Method for detecting face direction of a person
A method for detecting face direction of a person includes receiving a face image of the person. The method further includes determining whether the person is wearing glasses, based on the face image. The method also includes determining whether the number of reflection points of light in a glasses region of the face image is four or more at the time of detecting the glasses region. The method also includes aligning the reflection points of light in order of size, upon determining that the number of reflection points of light is four or more. The method also includes detecting two virtual images of the light, based on the aligning. The method also includes detecting a face direction vector based on the two virtual images of the light.
1. A method for detecting face direction, the method comprising:
receiving a face image of a person;
determining whether the person is wearing glasses, based on the face image;
determining whether a number of reflection points of light in a glasses region of the face image is four or more at a time of detecting the glasses region;
aligning reflection points of light in order of size, upon determining that the number of reflection points of light is four or more;
detecting two virtual images of the light, based on the aligning; and
detecting a face direction vector based on the two virtual images of the light.
2. The method according to claim 1 , wherein determining whether the person is wearing glasses comprises detecting a glasses region in the face image, and
when the glasses region is not detected, receiving a second face image of the person.
3. The method according to claim 1 , further comprising:
upon determining that the number of reflection points of light in the glasses is less than four, detecting a line of sight of the person.
4. The method according to claim 3 , further comprising:
determining a failure of the line of sight detection; and
applying additional light so as to increase the reflection points of the light.
5. The method according to claim 4 , further comprising:
upon applying the additional light, receiving a third face image of the person.
6. The method according to claim 3 , further comprising:
determining a success of the line of sight detection; and
determining whether the number of reflection points of light in the glasses region of the face image is four or more at the time of detecting the glasses region.
7. The method according to claim 1 , wherein the reflection points are reflection points of light by an inner surface of a lens of the glasses and reflection points of light by an outer surface thereof.
8. The method according to claim 1 , wherein the face image is received through a driver state surveillance system.