https://matjournals.net/engineering/index.php/JIIS <p class="contentStyle">Journal of Instrumentation and Innovation Sciences is a print e-journal focused towards the rapid Publication of fundamental research papers on all areas of Instrumentation. This Journal involves the basic principles of art and science of measurement and control of process variables within a production or manufacturing area. Focus and Scope includes Design and Develop Control Systems, Maintain the Existing Control Systems, Industrial Instrumentation, Process Control, Sensors, Monitoring of Processes and Operations, Control of Processes and Operations, Experimental Engineering Analysis, Collaborate with Design Engineers, Quality Standards.</p> <h6 class="mt-2"> </h6> <div class="card"> </div> en-US Journal of Instrumentation and Innovation Sciences https://matjournals.net/engineering/index.php/JIIS/article/view/4048 <p><em>Driver drowsiness is a major contributor to road accidents, as fatigue can reduce attention, slow reaction time, and impair driving performance. This paper presents Alerion, a privacy-preserving driver safety monitoring system that uses Eye Aspect Ratio (EAR), computer vision, and edge computing for real-time drowsiness detection. The proposed system employs a camera connected to a Raspberry Pi to continuously capture the driver's facial images and determine eye status from facial landmarks. EAR values are calculated locally on the edge device, eliminating the need to transmit or store raw facial video on external servers. An adaptive thresholding mechanism is incorporated to account for variations in individual facial characteristics, camera positioning, and lighting conditions. When prolonged eye closure indicative of drowsiness is detected, the system generates progressive visual and audio warnings and can initiate an SOS alert during critical conditions. A web-based dashboard provides real-time visualization of eye status, EAR values, alert levels, and event information while receiving only processed data and alert metadata. The proposed prototype was evaluated through simulated testing using accuracy, response time, and false-alert rate as key performance measures. The system achieved 92% detection accuracy, a 1.2 s response time, and an 8% false-alert rate, demonstrating promising performance compared with the selected conventional approaches. The combination of edge processing, non-intrusive monitoring, real-time alerting, and privacy preservation makes Alerion a promising low-cost solution for intelligent driver safety applications. Future work will focus on real-road testing and integration of additional fatigue indicators such as yawning and head-pose analysis.</em></p> R. Sathiyakala K. P. Senthil K. Pravarakya M. Sri Mouktika Copyright (c) 2026 Journal of Instrumentation and Innovation Sciences 2026-08-27 2026-08-27 11 3 1 12