Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images
With pedestrian crossings implicated in a significant proportion of vehicle-pedestrian accidents and the French government's initiatives to improve pedestrian safety, there is a pressing need for efficient, large-scale evaluation of pedestrian crossings. This study proposes the deployment of ad...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-02-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844025008084 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850194972156887040 |
|---|---|
| author | Dylan Russon Antoine Guennec Juan Naredo-Turrado Binbin Xu Cédric Boussuge Valérie Battaglia Benoit Hiron Emmanuel Lagarde |
| author_facet | Dylan Russon Antoine Guennec Juan Naredo-Turrado Binbin Xu Cédric Boussuge Valérie Battaglia Benoit Hiron Emmanuel Lagarde |
| author_sort | Dylan Russon |
| collection | DOAJ |
| description | With pedestrian crossings implicated in a significant proportion of vehicle-pedestrian accidents and the French government's initiatives to improve pedestrian safety, there is a pressing need for efficient, large-scale evaluation of pedestrian crossings. This study proposes the deployment of advanced deep learning neural networks to automate the assessment of pedestrian crossings and roundabouts, leveraging aerial and street-level imagery sourced from Google Maps and Google Street View. Utilizing ConvNextV2, ResNet50, and ResNext50 models, we conducted a comprehensive analysis of pedestrian crossings across various urban and rural settings in France, focusing on nine identified risk factors.Our methodology incorporates Mask R-CNN for precise segmentation and detection of zebra crossings and roundabouts, overcoming traditional data annotation challenges and extending coverage to underrepresented areas. The analysis reveals that the ConvNextV2 model, in particular, demonstrates superior performance across most tasks, despite challenges such as data imbalance and the complex nature of variables like visibility and parking proximity.The findings highlight the potential of convolutional neural networks in improving pedestrian safety by enabling scalable and objective evaluations of crossings. The study underscores the necessity for continued dataset augmentation and methodological advancements to tackle identified challenges. Our research contributes to the broader field of road safety by demonstrating the feasibility and effectiveness of automated, image-based pedestrian crossing audits, paving the way for more informed and effective safety interventions. |
| format | Article |
| id | doaj-art-dcfcc032a92b47b18a84a7835a7aff6d |
| institution | OA Journals |
| issn | 2405-8440 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj-art-dcfcc032a92b47b18a84a7835a7aff6d2025-08-20T02:13:52ZengElsevierHeliyon2405-84402025-02-01114e4242810.1016/j.heliyon.2025.e42428Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective imagesDylan Russon0Antoine Guennec1Juan Naredo-Turrado2Binbin Xu3Cédric Boussuge4Valérie Battaglia5Benoit Hiron6Emmanuel Lagarde7University of Bordeaux, INSERM BPH U1219, Bordeaux, F-33000, France; Corresponding author.University of Bordeaux, INSERM BPH U1219, Bordeaux, F-33000, FranceUniversity of Bordeaux, INSERM BPH U1219, Bordeaux, F-33000, FranceEuroMov Digital Health in Motion, Univ Montpellier, IMT Mines Ales, Ales, FranceCEREMA, FranceCEREMA, FranceCEREMA, FranceUniversity of Bordeaux, INSERM BPH U1219, Bordeaux, F-33000, FranceWith pedestrian crossings implicated in a significant proportion of vehicle-pedestrian accidents and the French government's initiatives to improve pedestrian safety, there is a pressing need for efficient, large-scale evaluation of pedestrian crossings. This study proposes the deployment of advanced deep learning neural networks to automate the assessment of pedestrian crossings and roundabouts, leveraging aerial and street-level imagery sourced from Google Maps and Google Street View. Utilizing ConvNextV2, ResNet50, and ResNext50 models, we conducted a comprehensive analysis of pedestrian crossings across various urban and rural settings in France, focusing on nine identified risk factors.Our methodology incorporates Mask R-CNN for precise segmentation and detection of zebra crossings and roundabouts, overcoming traditional data annotation challenges and extending coverage to underrepresented areas. The analysis reveals that the ConvNextV2 model, in particular, demonstrates superior performance across most tasks, despite challenges such as data imbalance and the complex nature of variables like visibility and parking proximity.The findings highlight the potential of convolutional neural networks in improving pedestrian safety by enabling scalable and objective evaluations of crossings. The study underscores the necessity for continued dataset augmentation and methodological advancements to tackle identified challenges. Our research contributes to the broader field of road safety by demonstrating the feasibility and effectiveness of automated, image-based pedestrian crossing audits, paving the way for more informed and effective safety interventions.http://www.sciencedirect.com/science/article/pii/S2405844025008084Pedestrian safetyPedestrian crossingsDeep learningConvolutional neural networksImage segmentationInfrastructure analysis |
| spellingShingle | Dylan Russon Antoine Guennec Juan Naredo-Turrado Binbin Xu Cédric Boussuge Valérie Battaglia Benoit Hiron Emmanuel Lagarde Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images Heliyon Pedestrian safety Pedestrian crossings Deep learning Convolutional neural networks Image segmentation Infrastructure analysis |
| title | Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images |
| title_full | Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images |
| title_fullStr | Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images |
| title_full_unstemmed | Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images |
| title_short | Evaluating pedestrian crossing safety: Implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images |
| title_sort | evaluating pedestrian crossing safety implementing and evaluating a convolutional neural network model trained on paired aerial and subjective perspective images |
| topic | Pedestrian safety Pedestrian crossings Deep learning Convolutional neural networks Image segmentation Infrastructure analysis |
| url | http://www.sciencedirect.com/science/article/pii/S2405844025008084 |
| work_keys_str_mv | AT dylanrusson evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages AT antoineguennec evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages AT juannaredoturrado evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages AT binbinxu evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages AT cedricboussuge evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages AT valeriebattaglia evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages AT benoithiron evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages AT emmanuellagarde evaluatingpedestriancrossingsafetyimplementingandevaluatingaconvolutionalneuralnetworkmodeltrainedonpairedaerialandsubjectiveperspectiveimages |