Open Access Open Access  Restricted Access Subscription Access

Real-Time Traffic Classification Using Weights & Biases for Model Monitoring

Raghu Ram Chowdary Velevela

Abstract


In today's transportation landscape, accurate sign prediction of traffic which is essential for enhancing safety on road and optimizing management of traffic. This paper presents an innovative solutions based on cloud which integrates YOLOv8, a cutting-edge object detection model to enhance and fine-tune traffic sign detection. The primary goal is to use YOLOv8 to design a cloud-based traffic sign recognition system while optimizing model development, training, and performance tracking by employing wandb's tools. YOLOv8 is perfect for quickly recognizing and locating traffic signs in intricate visual settings because of its real-time object detection feature. Cloud infrastructure is the foundation of the project.  Providing scalable computational resources vital for training sophisticated deep learning models like YOLOv8. Collaboration is another critical aspect of this project. Wandb’s collaborative tools enable seamless interaction between researchers and developers, regardless of location, fostering teamwork and continuous improvement. This paper aims to enhance road safety and improve traffic management by demonstrating the transformative power of cloud-based machine learning solutions.

Full Text:

PDF

References


Bhatt, N., Laldas, P., & Lobo, V. B. (2022). Real-time traffic sign detection and recognition system using CNNs and deep learning on hybrid datasets of German and Indian traffic signs. Journal of Computer Vision and Pattern Recognition, 14(3), 450-463.

Islam, M. T. (2019). Detection of globally recognized traffic signs using CNNs and image augmentation techniques. International Journal of Artificial Intelligence and Applications, 10(2), 97-107.

Kapoor, A., Nehra, N., & Deshwal, D. (2021). Comparative analysis of CNN architectures for traffic sign recognition. International Journal of Machine Learning and Data Science, 8(4), 234-245.

Macheev, E. M., Devyatkin, A. V., & Muzalevsky, A. R. (2019). Application of convolutional neural networks for traffic sign detection and recognition in UAV systems. Proceedings of the IEEE International Conference on Neural Networks and Signal Processing, 145-150.

Oza, R. M., Geisen, A., & Wang, T. (2021). Neural networks for traffic pattern categorization using image processing techniques. Journal of Advanced Neural Networks and AI Systems, 9(1), 112-120.

Sun, Y., Ge, P., & Liu, D. (2019). Traffic sign recognition system based on deep learning and image preprocessing. Journal of Intelligent Transportation Systems, 23(2), 125-136.

Zhu, Z., Liang, D., Zhang, S., Huang, X., Li, B., & Hu, S. (2016). Traffic-sign detection and classification in the wild. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2110-2118.

Cireşan, D. C., Meier, U., & Schmidhuber, J. (2012). Multi-column deep neural networks for traffic sign classification. Neural Networks, 32, 333-338.

Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 779-788.

Houben, S., Stallkamp, J., Salmen, J., Schlipsing, M., & Igel, C. (2013). Detection of traffic signs in real-world images: The German Traffic Sign Detection Benchmark. The IEEE International Joint Conference on Neural Networks (IJCNN), 1-8.


Refbacks

  • There are currently no refbacks.