Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11
Developing reliable railway fault detection systems is crucial for ensuring both safety and operational efficiency. Various artificial intelligence frameworks, especially deep learning models, have shown significant potential in enhancing fault detection within railway infrastructure. This study exp...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Infrastructures |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2412-3811/10/1/3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588337969364992 |
|---|---|
| author | Omar Rodríguez-Abreo Mario A. Quiroz-Juárez Idalberto Macías-Socarras Juvenal Rodríguez-Reséndiz Juan M. Camacho-Pérez Gabriel Carcedo-Rodríguez Enrique Camacho-Pérez |
| author_facet | Omar Rodríguez-Abreo Mario A. Quiroz-Juárez Idalberto Macías-Socarras Juvenal Rodríguez-Reséndiz Juan M. Camacho-Pérez Gabriel Carcedo-Rodríguez Enrique Camacho-Pérez |
| author_sort | Omar Rodríguez-Abreo |
| collection | DOAJ |
| description | Developing reliable railway fault detection systems is crucial for ensuring both safety and operational efficiency. Various artificial intelligence frameworks, especially deep learning models, have shown significant potential in enhancing fault detection within railway infrastructure. This study explores the application of deep learning models for railway fault detection, focusing on both transfer learning architectures and a novel classification framework. Transfer learning was utilized with architectures such as ResNet50V2, Xception, VGG16, MobileNet, and InceptionV3, which were fine-tuned to classify railway track images into defective and non-defective categories. Additionally, the state-of-the-art YOLOv11 model was adapted for the same classification task, leveraging advanced data augmentation techniques to achieve high accuracy. Among the transfer learning models, VGG16 demonstrated the best performance with a test accuracy of 89.18%. However, YOLOv11 surpassed all models, achieving a test accuracy of 92.64% while maintaining significantly lower computational demands. These findings underscore the versatility of deep learning models and highlight the potential of YOLOv11 as an efficient and accurate solution for railway fault classification tasks. |
| format | Article |
| id | doaj-art-432a02c0f4224714a36b4fcf8920a41c |
| institution | Kabale University |
| issn | 2412-3811 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Infrastructures |
| spelling | doaj-art-432a02c0f4224714a36b4fcf8920a41c2025-01-24T13:35:22ZengMDPI AGInfrastructures2412-38112024-12-01101310.3390/infrastructures10010003Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11Omar Rodríguez-Abreo0Mario A. Quiroz-Juárez1Idalberto Macías-Socarras2Juvenal Rodríguez-Reséndiz3Juan M. Camacho-Pérez4Gabriel Carcedo-Rodríguez5Enrique Camacho-Pérez6Centro de Física Aplicada y Tecnología Avanzada, Universidad Nacional Autónoma de México, Santiago de Querétaro 76230, MexicoCentro de Física Aplicada y Tecnología Avanzada, Universidad Nacional Autónoma de México, Santiago de Querétaro 76230, MexicoFacultad de Ciencias Agrarias, Universidad Estatal Península de Santa Elena, Santa Elena (UPSE), La Libertad 240204, EcuadorFacultad de Ingeniería, Universidad Autónoma de Querétaro, Querétaro 76010, MexicoDepartamento de Sistemas y Computación, Tecnológico Nacional de México/I.T. Mérida, Mérida 97118, MexicoRed de Investigación OAC Optimización, Automatización y Control, El Marqués, Queretaro 76240, MexicoRed de Investigación OAC Optimización, Automatización y Control, El Marqués, Queretaro 76240, MexicoDeveloping reliable railway fault detection systems is crucial for ensuring both safety and operational efficiency. Various artificial intelligence frameworks, especially deep learning models, have shown significant potential in enhancing fault detection within railway infrastructure. This study explores the application of deep learning models for railway fault detection, focusing on both transfer learning architectures and a novel classification framework. Transfer learning was utilized with architectures such as ResNet50V2, Xception, VGG16, MobileNet, and InceptionV3, which were fine-tuned to classify railway track images into defective and non-defective categories. Additionally, the state-of-the-art YOLOv11 model was adapted for the same classification task, leveraging advanced data augmentation techniques to achieve high accuracy. Among the transfer learning models, VGG16 demonstrated the best performance with a test accuracy of 89.18%. However, YOLOv11 surpassed all models, achieving a test accuracy of 92.64% while maintaining significantly lower computational demands. These findings underscore the versatility of deep learning models and highlight the potential of YOLOv11 as an efficient and accurate solution for railway fault classification tasks.https://www.mdpi.com/2412-3811/10/1/3deep learningtransfer learningYOLOv11railway fault detectionsafetymaintenance |
| spellingShingle | Omar Rodríguez-Abreo Mario A. Quiroz-Juárez Idalberto Macías-Socarras Juvenal Rodríguez-Reséndiz Juan M. Camacho-Pérez Gabriel Carcedo-Rodríguez Enrique Camacho-Pérez Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11 Infrastructures deep learning transfer learning YOLOv11 railway fault detection safety maintenance |
| title | Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11 |
| title_full | Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11 |
| title_fullStr | Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11 |
| title_full_unstemmed | Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11 |
| title_short | Automatic Detection of Railway Faults Using Neural Networks: A Comparative Study of Transfer Learning Models and YOLOv11 |
| title_sort | automatic detection of railway faults using neural networks a comparative study of transfer learning models and yolov11 |
| topic | deep learning transfer learning YOLOv11 railway fault detection safety maintenance |
| url | https://www.mdpi.com/2412-3811/10/1/3 |
| work_keys_str_mv | AT omarrodriguezabreo automaticdetectionofrailwayfaultsusingneuralnetworksacomparativestudyoftransferlearningmodelsandyolov11 AT marioaquirozjuarez automaticdetectionofrailwayfaultsusingneuralnetworksacomparativestudyoftransferlearningmodelsandyolov11 AT idalbertomaciassocarras automaticdetectionofrailwayfaultsusingneuralnetworksacomparativestudyoftransferlearningmodelsandyolov11 AT juvenalrodriguezresendiz automaticdetectionofrailwayfaultsusingneuralnetworksacomparativestudyoftransferlearningmodelsandyolov11 AT juanmcamachoperez automaticdetectionofrailwayfaultsusingneuralnetworksacomparativestudyoftransferlearningmodelsandyolov11 AT gabrielcarcedorodriguez automaticdetectionofrailwayfaultsusingneuralnetworksacomparativestudyoftransferlearningmodelsandyolov11 AT enriquecamachoperez automaticdetectionofrailwayfaultsusingneuralnetworksacomparativestudyoftransferlearningmodelsandyolov11 |