What are the benefits?

Versatility

Testing has shown the APL system to have tremendous versatility in monitoring drowsiness and fatigue as well as other human physiological factors. The results of a preliminary study has demonstrated good correlation between the APL Drowsy Driver measurements and those taken using the Carnegie-Mellon University (CMU) implementation of PERCLOS. PERCLOS is defined as the measurement of the percentage of time the pupils of the eyes are 80% or more occluded over a specified time interval. The APL system showed an advantage over the CMU system because it predicted the onset of drowsiness shortly before the PERCLOS system.

   

The technique monitors and quantitatively measures several indices, such as the general activity level, the speed, frequency and duration of eyelid closure, the rate of heartbeat and respiration, by analyzing the Doppler components present in the reflected signal.

Validity

A multiple sensor, drowsy driver system was designed and constructed to demonstrate the validity of the eye blink system. The detection system consisted of two time-synced detection systems, the one developed at APL and a video-based system. The video data is collected at 15 frames per second, providing detailed resolution of eyelid motion at 67 ms per frame. This system allows the user to correlate the data sets by directly matching the critical time information from the APL Drowsy Driver system to that of the video with time resolution of hundredths of a second. The examples below demonstrate the ability of the APL drowsy driver detection system to detect eyelid behavior associated with drowsiness and the excellent correlation between the APL system and the video system results.

SEE: Effectiveness in Monitoring Sleep: Graphic Displays