A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature Optimization

Railroad construction sites are high-risk environments where monitoring personnel safety is critical for preventing accidents and enhancing construction efficiency. Traditional manual monitoring and image processing methods exhibit deficiencies in real-time performance and accuracy. This paper propo...

Full description

Saved in:
Bibliographic Details
Main Authors: Jianqiu Chen, Huan Xiong, Shixuan Zhou, Xiang Wang, Benxiao Lou, Longtang Ning, Qingwei Hu, Yang Tang, Guobin Gu
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/7/2061
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850188933106761728
author Jianqiu Chen
Huan Xiong
Shixuan Zhou
Xiang Wang
Benxiao Lou
Longtang Ning
Qingwei Hu
Yang Tang
Guobin Gu
author_facet Jianqiu Chen
Huan Xiong
Shixuan Zhou
Xiang Wang
Benxiao Lou
Longtang Ning
Qingwei Hu
Yang Tang
Guobin Gu
author_sort Jianqiu Chen
collection DOAJ
description Railroad construction sites are high-risk environments where monitoring personnel safety is critical for preventing accidents and enhancing construction efficiency. Traditional manual monitoring and image processing methods exhibit deficiencies in real-time performance and accuracy. This paper proposes a railway worker detection method based on improved support vector machines (ISVM), while using non-local mean noise reduction and histogram equalisation pre-processing techniques to optimise image quality to improve detection efficiency and accuracy. Multiscale features are then extracted with Inception v3 and combined with principal component analysis (PCA) for dimensionality reduction. Finally, an SVM classification algorithm is employed for personnel detection. To process small sample categories, data enhancement techniques (e.g., random flip and rotation) and K-fold cross-validation are applied to optimize the model parameters. The experimental results demonstrate that the ISVM method significantly improves accuracy and real-time performance compared to traditional detection methods and single deep learning models. This method provides technical support for railroad construction safety monitoring and effectively addresses personnel detection tasks in complex construction environments.
format Article
id doaj-art-bf58cb9856764bddbcf214416e5bffdb
institution OA Journals
issn 1424-8220
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-bf58cb9856764bddbcf214416e5bffdb2025-08-20T02:15:46ZengMDPI AGSensors1424-82202025-03-01257206110.3390/s25072061A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature OptimizationJianqiu Chen0Huan Xiong1Shixuan Zhou2Xiang Wang3Benxiao Lou4Longtang Ning5Qingwei Hu6Yang Tang7Guobin Gu8Guangxi Key Laboratory of International Join for China-ASEAN Comprehensive Transportation, Nanning University, Nanning 530200, ChinaGuangxi Key Laboratory of Intelligent Transportation System (ITS), Guilin University of Electronic Technology, Guilin 541004, ChinaGuangxi Key Laboratory of International Join for China-ASEAN Comprehensive Transportation, Nanning University, Nanning 530200, ChinaGuangxi Key Laboratory of Intelligent Transportation System (ITS), Guilin University of Electronic Technology, Guilin 541004, ChinaSchool of Traffic and Transportation Engineering, Central South University, Changsha 410075, ChinaGuangxi Key Laboratory of International Join for China-ASEAN Comprehensive Transportation, Nanning University, Nanning 530200, ChinaGuangxi Key Laboratory of Intelligent Transportation System (ITS), Guilin University of Electronic Technology, Guilin 541004, ChinaGuangxi Key Laboratory of International Join for China-ASEAN Comprehensive Transportation, Nanning University, Nanning 530200, ChinaGuangxi Key Laboratory of International Join for China-ASEAN Comprehensive Transportation, Nanning University, Nanning 530200, ChinaRailroad construction sites are high-risk environments where monitoring personnel safety is critical for preventing accidents and enhancing construction efficiency. Traditional manual monitoring and image processing methods exhibit deficiencies in real-time performance and accuracy. This paper proposes a railway worker detection method based on improved support vector machines (ISVM), while using non-local mean noise reduction and histogram equalisation pre-processing techniques to optimise image quality to improve detection efficiency and accuracy. Multiscale features are then extracted with Inception v3 and combined with principal component analysis (PCA) for dimensionality reduction. Finally, an SVM classification algorithm is employed for personnel detection. To process small sample categories, data enhancement techniques (e.g., random flip and rotation) and K-fold cross-validation are applied to optimize the model parameters. The experimental results demonstrate that the ISVM method significantly improves accuracy and real-time performance compared to traditional detection methods and single deep learning models. This method provides technical support for railroad construction safety monitoring and effectively addresses personnel detection tasks in complex construction environments.https://www.mdpi.com/1424-8220/25/7/2061image recognitionpersonnel detectionISVMdeep learningimage preprocessingPCA
spellingShingle Jianqiu Chen
Huan Xiong
Shixuan Zhou
Xiang Wang
Benxiao Lou
Longtang Ning
Qingwei Hu
Yang Tang
Guobin Gu
A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature Optimization
Sensors
image recognition
personnel detection
ISVM
deep learning
image preprocessing
PCA
title A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature Optimization
title_full A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature Optimization
title_fullStr A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature Optimization
title_full_unstemmed A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature Optimization
title_short A Hybrid Deep Learning and Improved SVM Framework for Real-Time Railroad Construction Personnel Detection with Multi-Scale Feature Optimization
title_sort hybrid deep learning and improved svm framework for real time railroad construction personnel detection with multi scale feature optimization
topic image recognition
personnel detection
ISVM
deep learning
image preprocessing
PCA
url https://www.mdpi.com/1424-8220/25/7/2061
work_keys_str_mv AT jianqiuchen ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT huanxiong ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT shixuanzhou ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT xiangwang ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT benxiaolou ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT longtangning ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT qingweihu ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT yangtang ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT guobingu ahybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT jianqiuchen hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT huanxiong hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT shixuanzhou hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT xiangwang hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT benxiaolou hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT longtangning hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT qingweihu hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT yangtang hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization
AT guobingu hybriddeeplearningandimprovedsvmframeworkforrealtimerailroadconstructionpersonneldetectionwithmultiscalefeatureoptimization