Detecting redeye defects in digital images

• US 8,000,526 B2
• Filed: 06/27/2010
• Issued: 08/16/2011
• Est. Priority Date: 11/08/2007
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: Tessera Technologies Ireland Limited (Adeia Inc.)
• Inventors: Ciuc, Mihai
• Primary Examiner(s): Ge; Yuzhen
• Application Number:
  US12/824,214
• Publication Number:
  US 20100260414A1
• Time in Patent Office: 415 Days
• Field of Search: None
• US Class Current: 382/167
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction

Abstract

A method for detecting a redeye defect in a digital image containing an eye comprises converting the digital image into an intensity image, and segmenting the intensity image into segments each having a local intensity maximum. Separately, the original digital image is thresholded to identify regions of relatively high intensity and a size falling within a predetermined range. Of these, a region is selected having substantially the highest average intensity, and those segments from the segmentation of the intensity image whose maxima are located in the selected region are identified.

Claims

1. A method for correcting a defect in a digital image containing an eye, the method comprising:

converting the digital image into an intensity image,

identifying a group of pixels containing an eye region within the intensity image;

segmenting the group of pixels containing the eye region of the intensity image into segments each having a local intensity extremum,

thresholding a corresponding portion of the digital image to identify at least two regions of relatively extreme intensity, including a pupil and one or both of an iris or an eyelid;

selecting a region from the at least two regions identified in the thresholding step having the most extreme average intensity, and

identifying segments from the segmenting step intersecting the region selected at selecting step according to a pre-determined criterion, and

wherein the thresholding step uses a threshold obtained by identifying the most extreme intensity pixel inside the intensity image, computing the average intensity of the pixels inside a smaller rectangle centered on the most extreme intensity pixel, and assigning the threshold to the sum of the average intensity and a fixed quantity.

2. The method of claim 1 further comprising identifying regions from the thresholding step having a size falling within a predetermined range, and wherein said region of the most extreme average intensity is selected from said at least two regions.

3. The method of claim 1 wherein the identifying step comprises identifying those segments from the segmenting step whose extrema are located in the region selected at the selecting step.

4. The method of claim 1, comprising subjecting the intensity image to local averaging prior to mean shift segmentation.

5. The method of claim 1, wherein the converting step is performed by mean shift segmentation of the intensity image of the digital image.

6. The method of claim 1, further including:

computing an intensity histogram of a portion of the intensity image corresponding to identified segments,

smoothing said histogram,

determining a threshold from the smoothed histogram, and

thresholding the intensity histogram with the threshold from the determining step.

7. The method of claim 1 wherein the converting and segmenting are applied only to the eye region.

8. The method of claim 1 wherein the at least two regions comprise a pupil and an iris.

13. A digital image acquisition and processing apparatus including a lens, an image sensor, a processor, and a memory having code embedded therein for programming the processor the performs the method as claimed in any of claims 1-8.

14. One or more non-transitory computer readable media having code embedded therein for programming a processor the performs the method as claimed in any of claims 1-8.

9. A method for correcting a defect in a digital image containing an eye, the method comprising:

converting the digital image into an intensity image,

identifying a group of pixels containing an eye region within the intensity image;

segmenting the group of pixels containing the eye region of the intensity image into segments each having a local intensity extremum,

thresholding a corresponding portion of the digital image to identify at least two regions of relatively extreme intensity, including a pupil and one or both of an iris or an eyelid;

selecting a region from the at least two regions identified in the thresholding step having the most extreme average intensity, and

identifying segments from the segmenting step intersecting the region selected at selecting step according to a pre-determined criterion, and

wherein the average intensities of the at least two regions from the thresholding step having the most extreme and next most extreme average intensities differ by more than a predetermined amount, and the region selected at the selecting step comprises a region having the most extreme average intensity.

10. The method of claim 9, comprising subjecting the intensity image to local averaging prior to mean shift segmentation.

11. A method for correcting a defect in a digital image containing an eye, the method comprising:

converting the digital image into an intensity image,

identifying a group of pixels containing an eye region within the intensity image;

segmenting the group of pixels containing the eye region of the intensity image into segments each having a local intensity extremum,

thresholding a corresponding portion of the digital image to identify at least two regions of relatively extreme intensity, including a pupil and one or both of an iris or an eyelid;

selecting a region from the at least two regions identified in the thresholding step having the most extreme average intensity, and

identifying segments from the segmenting step intersecting the region selected at selecting step according to a pre-determined criterion, and

wherein the average intensities of the at least two regions from the thresholding step having the most extreme and next most extreme average intensities differ by less than a predetermined amount, and the region selected at the selecting step comprises a region having the most extreme average saturation.

12. The method of claim 11, wherein the thresholding step uses a threshold obtained by identifying the most extreme intensity pixel inside the intensity image, computing the average intensity of the pixels inside a smaller rectangle centered on the most extreme intensity pixel, and assigning the threshold to the sum of the average intensity and a fixed quantity.