Method and device for ascertaining a state of drowsiness of a driver

1. A method for ascertaining a state of drowsiness of a driver of a motor vehicle, comprising:

detecting, by an image sensor, images of an eye of the driver;

processing, by a processor, the images to determine, for each of the images, a respective distance of an upper eyelid of the eye from a lower eyelid of the eye as represented by the respective image;

for each of the determined distances, determining, by the processor what percentage the respective determined distance is of a predetermined normal degree of eye opening;

filtering out, by the processor, a first subset of the measurements for which the determined percentage is less than a first threshold and a second subset of the measurements for which the determined percentage is greater than a second threshold, the filtering resulting in a remaining third subset of the measurements;

determining, by the processor, a frequency with respect to time at which the measurements of the third subset occur;

determining, by the processor, that the identified frequency is greater than a predetermined threshold frequency corresponding to a trembling eyelid movement caused by drowsiness; and

responsive to the determination, outputting, by the processor, at least one of (a) a signal that warns of a critical state of drowsiness and (b) triggers an automatic vehicle driving maneuver.

2. The method as recited in claim 1 , wherein the filtering detects closing movements of the upper eyelid that are not complete closures of the eye.

3. The method as recited in claim 1 , wherein the first percentage is ≥60%.

4. The method as recited in claim 1 , wherein the first percentage is ≥70%.

5. The method as recited in claim 1 , wherein an identification of the driver is carried out and the threshold frequency is set as a function of the identification.

6. The method as recited in claim 1 , wherein the image sensor includes multiple camera sensors.

7. A device for ascertaining a state of drowsiness of a driver of a motor vehicle, comprising:

an image sensor configured to detect images of an eye of the driver; and

a safety device that includes a processor that is configured to:

process the images to determine, for each of the images, a respective distance of an upper eyelid of the eye from a lower eyelid of the eye as represented by the respective image;

for each of the determined distances, determine what percentage the respective determined distance is of a predetermined normal degree of eye opening;

filter out a first subset of the measurements for which the determined percentage is less than a first threshold and a second subset of the measurements for which the determined percentage is greater than a second threshold, the filtering resulting in a remaining third subset of the measurements;

determine a frequency with respect to time at which the measurements of the third subset occur;

determine that the identified frequency is greater than a predetermined threshold frequency corresponding to a trembling eyelid movement caused by drowsiness; and

responsive to the determination, output at least one of (a) a signal that warns of a critical state of drowsiness and (b) triggers an automatic vehicle driving maneuver.

8. The method as recited in claim 1 , wherein the normal degree of eye opening is an instantaneous normal degree of eye opening determined based on an identification of the driver.

9. The method as recited in claim 1 , wherein the signal that warns of the critical state of drowsiness is output.

10. The method as recited in claim 1 , wherein the signal that triggers the automatic vehicle driving maneuver is output.