Browsing by Author "Kafy, Md. Arafath"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Analyzing the Efficacy of Convolutional Neural Network Architectures for real-time Driver Behavior Detection: An In-depth Study of VGG16, ResNet50, InceptionV3, and Xception
    (Daffodil International University, 2024-01-12) Kafy, Md. Arafath; Afridi, Arafat Sahin
    Distracted driving is a significant contributor to traffic accidents worldwide, necessitating effective driver behavior monitoring systems. This study investigates the efficacy of state-of-the-art Convolutional Neural Network (CNN) architectures—VGG16, ResNet50, InceptionV3, Xception, and Xception_Customize_Model—for real-time driver behavior detection. The research employs a systematic methodology to evaluate these architectures on parameters such as accuracy, computational efficiency, and generalization ability. Through data preprocessing techniques like image augmentation and normalization, models were trained to classify driver behaviors such as safe driving, texting, talking on the phone, turning, and others. The experimental results highlight that Xception_Customize_Model achieved the highest validation accuracy of 97.07% with a consistently low validation loss, demonstrating superior performance and stability. The Xception model followed closely with competitive accuracy but exhibited occasional fluctuations in validation performance. ResNet50 also performed well, with a validation accuracy of 88.25%, reflecting robust classification capabilities despite requiring more epochs for stabilization. In contrast, InceptionV3, VGG16, and CNN demonstrated lower performance, with validation accuracies not exceeding 70%, largely due to higher training losses and limited generalization ability. The findings of this study contribute to the development of intelligent transportation systems by enhancing real-time detection of driver distractions, thereby promoting road safety. This research provides valuable insights into optimizing CNN architectures for real-world applications, offering a pathway for practitioners to implement scalable, efficient, and accurate driver monitoring solutions.
  • Thumbnail Image
    Item
    Traffic Congestion Prediction using Machine Learning
    (Scopus, 2024-04-18) Kafy, Md. Arafath; Faisal, Saimon Islam; Rahman, Md. Lutfor; Moni, Raka; Shanmuganathan, Harinee; Raza, Dewan Mamun
    Due to the world population growing rapidly over time, the number of personal and local vehicles are increasing which is one of the main causes of high traffic on the roads. For high traffic, the average speed of vehicles is decreasing which is known as traffic congestion. It is a very common and alarming problem in today’s world. Due to traffic congestion, civilians are facing different problems in this 21st century. Time is a precious thing and traffic congestion is killing the most precious times of our lives. In this paper, the authors aimed to offer a traffic congestion prediction model that will help to predict the traffic congestion of a particular area in a definite time period. During working with machine learning models or algorithms there is a concern about the accuracy of the result. To overcome this problem, 5 different machine learning models which are used decision tree, random forest, logistic regression, SVM, and MLP to predict the congestion rate. The authors compared those models with each other and calculated the mean absolute error for each of the models so that the prediction can be more accurate. Efforts are made to alleviate the traffic congestion reducing commute times and lower carbon emissions and to enhance the overall quality of life in cities.