The Problem of Motorcycle Accidents in Indonesia

Oct 13, 2025 by recoveryshake.com 60 views

Introduction

In recent years, the number of accidents and injuries involving motorbikes has increased significantly. According to the Central Statistics Agency (BPS) of Indonesia, the majority of fatal accidents that occurred in motorcyclists were caused by inadequate helmets or even not using a helmet at all. To overcome this problem, a system has been developed that can detect violations of the use of helmets in real-time and automatically record vehicle plate numbers. This study uses the You Only Look Once method version 8 (YOLOV8) combined with Easyocr. Yolov8 is able to detect helmet objects and vehicle number plates on the motorcycle, while Easyocr functions to extract the vehicle plate number.

The Problem of Motorcycle Accidents in Indonesia

Motorcycle accidents are a significant problem in Indonesia, with the number of accidents and injuries increasing by 6.8% compared to 2022. The majority of fatal accidents that occurred in motorcyclists were caused by inadequate helmets or even not using a helmet at all. This is a serious concern, as helmets are a crucial safety feature for motorcyclists. The lack of helmet use is not only a safety issue but also a legal one, as wearing a helmet is mandatory in Indonesia.

The Yolov8 Algorithm: A Solution to the Problem

The Yolov8 algorithm is a deep learning-based object detection method that is capable of detecting objects in images and videos. In this study, the Yolov8 algorithm is used to detect helmet objects and vehicle number plates on the motorcycle. The algorithm is combined with Easyocr, which is a library for optical character recognition (OCR). Easyocr is used to extract the vehicle plate number from the detected image.

Testing the Yolov8 Algorithm Model

Testing the Yolov8 algorithm model shows very promising results. The algorithm records an accuracy rate of 96% in detecting the use of helmets among the drivers, with an average precision value of 96.5%, recall reaching 100%, and F1-Score by 98%. These numbers indicate that this system is very effective in recognizing whether motorcycle riders use helmets in accordance with established safety standards. In addition, in detecting vehicle plate numbers, this system reached 89.5% accuracy with an average precision of 98%, recall 91%, and F1-Score 94%. This evaluation metric indicates that the detection system can function properly in identifying the use of helmets and vehicle number plates.

Advantages of the Application of Yolov8 and Easyocr

The advantage of the application of Yolov8 and Easyocr in this study lies in its ability to work in real-time, thus providing practical solutions to the safety problem of motorcycle riders. With this technology, the authorities can quickly follow up on violations that occur on the road. In addition, the use of Android-based devices makes this system more accessible and implemented, thereby increasing the awareness of driving safety among the people.

Conclusion

This system not only increases security on the road but can also function as analytical data for the authorities to understand the patterns of violations that occur. Thus, more precise and efficient prevention steps can be taken to reduce the number of motorcycle accidents in Indonesia. This innovation is a concrete example of how technology can be used to solve urgent social problems, while providing added value for the community.

Future Work

Future work can focus on improving the accuracy of the Yolov8 algorithm in detecting vehicle plate numbers. Additionally, the system can be integrated with other technologies, such as GPS and camera systems, to provide more comprehensive data on motorcycle accidents.

Implementation

The implementation of the Yolov8 algorithm and Easyocr can be done using the following steps:

Data Collection: Collect a dataset of images and videos of motorcycles with and without helmets.
Model Training: Train the Yolov8 algorithm using the collected dataset.
Model Testing: Test the trained model on a separate dataset to evaluate its accuracy.
System Integration: Integrate the Yolov8 algorithm with Easyocr and Android-based devices.
Deployment: Deploy the system on the road to detect helmet use and vehicle plate numbers.

Conclusion

Q: What is the Yolov8 algorithm and how does it work?

A: The Yolov8 algorithm is a deep learning-based object detection method that is capable of detecting objects in images and videos. It works by using a neural network to analyze the input image or video and identify the objects within it. In this study, the Yolov8 algorithm is used to detect helmet objects and vehicle number plates on the motorcycle.

Q: What is Easyocr and how does it work?

A: Easyocr is a library for optical character recognition (OCR) that is used to extract the vehicle plate number from the detected image. It works by analyzing the image and identifying the characters within it, allowing it to extract the vehicle plate number.

Q: How accurate is the Yolov8 algorithm in detecting helmet use and vehicle plate numbers?

A: The Yolov8 algorithm has been tested and shown to have an accuracy rate of 96% in detecting the use of helmets among the drivers, with an average precision value of 96.5%, recall reaching 100%, and F1-Score by 98%. In addition, in detecting vehicle plate numbers, this system reached 89.5% accuracy with an average precision of 98%, recall 91%, and F1-Score 94%.

Q: How does the system work in real-time?

A: The system works in real-time by using the Yolov8 algorithm to analyze the input image or video and identify the objects within it. The Easyocr library is then used to extract the vehicle plate number from the detected image. This process is repeated continuously, allowing the system to detect helmet use and vehicle plate numbers in real-time.

Q: What are the advantages of using the Yolov8 algorithm and Easyocr in this study?

A: The advantages of using the Yolov8 algorithm and Easyocr in this study include its ability to work in real-time, providing practical solutions to the safety problem of motorcycle riders. Additionally, the use of Android-based devices makes this system more accessible and implemented, thereby increasing the awareness of driving safety among the people.

Q: How can the system be integrated with other technologies?

A: The system can be integrated with other technologies, such as GPS and camera systems, to provide more comprehensive data on motorcycle accidents. This can include integrating the system with existing traffic management systems to provide real-time data on traffic conditions and motorcycle accidents.

Q: What are the potential applications of this system?

A: The potential applications of this system include:

Road safety: The system can be used to detect helmet use and vehicle plate numbers in real-time, providing a practical solution to the safety problem of motorcycle riders.
Traffic management: The system can be integrated with existing traffic management systems to provide real-time data on traffic conditions and motorcycle accidents.
Law enforcement: The system can be used by law enforcement agencies to detect and prevent motorcycle accidents and to identify and prosecute drivers who are not wearing helmets.

Q: What are the limitations of this system?

A: The limitations of this system include:

Accuracy: While the Yolov8 algorithm has been shown to have high accuracy in detecting helmet use and vehicle plate numbers, there may be cases where the system fails to detect these objects accurately.
Real-time processing: The system may require significant processing power to analyze the input image or video in real-time, which may be a limitation in certain scenarios.
Integration with other technologies: The system may require significant integration with other technologies, such as GPS and camera systems, to provide more comprehensive data on motorcycle accidents.