn/fase/EN/PDF/10.15302/J-FASE-2025613
work_keys_str_mv AT yanfeili improvedmethodforapedestriandetectionmodelbasedonyolo
AT chengyidong improvedmethodforapedestriandetectionmodelbasedonyolo