Adaptive White Balancing Using Illuminant Estimation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image capture devices struggle to accurately render images with dominant colors, as fixed color correction matrices assume neutral color integration, which is not valid for scenes like single-color expanses, leading to inaccurate white balancing and noticeable artifacts in video mode due to environmental noise and scene changes.

Innovation Solution

A method that analyzes scene parameters to determine the best guess illuminant and applies a corresponding color correction matrix, switching to a default matrix if the difference exceeds a tolerance, and using hysteresis tests to manage transitions between illuminant modes, ensuring stable white balancing and color correction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If a fixed color correction matrix is used for white balancing, then the process is simple and fast, but it fails to accurately render images with dominant colors like sky, water, or grass

Engineering Contradiction:

Improvesimplicity of white balancing processVSAvoidaccuracy of color rendering for dominant color images

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent transitions from a fixed color correction matrix to a dynamic system that adapts the white balancing coefficients based on scene analysis. The system determines whether to use a fixed matrix or compute adaptive coefficients by analyzing the captured image for dominant colors, allowing the white balancing process to dynamically adjust to different scene conditions and maintain accuracy across diverse imaging scenarios

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the parameters of the color correction process by introducing adaptive white balancing coefficients that vary based on scene characteristics. Instead of using a static matrix, the system modifies the balancing coefficients according to the detected dominant colors and lighting conditions, enabling accurate rendering across different color scenarios while maintaining computational efficiency

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If adaptive white balancing coefficients are computed from image statistics, then color rendering accuracy improves, but the processing time and computational complexity increase

Engineering Contradiction:

Improveaccuracy of white balancingVSAvoidprocessing time for white balancing

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-computing and storing lookup tables for white balancing coefficients corresponding to various illuminant conditions. During image processing, the system first analyzes the captured image to determine the likely illuminant, then retrieves the pre-computed coefficients from the lookup table, avoiding the need for time-consuming real-time computation while maintaining adaptive accuracy

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by creating a simplified model of the complex white balancing computation through lookup tables. Instead of performing full adaptive computation every time, the system copies pre-computed results from the lookup table based on the analyzed scene conditions, significantly reducing processing time while preserving the accuracy benefits of adaptive white balancing

Inventive Principle:

Principle #26Copying

3Measurement precision

If the system switches between fixed and adaptive white balancing modes, then accuracy for different scene types improves, but instability and artifacts occur in video mode due to environmental noise and scene changes

Engineering Contradiction:

Improveaccuracy for different scene typesVSAvoidstability of white balancing in video mode

Core Design Contradiction:

Measurement precisionVSStability of the object's composition

Solution Approach 1:

The patent applies beforehand cushioning by implementing hysteresis thresholds that prevent frequent switching between white balancing modes. The system requires a significant change in scene analysis before switching from fixed to adaptive mode, cushioning against premature transitions caused by environmental noise or minor scene variations, thereby maintaining stability in video mode while still adapting to genuine scene changes

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Solution Approach 2:

The patent uses feedback by continuously monitoring scene parameters and comparing them against hysteresis thresholds to determine when mode switching is appropriate. The system feeds back the scene analysis results to the white balancing decision logic, ensuring that mode changes are driven by significant and sustained scene characteristics rather than transient variations, thus maintaining compositional stability during video recording

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS7421121B2Spectral normalization using illuminant exposure estimation

Publication Date: 2008.09.02 APTINA IMAGING CORP

AI summary

After image capture, scene parameters are analyzed, e.g. lux, flicker, or world estimation. A best guess illuminant for the scene parameters is determined. At this point, the white balancing coefficients corresponding to that illuminant may be applied.