Panasonic Develops Technology to Detect/Control Drowsiness of Car Driver
Panasonic Corp developed a technology to control drowsiness and prevent drowsy driving.
The technology enables to detect unconscious slight drowsiness, predict the transition of drowsiness and realize drowsiness control for keeping awakefullness. For detection and prediction of drowsiness, camera images and data collected by an infrared sensor are used.
Also, the new technology keeps awakefullness by controlling temperature and air volume (airflow) in accordance with the predicted level of drowsiness. Therefore, compared with conventional methods using sound and vibration, it can improve comfort for the user.
Panasonic plans to start to offer samples in October 2017. The technology is expected to be used not only for general and commercial vehicles but also for offices and educational institutions.
For the detection of drowsiness, images taken by cameras are used. For example, the contour of the eyes is extracted to check the time variation of blinks based on the aperture of the eyelids. As a result, it turns out that, when drowsiness increases, the upper eyelids lower and the amplitude and speed of the eyelid movements become smaller and faster, respectively.
Panasonic made a database of measurement results regarding drowsiness, blinking, facial expressions, etc and analyzed the correlation between about 1,800 parameters and drowsiness. Furthermore, based on the analysis of drowsy facial expressions, which was conducted in a joint research with The Ohara Memorial Institute for Science of Labour, Panasonic developed an AI that estimates a drowsiness level. It can contactlessly detect light drowsiness with a high accuracy.