# US 2017/0289446 A1 · Automatic white balance using histograms from subsampled image

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2017/0289446 A1
- Document ID: US20170289446A1
- Status: Published application
- Kind: A1
- Application number: 15/084,829
- Filed: 2016-03-30
- Published: 2017-10-05
- Earliest claimed: 2016-03-30
- Assignees: Motorola Mobility LLC
- Applicants: Motorola Mobility LLC
- Inventors: JORGE DE JESUS GOMES LEANDRO; HENRIQUE CUNHA DANTAS; DANIELA RADAKOVIC; CASSIANO K. CASAGRANDE; NICOLAS LUIZ R. VEIGA
- Web page: https://patentgrep.com/patent/US20170289446A1

## CPC

- G06T 5/40: using histogram techniques
- G06T 5/70: Denoising; Smoothing
- H04N 23/10: for generating image signals from different wavelengths
- H04N 23/80: Camera processing pipelines; Components thereof
- H04N 23/88: for colour balance, e.g. white-balance circuits or colour temperature control
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A method includes receiving statistical image data for a set of image data. The statistical image data includes composite intensity metrics for each of a plurality of color channels for each of a plurality of cells of a subsampling grid. A histogram is generated for each of the color channels based on the composite intensity metrics. A gain for each of the color channels is determined based on the histograms. The gains for each of the color channels are applied to the set of image data. The set of image data is stored after applying the gains to generate an image data file.

## Claims

### Claim 1

A method, comprising:
receiving statistical image data for a set of image data, the statistical image data including composite intensity metrics for each of a plurality of color channels for each of a plurality of cells of a subsampling grid;
generating a histogram for each of the color channels based on the composite intensity metrics;
determining a gain for each of the color channels based on the histograms that maximizes a degree of overlap between the histograms, wherein determining the gain comprises:
holding a gain for a selected one of the color channels at unity;
iteratively varying the gains of the color channels other than the selected one; and
determining the degree of overlap for each iteration;
applying the gains for each of the color channels to the set of image data; and
storing the set of image data after applying the gains to generate an image data file.

### Claim 2 (depends on 1)

The method of claim 1, wherein the composite intensity metric comprises an average intensity metric.

### Claim 3 (depends on 1)

The method of claim 1, wherein the plurality of color channels comprise a red channel, a green channel, and a blue channel.

### Claim at position 4

-5. (canceled)

### Claim 6 (depends on 1)

The method of claim 1, wherein the plurality of color channels comprise a red channel, a green channel, and a blue channel, and the selected channel comprises the green channel.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
for each iteration:
determining a candidate gain for each of the color channels other than the selected one;
applying the candidate gains to the histograms;
determining the degree of overlap associated with the candidate gains; and
selecting a set of candidate gains having the highest degree of overlap across all of the iterations as the gains for applying to the set of image data.

### Claim 8 (depends on 1)

The method of claim 1, further comprising normalizing the intensity data for each of the cells of the subsampling grid based on the composite intensity metrics for each of the color channels using a Euclidian distance generated based on the composite intensity metrics for each of the color channels.

### Claim 9 (depends on 1)

The method of claim 1, wherein generating the histograms comprises clipping values in the histogram.

### Claim 10 (depends on 1)

The method of claim 1, wherein generating the histograms comprises applying a smoothing filter to the histogram.

### Claim 11

A device, comprising:
an image sensor to generate image data for a plurality of color channels; and
a processor complex to implement a video front end to generate statistical image data from the set of image data, the statistical image data including composite intensity metrics for each of the plurality of color channels for each of a plurality of cells of a subsampling grid, and an automatic white balancing unit to generate a histogram for each of the color channels based on the composite intensity metrics and determine a gain for each of the color channels based on the histograms that maximizes a degree of overlap between the histograms, wherein the automatic white balancing unit is to hold a gain for a selected one of the color channels at unity, iteratively vary the gains of the color channels other than the selected one, and determining the degree of overlap for each iteration, and wherein the processor complex is to apply the gains for each of the color channels to the set of image data and store the set of image data after applying the gains to generate an image data file.

### Claim 12 (depends on 11)

The device of claim 11, wherein the composite intensity metric comprises an average intensity metric.

### Claim 13 (depends on 11)

The device of claim 11, wherein the plurality of color channels comprise a red channel, a green channel, and a blue channel.

### Claim at position 13

-15. (canceled)

### Claim 16 (depends on 11)

The device of claim 11, wherein the plurality of color channels comprise a red channel, a green channel, and a blue channel, and the selected channel comprises the green channel.

### Claim 17 (depends on 11)

The device of claim 11, wherein the automatic white balancing unit is to, for each iteration, determine a candidate gain for each of the color channels other than the selected one, apply the candidate gains to the histograms, determine the degree of overlap associated with the candidate gains, and select a set of candidate gains having the highest degree of overlap across all of the iterations as the gains for applying to the set of image data.

### Claim 18 (depends on 11)

The device of claim 11, wherein the automatic white balancing unit is to normalize the intensity data for each of the cells of the subsampling grid based on the composite intensity metrics for each of the color channels using a Euclidian distance generated based on the composite intensity metrics for each of the color channels.

### Claim 19 (depends on 11)

The device of claim 11, wherein the automatic white balancing unit is to clip values in the histograms.

### Claim 20 (depends on 11)

The device of claim 11, wherein the wherein automatic white balancing unit is to apply a smoothing filter to the histograms.
