Method for statistical analysis of images for automatic white balance of color channel gains for image sensors

• US 8,300,930 B2
• Filed: 06/23/2006
• Issued: 10/30/2012
• Est. Priority Date: 06/04/2002
• Status: Active Grant

• Current Assignee: Micron Technology, Inc.
• Original Assignee: Micron Technology, Inc.
• Inventors: Kaplinsky, Michael, Subbotin, Igor
• Primary Examiner(s): Newman, Michael A
• Application Number:
  US11/473,310
• Publication Number:
  US 20070008417A1
• Time in Patent Office: 2,321 Days
• Field of Search: 348/223.1, 348/225.1, 348/229.1, 382/160, 382/164, 382/165, 382/228, 382/272, 382/167
• US Class Current: 382/167
• CPC Class Codes: H04N 23/88 for colour balance, e.g. white-balance circuits or colour temperature control

Abstract

A process for performing white balancing of an image is performed by subdividing an image into a plurality of subframes, and then analyzing each subframe to determine if that subframe is predominantly monochromatic other than gray. If so, that subframe is excluded from the computation of the gain adjustments in the white balancing operation. As a result, the white balance process is performed using only the multicolored and/or gray subframes, thus allowing the overall white-balance of the image to be shifted only when a change in the color average is due to a change in the spectra of illumination, and not a presence of large monochromatic areas in the image.

Claims

1. A method of performing a white balance operation comprising:

receiving an image;

dividing the image into a plurality of individual subframes;

determining whether or not each subframe is substantially monochromatic, wherein the determination is made independently for each individual subframe and is not affected by other subframes in the image, the determining step comprising;

obtaining a hue value for each pixel in the subframe,

calculating an estimated mean hue for the subframe based on the obtained hue values for a subset of lines at a beginning of the subframe, wherein the subset of lines comprises less than all lines of the subframe,

calculating a standard deviation of the hue variances for the subframe using sums of the squares of the differences between the hue value for each pixel in the remaining lines of the subframe not included in the subset of lines and the mean hue, and

comparing the standard deviation with a threshold value,

wherein the hue value for each pixel in the subframe is used for only one of calculating the estimated mean hue or calculating the standard deviation of the hue variance, such that the determining step occurs in a single pass; and

performing a white balance operation on the entire image using hue values only for the subframes which are determined to be not substantially monochromatic to make white balance determinations.

2. The method of claim 1, wherein the hue value obtained for each pixel is expressed as a value within a range from 0 to 360.

3. The method of claim 1, wherein the hue value obtained for each pixel is expressed as a value within a range from 0 to 96.

4. The method of claim 1, wherein performing the white balance operation comprises:

calculating a respective sum of all hue values for each color component in the subframes which are determined to be not substantially monochromatic;

determining a weight for each respective sum so that the three components are equal; and

adjusting the hue values for each pixel in the image according to the determined weight for each color component.

5. A method of performing a white balance operation comprising:

receiving an image;

dividing the image into a plurality of individual subframes;

determining whether or not each subframe is substantially monochromatic, wherein the determination is made independently for each individual subframe and is not affected by other subframes in the image, the determining step comprising;

obtaining a hue value for each pixel in the subframe,

calculating an estimated mean hue for the subframe based on the obtained hue values for a subset of lines at a beginning of the subframe, wherein the subset of lines comprises less than all lines of the subframe,

calculating a standard deviation of the hue variances for the subframe using sums of the absolute values of the differences between the hue value for each pixel in the remaining lines of the subframe not included in the subset of lines and the mean hue, and

comparing the standard deviation with a threshold value,

wherein the hue value for each pixel in the subframe is used for only one of calculating the estimated mean hue or calculating'"'"'the standard deviation of the hue variance, such that the determining step occurs in a single pass; and

performing a white balance operation on the entire image using hue values only for the subframes which are determined to be not substantially monochromatic to make white balance determinations.

6. The method of claim 5, wherein the hue value obtained for each pixel is expressed as a value within a range from 0 to 360.

7. The method of claim 5, wherein the hue value obtained for each pixel is expressed as a value within a range from 0 to 96.

8. The method of claim 5, wherein performing the white balance operation comprises:

calculating a respective sum of all hue values for each color component in the subframes which are determined to be not substantially monochromatic;

determining a weight for each respective sum so that the three components are equal; and

adjusting the hue values for each pixel in the image according to the determined weight for each color component.

9. A method of performing a white balance operation comprising:

receiving an image;

dividing the image into a plurality of individual subframes;

subdividing each subframe into a plurality of macropixels, each macropixel comprising a plurality of pixels of the subframe;

determining whether or not each subframe is substantially monochromatic, wherein the determination is made independently for each individual subframe and is not affected by other subframes in the image, the determining step comprising;

obtaining a single hue value for each macropixel in the subframe as a whole,

calculating a mean hue for the subframe based on the obtained macropixel hue values,

calculating a standard deviation of the hue variances for the subframe using sums of the squares of the differences between the hue value for each macropixel in the subframe and the mean hue, and

comparing the standard deviation with a threshold value; and

performing a white balance operation on the entire image using hue values only for the subframes which are determined to be not substantially monochromatic to make white balance determinations.

10. The method of claim 9, wherein the hue value obtained for each pixel is expressed as a value within a range from 0 to 360.

11. The method of claim 9, wherein the hue value obtained for each pixel is expressed as a value within a range from 0 to 96.

12. The method of claim 9, wherein performing the white balance operation comprises:

calculating a respective sum of all hue values for each color component in the subframes which are determined to be not substantially monochromatic;

determining a weight for each respective sum so that the three components are equal; and

adjusting the hue values for each pixel in the image according to the determined weight for each color component.