Lightweight Detection Methods for Multi-Scale Targets in Complex Scenarios

Detecting multi-scale objects in complex scenes is crucial for real-time applications in building safety, where accurate monitoring of safety equipment under challenging conditions is essential. In this paper, we propose LAP-YOLO (Lightweight Aggregate Perception based on “You Only Look O...

Full description

Saved in:
Bibliographic Details
Main Authors: Anjun Yu, Zhichao Rao, Yonghua Xiong, Jinhua She
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10965668/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850201364964048896
author Anjun Yu
Zhichao Rao
Yonghua Xiong
Jinhua She
author_facet Anjun Yu
Zhichao Rao
Yonghua Xiong
Jinhua She
author_sort Anjun Yu
collection DOAJ
description Detecting multi-scale objects in complex scenes is crucial for real-time applications in building safety, where accurate monitoring of safety equipment under challenging conditions is essential. In this paper, we propose LAP-YOLO (Lightweight Aggregate Perception based on “You Only Look Once”), which integrates a Feature-Aware Aggregation (FAA) module to enhance feature representation and a Lightweight Information Diffusion (LID) detection head to improve small-object detection efficiency with minimal computational overhead. Experimental results demonstrate that LAP-YOLO surpasses other real-time models on the safety equipment dataset, achieving a 39% reduction in parameters and a 57% decrease in FLOPs compared to YOLOv8, while maintaining detection accuracy under the same conditions. This lightweight and effective approach presents a promising solution for enhancing construction site safety.
format Article
id doaj-art-db3f43ebb0374fd18f3a3aad061294b6
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-db3f43ebb0374fd18f3a3aad061294b62025-08-20T02:12:02ZengIEEEIEEE Access2169-35362025-01-0113663936640410.1109/ACCESS.2025.356066410965668Lightweight Detection Methods for Multi-Scale Targets in Complex ScenariosAnjun Yu0Zhichao Rao1https://orcid.org/0009-0002-0418-6503Yonghua Xiong2https://orcid.org/0000-0002-8672-0193Jinhua She3https://orcid.org/0000-0003-3165-5045Jiangxi Ganyue Expressway Company Ltd., Nanchang, ChinaSchool of Automation, China University of Geosciences (Wuhan), Wuhan, ChinaSchool of Automation, China University of Geosciences (Wuhan), Wuhan, ChinaSchool of Automation, China University of Geosciences (Wuhan), Wuhan, ChinaDetecting multi-scale objects in complex scenes is crucial for real-time applications in building safety, where accurate monitoring of safety equipment under challenging conditions is essential. In this paper, we propose LAP-YOLO (Lightweight Aggregate Perception based on “You Only Look Once”), which integrates a Feature-Aware Aggregation (FAA) module to enhance feature representation and a Lightweight Information Diffusion (LID) detection head to improve small-object detection efficiency with minimal computational overhead. Experimental results demonstrate that LAP-YOLO surpasses other real-time models on the safety equipment dataset, achieving a 39% reduction in parameters and a 57% decrease in FLOPs compared to YOLOv8, while maintaining detection accuracy under the same conditions. This lightweight and effective approach presents a promising solution for enhancing construction site safety.https://ieeexplore.ieee.org/document/10965668/Complex backgroundlightweightmulti-scalesafety equipment
spellingShingle Anjun Yu
Zhichao Rao
Yonghua Xiong
Jinhua She
Lightweight Detection Methods for Multi-Scale Targets in Complex Scenarios
IEEE Access
Complex background
lightweight
multi-scale
safety equipment
title Lightweight Detection Methods for Multi-Scale Targets in Complex Scenarios
title_full Lightweight Detection Methods for Multi-Scale Targets in Complex Scenarios
title_fullStr Lightweight Detection Methods for Multi-Scale Targets in Complex Scenarios
title_full_unstemmed Lightweight Detection Methods for Multi-Scale Targets in Complex Scenarios
title_short Lightweight Detection Methods for Multi-Scale Targets in Complex Scenarios
title_sort lightweight detection methods for multi scale targets in complex scenarios
topic Complex background
lightweight
multi-scale
safety equipment
url https://ieeexplore.ieee.org/document/10965668/
work_keys_str_mv AT anjunyu lightweightdetectionmethodsformultiscaletargetsincomplexscenarios
AT zhichaorao lightweightdetectionmethodsformultiscaletargetsincomplexscenarios
AT yonghuaxiong lightweightdetectionmethodsformultiscaletargetsincomplexscenarios
AT jinhuashe lightweightdetectionmethodsformultiscaletargetsincomplexscenarios