Red eye detection aided by face detection

• US 8,929,614 B2
• Filed: 10/19/2009
• Issued: 01/06/2015
• Est. Priority Date: 10/19/2009
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Oicherman, Boris, Waidman, Ran, Gaubatz, Matthew, Harush, Shlomo
• Primary Examiner(s): Patel, Jayesh A
• Application Number:
  US13/387,065
• Publication Number:
  US 20120134546A1
• Time in Patent Office: 1,905 Days
• Field of Search: 382/103, 382115-118, 382/190, 382/199, 382/225, 382/282
• US Class Current: 382/117
• CPC Class Codes: G06T 2207/30216 Redeye defect
• CPC Class Codes: G06T 7/75 involving models

Abstract

A method for identifying a pair of genuine red eye artifacts in a captured image includes the steps of determining the presence of a face in the captured image, substantially encompassing the face within a shape, and determining the presence of three or more candidate red eye artifacts within the shape. The method continues with measuring the distance from an edge of the shape to each of the three or more candidate red eye artifacts and identifying, as genuine red eye artifacts, two candidate red eye artifacts of the three or more candidate red eye artifacts that are within a predetermined vertical distance from the edge of the shape.

Claims

1. A method for identifying a pair of genuine red eye artifacts in a captured image, comprising:

determining the presence of a face in the captured image;

substantially encompassing the face within a rectangular shape;

determining the presence of three or more candidate red eye artifacts within the rectangular shape;

identifying, as genuine red eye artifacts, two candidate red eye artifacts of the three or more candidate red eye artifacts by;

measuring a color distance between each of the three or more candidate red eye artifacts within the rectangular shape;

measuring, for each of the three or more candidate red eye artifacts, a vertical distance therefrom to a bottom edge of the rectangular shape and determining whether a group of candidate red eye artifacts of the three or more candidate red eye artifacts that are each within a predetermined vertical distance of the bottom edge of the rectangular shape exists;

after and in response to determining that the group exists, measuring a horizontal distance between each pair of candidate red eye artifacts of the group and determining whether a pair of candidate red eye artifacts of the group separated by a predetermined horizontal distance exists; and

in response to determining that the pair of candidate red eye artifacts of the group separated by the predetermined horizontal distance exists, setting the pair of candidate red eye artifacts as the genuine red eye artifacts.

2. The method of claim 1, wherein the bottom edge of the rectangular shape corresponds to a narrow side of the rectangle, wherein the narrow side of the rectangle substantially coincides with the chin of the face.

3. The method of claim 1, further comprising comparing an area of a first of the three or more candidate red eye artifacts with an area of the second of the three or more candidate red eye artifacts.

4. A logic module for identifying a pair of red eye artifacts in a captured image, comprising:

hardware, including a processor;

logic implemented at least by the hardware for determining that at least one face is present in the captured image;

logic implemented at least by the hardware for encompassing each one of the at least one face in a two-dimensional rectangular shape;

logic implemented at least by the hardware for determining that three or more candidate red eye artifacts are present in the two-dimensional shape;

logic implemented at least by the hardware for determining two genuine red eye artifacts within the rectangular two- dimensional shape by;

measuring a color distance between each of the three or more candidate red eye artifacts within the rectangular two-dimensional shape;

measuring, for each of the three or more candidate red eye artifacts, a vertical distance therefrom to a bottom edge of the rectangular two-dimensional shape and determining whether a group of candidate red eye artifacts of the three or more candidate red eye artifacts that are each within a predetermined vertical distance of the bottom edge of the rectangular two-dimensional shape exists;

after and in response to determining that the group exists, measuring a horizontal distance between each pair of candidate red eye artifacts of the group and determining whether a pair of candidate red eye artifacts of the group separated by a predetermined horizontal distance exists; and

in response to determining that the pair of candidate red eye artifacts of the group separated by the predetermined horizontal distance exists, setting the pair of candidate red eye artifacts as the genuine red eye artifacts.

5. The logic module of claim 4, further comprising logic for determining that a first of the two genuine red eye artifacts is of a color comparable to that of a second of the two genuine red eye artifacts.

6. The logic module of claim 4, further comprising logic for determining that a first of the two genuine red eye artifacts is of an area comparable to that of a second of the two genuine red eye artifacts.

7. The logic module of claim 4, wherein the bottom edge of the rectangular two-dimensional shape corresponds to a narrow side that substantially coincides with the chin of the at least one face present in the captured image.

8. The logic module of claim 4, wherein the logic for determining that two red eye artifacts are among the three or more candidate red eye artifacts present includes logic for assigning a probability to each of the candidate red eye artifacts, wherein the probability pertains to the likelihood that the candidate red eye artifact is a red eye artifact.