Optical recognition methods for locating eyes

• US 6,419,638 B1
• Filed: 08/01/2000
• Issued: 07/16/2002
• Est. Priority Date: 07/20/1993
• Status: Expired due to Fees

• Current Assignee: Guice, David L., Sam H. Hay
• Original Assignee: Guice, David L., Sam H. Hay
• Inventors: Guice, David L., Hay, Sam H.
• Primary Examiner(s): HINDENBURG, MAX F
• Application Number:
  US09/630,178
• Time in Patent Office: 714 Days
• Field of Search: 600/558, 351/200, 351/221
• US Class Current: 600/558
• CPC Class Codes: A61B 3/0025 characterised by electronic signal processing, e.g. eye models
• CPC Class Codes: A61B 3/11 for measuring interpupillary distance or diameter of pupils
• CPC Class Codes: G06T 2207/30041 Eye; Retina; Ophthalmic

Abstract

A method for processing an image of eyes of a subject is provided. The image is divided into two halves, each containing an eye of the subject. The halves are then processed to determine at least a position of a pupil within the first half and second half. After positions of the pupils are determined, horizontal and vertical histograms are developed from the pupil in each of the halves. Several ways of plotting the respective histograms are disclosed, these plots provided with various indicators for detecting abnormal conditions of the eyes.

Claims

1. A method for processing an image including eyes of a subject registered by a reflex photometer comprising the steps of:

1) obtaining at least a partial image of a face of a subject containing eyes of said subject,

2) dividing said image into a first half and a second half so that said first half and said second half each contain an image of one of said eyes,

3) processing said first half and said second half to at least determine a position of a pupil of a one of said eves in said first half and said second half.

2. A method as set forth in claim 1 further comprising the step of excluding an area around an eye in each of said first half and said second half prior to initiating a search for said pupils of said eyes.

3. A method as set forth in claim 2 further comprising the step of sequentially applying a series of tests to each pixel in a searched area of each said first half and said second half to determine if the pixel under test has characteristics of a pupil of an eye.

4. A method as set forth in claim 3 wherein one of said tests further comprises the step of testing a said pixel under test to determine if it exceeds an intensity threshold of a lowest expected intensity value of a pixel registering a pupil.

5. A method as set forth in claim 3 wherein one of said tests further comprises the step of testing a said pixel under test to determine if it is a hot pixel.

6. A method as set forth in claim 3 wherein one of said tests further comprises the step of testing an area immediately around a said pixel under test to determine if said area is of an intensity value generally greater than an overall pixel intensity value in a respective said half said pixel under test is in.

7. A method as set forth in claim 3 wherein one of said tests further comprises the step of determining if circular structures exist in a region around the pixel under test.

8. A method as set forth in claim 7 further comprising the steps of:

establishing a reference intensity level that is generally lower than intensity levels of a pupil of an eye,

beginning at a pixel having an intensity value characteristic of a corneal spike, successively comparing pixel intensities in radially opposed directions from said pixel having an intensity value characteristic of a corneal spike with said reference intensity level until pixels having a lower intensity value than said reference value are detected,

in a plurality of lines generally perpendicular to said radially opposed directions, sequentially testing pixels until first transition intensities lower than said reference intensity level are located,

storing pixel locations registering said first transition intensities.

9. A method as set forth in claim 8 further comprising the step of sequentially testing pixels in directions normal to each said plurality of lines until second transition intensities lower than said reference intensity level are located,

storing pixel locations registering said second transition intensities.

10. A method as set forth in claim 9 further comprising the step of averaging said pixel locations of said first transition intensities and said second transition intensities to obtain an approximate location of a center of said pupil.

