System for analyzing eye responses to automatically track size, location, and movement of the pupil

• US 7,798,643 B2
• Filed: 11/02/2007
• Issued: 09/21/2010
• Est. Priority Date: 11/02/2007
• Status: Active Grant

• Current Assignee: Visionetx Incorporated
• Original Assignee: Visionetx Incorporated
• Inventors: Waldorf, Ronald, Mauch, Barbara
• Primary Examiner(s): DANG, HUNG XUAN
• Application Number:
  US11/934,414
• Publication Number:
  US 20090115965A1
• Time in Patent Office: 1,054 Days
• Field of Search: 351/209, 351/210, 351/211, 351/212, 351/221, 351/246, 351/205, 351/247
• US Class Current: 351/209
• CPC Class Codes: A61B 3/113 for determining or recording eye movement
• CPC Class Codes: A61B 5/162 Testing reaction times
• CPC Class Codes: A61B 5/163 by tracking eye movement, gaze, or pupil change
• CPC Class Codes: G06V 20/597 Recognising the driver's state or behaviour, e.g. attention or drowsiness
• CPC Class Codes: G06V 40/19 Sensors therefor
• CPC Class Codes: G08B 21/06 indicating a condition of sleep, e.g. anti-dozing alarms

Abstract

The present system for analyzing eye responses accurately and automatically tracks the size, location, and movement of a subject'"'"'s pupil(s) in response to a predetermined test protocol. These eye responses include both ocularmotor and pupillary responses to illumination and target tracking for either eye or both eyes simultaneously. The series of digital image frames of a subject'"'"'s eyes, taken during the execution of a predetermined stimulus protocol, are automatically processed to determine pupil location, size, and responsiveness to changes in illumination.

Claims

1. A pupil tracking system for automatically determining pupil responses, as recorded in a time-wise sequence of digital images of at least one of the subject'"'"'s eyes, comprising:

digital image processing means for identifying a location of a pupil of at least one of said subject'"'"'s eyes in a plurality of said digital images in said sequence of digital images, comprising;

center point determining means for identifying a darkest spot in the pupil of said eye, pupil geometry determining means for ensuring said pupil satisfies a minimum shape configuration, comprising;

region-growing means for applying a region-growing procedure to said identified darkest spot of said pupil to expand the boundaries of said pupil to include regions of equivalent decreasing darkness; and

pupil size means for computing a size of said pupil in said plurality of said digital images in said sequence of digital images.

2. The pupil tracking system of claim 1 further comprising:

pupil movement determining means, responsive to data indicative of said location of said pupil, for tracking eye movement in said plurality of digital images.

3. The pupil tracking system of claim 1 further comprising:

filtering means for removing light specularities from said digital image.

4. The pupil tracking system of claim 1 wherein said pupil geometry determining means further comprises:

shape means for fitting an ellipse over the extracted pupil region to smooth the contours of the boundary.

5. The pupil tracking system of claim 4 wherein said pupil geometry determining means further comprises:

editing means for discarding regions with very small area or not being included within an elliptical shape from further analysis.

6. The pupil tracking system of claim 4 wherein said pupil geometry determining means further comprises:

geometric feature determining means for testing said pupil for at least one of its perimeter, roundness, and circularity.

7. The pupil tracking system of claim 6 wherein said geometric feature determining means comprises:

computation means for computing the perimeter of said pupil as the length of the boundary, and the roundness and circularity measures as;

$\mathrm{Roundness}=\frac{{<br><br>\mathrm{Perimerter}<br><br>}^2}{\mathrm{Area}}$ $\mathrm{and}$ $\mathrm{Circularity}=\frac{{\mathrm{\mu <br><br>}}_R}{{\mathrm{\sigma <br><br>}}_R}$

where μ

_o and σ

_o are the mean and standard deviation of the radial distance from the center of the pupil to the shape of the boundary.

8. The pupil tracking system of claim 1 further comprising:

blink determining means for rejecting a digital image when a distribution of intensities of the current digital image drastically changes from the distribution of intensities obtained from the previous digital image.

9. A method for automatically determining pupil responses, as recorded in a time-wise sequence of digital images of at least one of the subject'"'"'s eyes, comprising:

identifying a location of a pupil of at least one of said subject'"'"'s eyes in a plurality of said digital images in said sequence of digital images, comprising;

identifying a darkest spot in the pupil of said eye,

ensuring said pupil satisfies a minimum shape configuration, comprising;

applying a region-growing procedure to said identified darkest spot of said pupil to expand the boundaries of said pupil to include regions of equivalent decreasing darkness; and

computing a size of said pupil in said plurality of said digital images in said sequence of digital images.

10. The method for automatically determining pupil responses of claim 9 further comprising:

tracking, in response to data indicative of said location of said pupil, for tracking eye movement in said plurality of digital images.

11. The method for automatically determining pupil responses of claim 9 further comprising:

removing light specularities from said digital image.

12. The method for automatically determining pupil responses of claim 9 wherein said step of ensuring further comprises:

fitting an ellipse over the extracted pupil region to smooth the contours of the boundary.

13. The method for automatically determining pupil responses of claim 12 wherein said step of ensuring further comprises:

discarding regions with very small area or not being included within an elliptical shape from further analysis.

14. The method for automatically determining pupil responses of claim 12 wherein said step of ensuring further comprises:

testing said pupil for at least one of its perimeter, roundness, and circularity.

15. The method for automatically determining pupil responses of claim 14 wherein said step of testing comprises:

computing the perimeter of said pupil as the length of the boundary, and the roundness and circularity measures as;

$\mathrm{Roundness}=\frac{{<br><br>\mathrm{Perimerter}<br><br>}^2}{\mathrm{Area}}$ $\mathrm{and}$ $\mathrm{Circularity}=\frac{{\mathrm{\mu <br><br>}}_R}{{\mathrm{\sigma <br><br>}}_R}$

where μ

_o and σ

_o are the mean and standard deviation of the radial distance from the center of the pupil to the shape of the boundary.

16. The method for automatically determining pupil responses of claim 9 further comprising:

rejecting a digital image when a distribution of intensities of the current digital image drastically changes from the distribution of intensities obtained from the previous digital image.