Driver drowsiness is one of the major factors contributing to road accidents, as fatigue and lack of sleep can reduce a driver's attention, reaction time, and ability to control a vehicle safely. NidraNet: AI-Based Driver Drowsiness Detection System is a proposed intelligent system designed to monitor driver alertness using Artificial Intelligence (AI), Deep Learning, and Computer Vision techniques. The system uses a camera or webcam to capture the driver's facial images and analyzes visual indicators such as eye closure, blinking patterns, yawning, and facial movements to identify possible signs of drowsiness. Image processing techniques using OpenCV help process the captured frames, while a deep-learning model, such as a Convolutional Neural Network (CNN) or EfficientNet, can be used to classify the driver's state as alert or drowsy. When drowsiness is detected, the system can generate an audible warning to alert the driver and encourage timely action. A web-based interface allows users to interact with the system, while SQLite can be used to store relevant user information and detection records when database functionality is implemented. The main objective of NidraNet is to support early detection of driver fatigue and improve road safety through timely warnings. The system may be further enhanced with infrared cameras, mobile notifications, and vehicle-integrated safety features. NidraNet demonstrates the potential of AI-based driver monitoring as a supportive technology for reducing fatigue-related driving risks.

The primary objective of NidraNet is to support the early identification of visual signs associated with driver drowsiness and improve awareness of fatigue-related driving risks. The proposed approach offers the advantages of automated monitoring, real-time processing potential, and compatibility with commonly available computing devices and cameras. However, the reliability of the system depends on factors such as camera placement, lighting conditions, facial visibility, dataset quality, model performance, and processing speed.

Keywords—NidraNet, Driver Drowsiness Detection, Artificial Intelligence, Deep Learning, Computer Vision, OpenCV, CNN, EfficientNet, Driver Monitoring, Road Safety.