11. A method as set forth in claim 10 further comprising the steps of:

developing a 2 dimensional eye array for each eye of a subject, each said eye array including a pupil and at least a portion of an iris of a respective said eye, with each said pupil generally centered within an associated said eye array,

for each said 2 dimensional eye array, developing a first histogram array of intensity values developed from rows of pixels across at least said pupil in each said 2 dimensional eye array,

for each said eye array, developing a second histogram array of intensity values developed from columns of pixels across at least said pupil in each said 2 dimensional eye array,

locating said first histogram array horizontally aligned with respect to an associated said eye array,

locating said second histogram array vertically aligned with respect to an associated said eye array,

dividing pixel intensities into a range of equally separated ranges and assigning a particular color to each said range.

12. A method as set forth in claim 11 further comprising the steps of:

locating a geometric center of said pupil at a center of each said eye array,

providing at least one indica on two adjacent sides of each said eye array, said indica showing a geometric center of said pupil and said eye array,

comparing a position of a corneal spike in said first histogram array and said second histogram array to respective said indica to determine vertical and horizontal deviation of an eye imaged in each said eye array.

13. A method as set forth in claim 11 further comprising the step of developing said first histogram array and said second histogram array in the form of envelopes each showing an outline of said histogram arrays.

14. A method as set forth in claim 13 further comprising the step of indicating data including average intensity value, standard deviation, and a median value within each said envelope.

15. A method as set forth in claim 7 further comprising the step of determining pixel intensity values along radial lines extending from the pixel under test.

16. A method as set forth in claim 15 further comprising the step of detecting a drop of said pixel intensity values along said radial lines to locate an edge of said pupil.

17. A method as set forth in claim 16 further comprising the step of testing said pixel intensity values along said radial lines to locate pupil to iris transitions along each said radial line.

18. A method for displaying eyes of a subject imaged by a reflex photometer and comprising the steps of:

vertically dividing an image of said eyes into a first half and a second half,

developing a two-dimensional plot of said first half and said second half,

dividing light intensity values of pixels in said first half and said second half into a plurality of predetermined ranges,

assigning a different color for each of said predetermined ranges to develop a two-dimensional plot for each said first half and said second half.

19. A method as set forth in claim 18 further comprising the step of developing a histogram of intensity values of said first half and said second half.

20. A method as set forth in claim 18 further comprising the step of designating the range of intensity values including a highest number of pixels as a most probable intensity value.

21. A method as set forth in claim 18 further comprising the step of narrowing a span of said predetermined ranges of intensity values just above said most probable intensity values so that said predetermined ranges of intensity values just above said most probable intensity value is of a greater resolution.

22. A method as set forth in claim 18 further comprising the step of including a light intensity range between the most probable intensity value and approximately 10% of the highest intensity value.

23. A method as set forth in claim 18 wherein a three-dimensional plot of said first half and said second half is developed by vertically and horizontally offsetting light intensity values for each row of pixels.

24. A method as set forth in claim 23 further comprising the step of storing said two dimensional plot and said three dimensional plot for further analysis.

25. A method as set forth in claim 18 wherein a portion of said first half and said second half containing said eyes of the subject are transmitted to a different location for processing.

26. A method as set forth in claim 25 further including the steps of:

transmitting coordinates of the eyes in the full-face image,

transmitting interpupillary distance between said pupils of said eyes.

27. A method as set forth in claim 18 further comprising the steps of:

locating a predetermined number of groups of pixels registering higher levels of light intensity,

performing a series of tests to determine if the region around each group of pixels has characteristics of at least a pupil.

28. A method as set forth in claim 18 further comprising the steps of:

1) scanning each said array for pixels registering light intensity peaks,

2) applying a series of tests to determine whether any of said light

3) After a pupil is located, storing light intensity values of a small array just around the pupil.

29. A method as set forth in claim 28 further comprising the step of developing 2 dimensional and three dimensional images of said small array of pixels.

30. A method as set forth in claim 27 wherein one of said series of tests comprises the step of, for each of said predetermined number of groups of pixels registering higher levels of light intensity, testing pixel intensity values along a predetermined number of radial lines beginning from said group of pixels until light intensities representative of an edge of said pupil is found.