Driver Drowsiness Detection using ML and IOT

Abstract

According to new statistics across the globe, drowsy driving accidents become one of the very significant problems related to safety. This proposed Driver Sleep Detection System (DDS) takes a proactive approach with the help of ML and IoT technologies to track the behavior of the driver through features in facial expressions, head movements, and eye activity monitored by camera through the vehicle. Computer vision methods, such as Haar Cascades or CNNs, determine eye closure and yawning as drowsiness symptoms. The ML model enhances precision even more by training on diverse facial data while constantly adapting to varying driving environments and individual drivers. The key performance metrics are accuracy and response time, for the assessment of the system's performance. IoT integration allows for real-time alerts over an onboard buzzer or vibration system and can even forward data to a monitoring system or mobile device for remote viewing. Data is also stored in the cloud for long-term analysis.