Automatic White Balance Control Using Gray Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional white balance control methods fail to accurately detect and adjust for light source changes, leading to incorrect color representation in images due to the introduction of chromatic components into the gray region, especially when dominant colors are present.
Innovation Solution
An apparatus and method that convert input image data into a chromatic color space, extract a gray region based on plural light sources, calculate initial light source information using color gamut average and center values, and adjust this information to exclude chromatic components by dividing the gray region into low and high luminance regions, applying gains to control the white balance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional white balance control methods are used, then the process is simple, but chromatic components are introduced into the gray region causing inaccurate color detection
Solution Approach 1:
The patent segments the gray region into multiple sub-regions based on luminance values (dividing into high luminance region and low luminance region). This segmentation allows selective processing of different gray region portions to eliminate chromatic components while maintaining accurate light source detection, thus resolving the contradiction between measurement precision and process complexity.
Solution Approach 2:
The patent applies different processing strategies to different parts of the gray region. Specifically, it identifies and excludes chromatic components from specific sub-regions while preserving the overall gray region structure. This local quality approach improves light source detection accuracy without requiring complete restructuring of the white balance control process.
2Measurement precision
If the gray region is extracted without considering luminance differences, then the process is straightforward, but chromatic components contaminate the detected gray region
Solution Approach 1:
The patent divides the gray region into multiple sub-regions based on luminance thresholds (high luminance region where Y > Yth and low luminance region where Y ≤ Yth). This segmentation enables precise identification of chromatic component locations while maintaining a systematic detection approach, thereby improving extraction accuracy without excessively increasing detection difficulty.
Solution Approach 2:
The patent introduces luminance (Y component) as an additional parameter for gray region analysis. By incorporating luminance thresholds and analyzing RGB differences within specific luminance ranges, the method transforms the detection process into a multi-parameter analysis that improves accuracy while maintaining manageable complexity through structured parameter evaluation.
3Measurement precision
If dominant colors are present in the input image, then the image has vivid color characteristics, but the gray region detection becomes inaccurate due to chromatic component introduction
Solution Approach 1:
The patent applies local quality processing by identifying chromatic components in specific sub-regions of the gray region and selectively eliminating them. This approach maintains adaptability to diverse lighting conditions (including dominant colors) while improving white balance accuracy, as it processes only the affected regions rather than requiring complete reprocessing of the entire image.
Solution Approach 2:
The patent implements a feedback mechanism by calculating RGB differences within the gray region, identifying chromatic components based on these differences, and then adjusting the white balance control accordingly. This feedback loop enables the system to adapt to various lighting conditions including dominant colors while maintaining accurate white balance control.
Data Source
AI summary
An automatic white balance control image apparatus, medium, and method. The automatic white balance control imaging apparatus may include a color space conversion unit converting the input image data into color space data having a chromatic component, a gray region extraction unit extracting a gray region for the input image data based on gray regions for light sources, the gray regions being detected based on plural light sources and a predetermined color checker, a detection unit extracting initial light source information using a color gamut average value and a color gamut center value, and adjusting the initial light source information in order for a chromatic component introduced into the extracted gray region to be excluded, a gain control unit calculating gains to scale the input image data based on the adjusted initial light source information, and controlling a white balance by applying the calculated gains to the input image data.