Eyelid detection device, drowsiness determination device, and eyelid detection method

1. An eyelid detection device comprising:

processing circuitry

to detect outer and inner eye corners as face feature points from a captured image;

to detect an eye region including both an upper eyelid area and an eye area from the captured image on a basis of the face feature points; and

to detect, for each of a plurality of divided areas into which the eye region is divided, a boundary between the upper eyelid area and the eye area on a basis of a pattern of arrangement of intensity values of pixels aligned in a divided area, and to connect the boundaries each detected in the divided area to determine an upper eyelid,

wherein the processing circuitry is configured to use, in one of the divided area, a first threshold and a second threshold greater than the first threshold to determine as belonging to the upper eyelid area an arrangement pattern in which pixels each having an intensity value equal to or greater than the first threshold and less than the second threshold are continuously arranged, and determine as belonging to the eye area an arrangement pattern in which a pixel having an intensity value less than the first threshold and a pixel having an intensity value equal to or greater than the second threshold are continuously arranged, in which pixels each having an intensity value less than the first threshold are continuously arranged, or in which pixels each having an intensity value equal to or greater than the second threshold are continuously arranged.

2. An eyelid detection device comprising:

processing circuitry

to detect outer and inner eye corners as face feature points from a captured image;

to detect an eye region including both an upper eyelid area and an eye area from the captured image on a basis of the face feature points;

to detect, for each of a plurality of divided areas into which the eye region is divided, a boundary between the upper eyelid area and the eye area on a basis of a pattern of arrangement of intensity values of pixels aligned in a divided area, and to connect the boundaries each detected in the divided area to determine an upper eyelid; and

to detect a temporary upper eyelid and a lower eyelid from the eye region on a basis of the face feature points, wherein

the processing circuitry is configured to determine that, if the detected temporary upper eyelid and the detected lower eyelid do not match, the detected upper eyelid is a true upper eyelid.

3. The eyelid detection device according to claim 2 , the processing circuitry is further configured to calculate an degree of eye openness by using both the detected true upper eyelid and the detected lower eyelid detected.

4. The eyelid detection device according to claim 1 , wherein each divided area is an area connecting an end of the eye region to an opposite end of the eye region.

5. The eyelid detection device according to claim 2 , wherein each divided area is an area connecting an end of the eye region to an opposite end of the eye region.

6. The eyelid detection device according to claim 3 , wherein each divided area is an area connecting an end of the eye region to an opposite end of the eye region.

7. The eyelid detection device according to claim 1 , the processing circuitry is further configured to detect, if a plurality of upper eyelids is detected, a degree of change in intensity value for each of the plurality of upper eyelids, wherein

to determine that an upper eyelid having a largest degree of change in intensity value detected by the processing circuitry is a true upper eyelid among the plurality of upper eyelids.

8. The eyelid detection device according to claim 2 , the processing circuitry is further configured to detect, if a plurality of upper eyelids is detected, a degree of change in intensity value for each of the plurality of upper eyelids, wherein

to determine that an upper eyelid having a largest degree of change in intensity value detected by the processing circuitry is a true upper eyelid among the plurality of upper eyelids.

9. The eyelid detection device according to claim 3 , the processing circuitry is further configured to detect, if a plurality of upper eyelids is detected, a degree of change in intensity value for each of the plurality of upper eyelids, wherein

to determine that an upper eyelid having a largest degree of change in intensity value detected by the processing circuitry is a true upper eyelid among the plurality of upper eyelids.

10. The eyelid detection device according to claim 4 , the processing circuitry is further configured to detect, if a plurality of upper eyelids is detected, a degree of change in intensity value for each of the plurality of upper eyelids, wherein

to determine that an upper eyelid having a largest degree of change in intensity value detected by the processing circuitry is a true upper eyelid among the plurality of upper eyelids.

11. The eyelid detection device according to claim 5 , the processing circuitry is further configured to detect, if a plurality of upper eyelids is detected, a degree of change in intensity value for each of the plurality of upper eyelids, wherein

to determine that an upper eyelid having a largest degree of change in intensity value detected by the processing circuitry is a true upper eyelid among the plurality of upper eyelids.

12. The eyelid detection device according to claim 6 , the processing circuitry is further configured to detect, if a plurality of upper eyelids is detected, a degree of change in intensity value for each of the plurality of upper eyelids, wherein

to determine that an upper eyelid having a largest degree of change in intensity value detected by the processing circuitry is a true upper eyelid among the plurality of upper eyelids.

13. A drowsiness determination device comprising:

the eyelid detection device according to claim 3 ; and

the processing circuitry is further configured to determine a drowsy state on a basis of the degree of eye openness calculated by the eyelid detection device.

14. An eyelid detection method comprising:

detecting outer and inner eye corners as face feature points from a captured image;

detecting an eye region including both an upper eyelid area and an eye area from the captured image on a basis of the face feature points; and

detecting for each of a plurality of divided areas into which the eye region is divided, a boundary between the upper eyelid area and the eye area on a basis of a pattern of arrangement of intensity values of pixels aligned in a divided area, and connecting the boundaries each detected in the divided area to determine an upper eyelid,

using, in one of the divided area, a first threshold and a second threshold greater than the first threshold to determine as belonging to the upper eyelid area an arrangement pattern in which pixels each having an intensity value equal to or greater than the first threshold and less than the second threshold are continuously arranged, and determine as belonging to the eye area an arrangement pattern in which a pixel having an intensity value less than the first threshold and a pixel having an intensity value equal to or greater than the second threshold are continuously arranged, in which pixels each having an intensity value less than the first threshold are continuously arranged, or in which pixels each having an intensity value equal to or greater than the second threshold are continuously arranged.