Abstract
The growing rate of urbanization in the city of Dubai has contributed to the increase in vehicular activity and the prevalence of risky driving behavior on the road. This project proposes a deep-learning-based system, VGG-16, for the identification and classification of different patterns of driving behavior. The research methodology used for the study relies on the compilation of datasets from various resources, including ImageNet, Kaggle, UAH-DriverSet, D²-City Dataset, and confidential videos obtained from the Dubai Police command center. The findings of the study will be categorized into two parts. The first part will involve the compilation of the features based on patterns of behaviors of drivers captured from the datasets. The second part, on the other hand, relies on the feature classifiers for the categorization of different patterns of driving behavior into different classes. The implication of this research is that it provides relevant authorities with a robust system for traffic management and driver behavior regulations based on real-time data.
Publication Date
5-21-2024
Document Type
Thesis
Student Type
Graduate
Degree Name
Professional Studies (MS)
Department, Program, or Center
Graduate Programs & Research
Advisor
Ayman Ibrahim
Recommended Citation
Alhaddad, Fatma Abdulla Desmal, "Deep Learning Approach for Detecting and Classifying Chaotic Driving Activity Using Video Surveillance" (2024). Thesis. Rochester Institute of Technology. Accessed from
https://repository.rit.edu/theses/12816
Campus
RIT Dubai
