Multi-Illuminant White Balance via Illumination Map Segmentation

Solution Overview

Problem

Traditional digital image processing systems struggle to accurately correct white balance in scenes with multiple illuminants, often resulting in incomplete color correction and visible color casts, as they are programmed to correct for only a single dominant illuminant.

Innovation Solution

A method for image processing that generates an illumination map to estimate multiple illuminants in an image, divides the image into regions corresponding to each illuminant, and applies white balancing operations independently to each region using clustering algorithms and constraints to refine illuminant estimates, ensuring accurate color correction across the image.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional AWB algorithms correct for a single dominant illuminant, then the processing complexity is low, but the white balance accuracy in multi-illuminant scenes deteriorates

Engineering Contradiction:

Improvewhite balance accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The image is divided into multiple regions based on illuminant estimation, with each region processed independently for white balance correction. This segmentation allows the system to handle multiple illuminants separately, improving accuracy without requiring complex global processing of the entire image at once.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different white balance corrections are applied to different regions of the image based on their local illuminant characteristics. Each region receives a customized correction tailored to its specific lighting conditions, rather than applying a uniform correction across the entire image.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If traditional AWB algorithms process the entire image with a single illuminant correction, then the algorithm simplicity is maintained, but the color correction completeness deteriorates

Engineering Contradiction:

Improvecolor correction accuracyVSAvoidalgorithm complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The image is divided into multiple regions based on illuminant estimation, with each region processed independently for white balance correction. This segmentation allows the system to handle multiple illuminants separately, improving accuracy without requiring complex global processing of the entire image at once.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The algorithm dynamically adjusts the number of illuminants and region boundaries based on the specific characteristics of each image. This adaptive approach allows the system to increase complexity only when necessary, maintaining simplicity for single-illuminant scenes while enabling multi-illuminant handling when needed.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If multiple illuminants are corrected independently in different regions, then the white balance accuracy improves, but the computational resources required increase

Engineering Contradiction:

Improveilluminant estimation accuracyVSAvoidcomputational energy

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The image is divided into multiple regions based on illuminant estimation, with each region processed independently for white balance correction. This segmentation allows the system to handle multiple illuminants separately, improving accuracy without requiring complex global processing of the entire image at once.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The algorithm applies full computational resources to identify and correct the dominant illuminants, while using simplified approaches for secondary illuminants. This partial action strategy ensures accurate correction of the most significant color casts while reducing overall computational burden.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS11323676B2Image white balance processing system and method

Publication Date: 2022.05.03 APPLE INC

AI summary

One limitation of traditional imaging systems is that they are only programmed to correct for a single color of illuminant in a scene. In multi-illuminant scenes, the detected illuminant color may correspond to some mixture of scene illuminants. This may lead to incomplete color correction, wherein, e.g., the dominant illuminant is corrected for but the color cast caused by secondary illuminants is still visible, or an at least partially visible color cast remains from multiple of the scene illuminants. Thus, the techniques disclosed herein comprise: obtaining an image of a scene; generating an illumination map for the obtained image; dividing the values in the illumination map to determine a number of estimated illuminant regions, wherein each region corresponds to at least one estimated illuminant present in the captured scene; estimating a white point for each region; and applying white balancing operations, based on the estimated white points for each region.