System and method of detecting eye closure based on line angles
A system and method are provided for monitoring an eye to determine eye closure. The system includes a video imaging camera oriented to generate images of an eye, and a video processor for processing the images. The processor includes an edge detector for detecting edges of the eye in the image, and a line selector for generating lines at any of select angles representative of the detected edges. The processor determines the state of eye closure, such as an open eye or closed eye, based on an angle between two of the lines that select angles. The system may be employed on a vehicle to determine the eye closure state of a driver of the vehicle.
1. A system for monitoring an eye and determining eye closure, said system comprising:
a video imaging camera oriented to generate images of an eye; and
a video processor for processing the images generated with the video imaging camera, said video processor comprising an edge detector for detecting edges of the eye in the image and a line selector for selecting a line representative of the detected edges at any of multiple selected angles, said processor determining a state of eye closure based on an angle between two of the lines at two of the selected angles.
2. The system as defined in claim 1 , wherein each said line is substantially linear.
3. The system as defined in claim 1 , wherein said processor determines the state of eye closure is open when the angle between the two lines exceeds an open threshold angle.
4. The system as defined in claim 1 , wherein said processor determines the eye is closed when the angle between the two lines is less than a closed threshold angle.
5. The system as defined in claim 1 , wherein said processor further determines the eye is closed when only one or zero total lines are present among all of the selected angles.
6. The system as defined in claim 1 , wherein said processor determines the eye is open at a reduced confidence when the angle is greater than a reduced threshold angle that is less than the open threshold angle.
7. The system as defined in claim 1 , wherein said processor performs a Hough Transformation to transform each line from Cartesian coordinates to polar space and further determines the line at each of the multiple selected angles.
8. The system as defined in claim 1 , wherein said detected edges comprise a gray-scale representation of pixels imaging the eye.
9. The system as defined in claim 1 , wherein the system is employed in a vehicle for monitoring an eye of a driver of the vehicle.
10. A vehicle eye monitoring system for monitoring an eye of a person in a vehicle and determining eye closure, said system comprising:
a video imaging camera located on a vehicle and oriented to generate images of an eye of a person in the vehicle; and
a video processor for processing the images generated with the video imaging camera, said video processor comprising an edge detector for detecting edges of the eye in the image and a line selector for selecting a line representative of the detected edges at any of multiple angles, said processor determining a state of eye closure based on an angle between two of the lines at two of the selected angles.
11. A method of monitoring an eye and determining eye closure, said method comprising the steps of:
arranging a video imaging camera to generate images of an eye;
generating an image of the eye;
processing the image to identify edges of the eye in the image;
generating a line representative of an edge at any of multiple selected angles; and
determining an eye closure state based on an angle between two of the lines at two of the selected angles.
12. The method as defined in claim 11 , wherein the step of generating a line comprises generating a line that is substantially linear.
13. The method as defined in claim 11 , wherein the step of determining an eye closure state comprises determining the eye is open when the angle exceeds an open eye threshold.
14. The method as defined in claim 11 , further comprising the step of determining the eye is closed when the angle is less than a closed threshold angle.
15. The method as defined in claim 11 , further comprising the step of determining the eye is closed when only one or zero total lines are present among all of the selected angles.
16. The method as defined in claim 11 , further comprising the step of determining the eye is open at a reduced confidence (50%) when the angle is greater than a reduced threshold angle that is less than the open threshold angle.
17. The method as defined in claim 11 , wherein the step of generating a lines comprises transforming said line from Cartesian coordinates to polar space with a Hough Transformation and determining the line at each of any of the multiple selected angles.
18. The method as defined in claim 11 , wherein said step of arranging a video imaging camera comprises arranging the video imaging camera in a vehicle to monitor the eye of a driver of the vehicle.
19. The method as defined in claim 11 , further comprising the step of generating a gray-scale representation of pixels imaging the eye.