Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter Configurations
The automatic identification of traffic collisions is an emerging topic in modern traffic surveillance systems. The increasing number of surveillance cameras at urban intersections connected to traffic surveillance systems has created new opportunities for leveraging computer vision techniques for a...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/13/7001 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850118469546147840 |
|---|---|
| author | Muhammad Shahid Martin Gregurić Amirhossein Hassani Marko Ševrović |
| author_facet | Muhammad Shahid Martin Gregurić Amirhossein Hassani Marko Ševrović |
| author_sort | Muhammad Shahid |
| collection | DOAJ |
| description | The automatic identification of traffic collisions is an emerging topic in modern traffic surveillance systems. The increasing number of surveillance cameras at urban intersections connected to traffic surveillance systems has created new opportunities for leveraging computer vision techniques for automatic collision detection. This study investigates the effectiveness of transfer learning utilizing pre-trained deep learning models for collision detection through dashcam images. We evaluated several state-of-the-art (SOTA) image classification models and fine-tuned them using different hyperparameter combinations to test their performance on the car collision detection problem. Our methodology systematically investigates the influence of optimizers, loss functions, schedulers, and learning rates on model generalization. A comprehensive analysis is conducted using 7 performance metrics to assess classification performance. Experiments on a large dashcam-based images dataset show that ResNet50, optimized with AdamW, a learning rate of 0.0001, CosineAnnealingLR scheduler, and Focal Loss, emerged as the top performer, achieving an accuracy of 0.9782, F1-score of 0.9617, and IoU of 0.9262, indicating a strong ability to reduce false negatives. |
| format | Article |
| id | doaj-art-38ba1e97449b47b1bbac98b5c7d0eadd |
| institution | OA Journals |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-38ba1e97449b47b1bbac98b5c7d0eadd2025-08-20T02:35:51ZengMDPI AGApplied Sciences2076-34172025-06-011513700110.3390/app15137001Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter ConfigurationsMuhammad Shahid0Martin Gregurić1Amirhossein Hassani2Marko Ševrović3Faculty of Transport and Traffic Sciences, University of Zagreb, Vukelićeva Street 4, HR-10000 Zagreb, CroatiaFaculty of Transport and Traffic Sciences, University of Zagreb, Vukelićeva Street 4, HR-10000 Zagreb, CroatiaFaculty of Transport and Traffic Sciences, University of Zagreb, Vukelićeva Street 4, HR-10000 Zagreb, CroatiaFaculty of Transport and Traffic Sciences, University of Zagreb, Vukelićeva Street 4, HR-10000 Zagreb, CroatiaThe automatic identification of traffic collisions is an emerging topic in modern traffic surveillance systems. The increasing number of surveillance cameras at urban intersections connected to traffic surveillance systems has created new opportunities for leveraging computer vision techniques for automatic collision detection. This study investigates the effectiveness of transfer learning utilizing pre-trained deep learning models for collision detection through dashcam images. We evaluated several state-of-the-art (SOTA) image classification models and fine-tuned them using different hyperparameter combinations to test their performance on the car collision detection problem. Our methodology systematically investigates the influence of optimizers, loss functions, schedulers, and learning rates on model generalization. A comprehensive analysis is conducted using 7 performance metrics to assess classification performance. Experiments on a large dashcam-based images dataset show that ResNet50, optimized with AdamW, a learning rate of 0.0001, CosineAnnealingLR scheduler, and Focal Loss, emerged as the top performer, achieving an accuracy of 0.9782, F1-score of 0.9617, and IoU of 0.9262, indicating a strong ability to reduce false negatives.https://www.mdpi.com/2076-3417/15/13/7001car collision detectioncar accidents detectiondeep learningroad safety |
| spellingShingle | Muhammad Shahid Martin Gregurić Amirhossein Hassani Marko Ševrović Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter Configurations Applied Sciences car collision detection car accidents detection deep learning road safety |
| title | Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter Configurations |
| title_full | Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter Configurations |
| title_fullStr | Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter Configurations |
| title_full_unstemmed | Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter Configurations |
| title_short | Optimizing Car Collision Detection Using Large Dashcam-Based Datasets: A Comparative Study of Pre-Trained Models and Hyperparameter Configurations |
| title_sort | optimizing car collision detection using large dashcam based datasets a comparative study of pre trained models and hyperparameter configurations |
| topic | car collision detection car accidents detection deep learning road safety |
| url | https://www.mdpi.com/2076-3417/15/13/7001 |
| work_keys_str_mv | AT muhammadshahid optimizingcarcollisiondetectionusinglargedashcambaseddatasetsacomparativestudyofpretrainedmodelsandhyperparameterconfigurations AT martingreguric optimizingcarcollisiondetectionusinglargedashcambaseddatasetsacomparativestudyofpretrainedmodelsandhyperparameterconfigurations AT amirhosseinhassani optimizingcarcollisiondetectionusinglargedashcambaseddatasetsacomparativestudyofpretrainedmodelsandhyperparameterconfigurations AT markosevrovic optimizingcarcollisiondetectionusinglargedashcambaseddatasetsacomparativestudyofpretrainedmodelsandhyperparameterconfigurations |