This paper proposes a system to detect and measure blink rate to determine fatigue levels. The method involved analysing specific frames to determine that a blink occurred, and then monitoring the time between successive blinks. The program was simulated in python using a Raspberry Pi Zero and a standard USB camera. For the blink rate detection block, a gate level schematic was implemented in Cadence software using 65nm CMOS technology. The design was based around an asynchronous 6-bit based edge counter which was designed using D-flip-flops. The simulation calculated the average blink rate and compared this to the most recent blink rate. The outcome would determine if an alarm signal should be sent to the alarm. The system consumed 130uA from a 1.2V power supply.
Yassine, Nabil Barker, Stephen Hayatleh, KhaledChoubey, BhaskarNagulapalli, Rajasekhar
Faculty of Technology, Design and Environment\School of Engineering, Computing and Mathematics
Year of publication: 2018Date of RADAR deposit: 2018-03-02