Publication: Traffic sign detection and recognition for autonomous car driving assistance
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering::Electrical and electronic engineering | |
| dc.contributor.author | Rasydan bin Megat Ridzuan | |
| dc.date.accessioned | 2025-05-22T09:36:28Z | |
| dc.date.available | 2025-05-22T09:36:28Z | |
| dc.date.issued | 2024-08 | |
| dc.description.abstract | The rise of autonomous technology has ushered in radical changes to transportation recently and is paving the way for a future dominated by self-driving cars. Both in the automotive domain of self-driving vehicles and Advanced Driving Assistance Systems (ADAS), it is crucial to recognize the environment accurately. One important aspect of this environmental awareness is the detection and recognition of road signs, as they provide crucial information for vehicles and drivers alike. This technology helps minimize accidents caused by human error or missing traffic signs. This project presents a traffic sign detection and recognition system using YOLOv5, embedded on a Raspberry Pi 4 Model B. The device is mounted on vehicles to capture and analyze real-time road conditions. It identifies traffic signs and displays visual information to the driver, providing voice feedback to alert the driver. This work has three stages, which involve data collection and pre-processing regarding traffic signs, training the model on a comprehensive dataset, evaluating the model, embedding the model inside the Raspberry Pi 4 model B that has been setup, and deploying the trained YOLOv5 model to vehicles to evaluate performance based on precision, recall, F1 score, and mean average precision. Finally, the resultant performance of the model to detect and recognize traffic sign detection and recognition achieved 69% for precision, 67% for recall, 68% in F1-score, and 66% mean average precision at the 0.5 threshold. Overall, the results demonstrate the system’s potential to provide valuable assistance to human drivers and enhance road safety. | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/21797 | |
| dc.language.iso | en | |
| dc.title | Traffic sign detection and recognition for autonomous car driving assistance | |
| dc.type | Resource Types::text::report::technical report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |