Flexible ROI Color Processing for Mixed Illuminant Scenes

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Image capture devices face challenges in setting accurate white balance, particularly when a region-of-interest (ROI) is illuminated by a different light source than the environmental illuminant, leading to color tone shifts in captured images.

Innovation Solution

The implementation of a technique that determines both global and local automatic white balance (AWB) parameters, allowing for flexible local face color improvement by identifying specific illuminants for ROIs, such as faces, and adjusting pixels accordingly to minimize color casts in mixed light scenes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If global AWB parameters are applied uniformly to the entire image, then the overall white balance is improved, but color casts appear in regions illuminated by different illuminants

Engineering Contradiction:

Improvewhite balance accuracyVSAvoidcolor cast in ROI

Core Design Contradiction:

Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent segments the image into multiple regions including global regions and specific ROIs (such as face regions). Different AWB parameters are calculated and applied to different segments, allowing the global image to maintain overall white balance while ROIs illuminated by different illuminants are corrected independently to eliminate color casts.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements local quality by determining that different regions of the image require different white balance characteristics. Specifically, ROIs are identified and assigned dedicated AWB parameters based on their local illuminant conditions, while the global image uses overall AWB parameters, ensuring each region receives appropriate color correction for its specific lighting conditions.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If separate AWB parameters are determined for ROI and global image, then color accuracy in ROI is improved, but processing complexity increases

Engineering Contradiction:

Improvecolor accuracy in ROIVSAvoidAWB processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the AWB processing into segmented steps: global AWB parameter determination for the entire image, followed by ROI detection and separate AWB parameter determination for identified ROIs. This segmentation allows systematic handling of different regions with different complexity levels, improving ROI color accuracy while managing processing complexity through structured approach.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary action by first determining global AWB parameters for the entire image before proceeding to identify ROIs and calculate their specific AWB parameters. This preliminary global processing establishes a baseline that simplifies subsequent ROI-specific processing, reducing overall computational complexity while maintaining high color accuracy in ROIs.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12052516B2Flexible region of interest color processing for cameras

Publication Date: 2024.07.30 QUALCOMM INC

AI summary

A method of camera processing including receiving a first image, determining one or more first automatic white balance (AWB) parameters for the first image, determining one or more second AWB parameters for a region-of-interest (ROI) of the first image, applying the one or more first AWB parameters to one or more of the first image or a second image, and adjusting the ROI of one or more of the first image or the second image based on the one or more second AWB parameters.