Research

SafeTrax: Smart Collision Prediction and Alert System Using IoT for Sustainable Traffic Safety

With the alarming rise in traffic accidents globally, which lead to significant loss of life, injuries, and economic hardships, there is an urgent need for innovative solutions to enhance road safety. SafeTrax addresses this critical issue by leveraging deep learning and IoT technologies to predict potential traffic accidents and provide real-time alerts to drivers. By analyzing data from sensors and onboard cameras, SafeTrax proactively identifies potential collisions, offering timely warnings to prevent accidents before they occur and ensuring safer roads for everyone. In addition, the system is designed to seamlessly integrate with existing vehicle technologies, making it adaptable to a wide range of vehicles. The use of cloud-based architecture enables data synchronization and scalability, further enhancing the system's utility. Ultimately, SafeTrax aims to create a safer, smarter, and more sustainable transportation ecosystem.

Enhanced ROI Guided Deep Learning Model for Alzheimer’s Detection Using 3D MRI Images

Alzheimer’s disease is a progressive neurological disorder with no known cure, causing significant shrinkage in brain regions and disrupting neuronal connections. Current detection methods using 3D MRI images are resource-intensive, time-consuming, and computationally expensive, making them less practical for widespread use. Our research addresses these challenges by proposing an ROI-guided approach that selectively focuses on the six most affected brain regions, reducing computational overhead while improving detection accuracy. Using a 3D ResNet with a Convolutional Block Attention Module, our model achieves high accuracy on benchmark datasets (ADNI and OASIS), demonstrating that targeted analysis of Regions of Interest (ROIs) significantly enhances efficiency and diagnostic precision for Alzheimer’s disease detection.

Vehicle Number Plate Detection and Encryption in Digital Images Using YOLOv8 and Chaotic-Based Encryption Scheme

With the rapid growth of smart cities and intelligent transportation systems, ensuring the security and privacy of vehicular data has become a critical concern. Vehicle number plates contain sensitive information that, if accessed unauthorizedly, could lead to privacy violations, identity theft, or criminal misuse. To address these challenges, our research proposes a comprehensive system that combines YOLOv8, a state-of-the-art object detection model, with a chaotic-based logistic map encryption scheme. This system not only ensures accurate and real-time detection of number plates but also encrypts them to provide robust security. By integrating detection and encryption in a seamless pipeline, our approach enhances the confidentiality of sensitive information, safeguards vehicle identity, and resolves privacy concerns, making it highly suitable for real-time applications like traffic surveillance and law enforcement.