HDR Sensor Auto White Balance for Mixed Illumination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital cameras face difficulties in performing accurate automatic white balancing, especially when an image is illuminated by multiple light sources, leading to unsightly color casts due to the interference of different illumination sources.
Innovation Solution
The method involves using High Dynamic Range (HDR) merge information to generate white balance values and adjust each pixel's balance individually, segmenting the image based on multiple illuminators, and applying different white balance values to different regions of the image, allowing for proper management of mixed illumination scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single global white balance value is applied to the entire image, then the processing is simple and fast, but the white balance accuracy deteriorates in mixed illumination scenes
Solution Approach 1:
The patent divides the image into multiple regions based on illumination characteristics, identifying different light sources and their respective areas. Each region is then processed with its own white balance value, allowing accurate color correction in mixed illumination scenes while maintaining manageable processing complexity through systematic segmentation.
Solution Approach 2:
The patent applies different white balance values to different regions of the image based on local illumination characteristics. Instead of using a single global value, each region receives a customized white balance adjustment tailored to its specific lighting conditions, thereby improving overall white balance accuracy in complex lighting environments.
2Measurement precision
If multiple white balance values are generated for different regions, then the white balance accuracy improves in mixed illumination, but the processing complexity increases
Solution Approach 1:
The image is segmented into distinct regions based on illumination characteristics, allowing efficient processing by treating each region independently. This segmentation enables the system to generate multiple white balance values targeted to specific areas, improving accuracy while maintaining processing efficiency through organized regional handling.
Solution Approach 2:
The patent applies white balance adjustment selectively to regions where it is most needed, rather than uniformly processing the entire image. By focusing computational resources on identifying and correcting problematic regions with mixed or inaccurate illumination, the system achieves high accuracy without excessive processing overhead across the whole image.
3Adaptability or versatility
If traditional auto white balance algorithms are used, then the processing is fast, but the ability to handle multiple light sources deteriorates
Solution Approach 1:
The patent implements illumination segmentation that identifies and separates different light sources in the scene. By dividing the image into regions affected by different illuminators and processing each region with appropriate white balance values, the system achieves versatile handling of mixed illumination while maintaining reasonable processing speeds through efficient regional analysis.
Solution Approach 2:
The system performs preliminary analysis to identify different illumination sources and their characteristics before applying white balance adjustments. This preliminary segmentation and characterization of light sources enables the algorithm to adapt to mixed illumination scenarios effectively, improving versatility while optimizing processing time by preparing region-specific parameters in advance.
Data Source
AI summary
Systems and methods for setting the white balance of an image are described. Embodiments of the systems and methods may receive image data comprising a plurality of exposures, generate a plurality of white balance values based on merge information from a high dynamic range (HDR) merge of the exposures, and adjust a white balance of each pixel of the image data based on the white balance values.