White Balance Processing Using Light Source and Object Color Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing white balance processing methods often result in objects appearing grayish when the light source color is identical to the object's natural color, leading to poor white balance effects.
Innovation Solution
A method that determines whether the light source color is identical to the object's natural color, and if so, performs white balance processing based on the scene type, or if different, determines the light source's color temperature to adjust the white balance accordingly, using a combination of histogram analysis, pixel value ratios, and a classifier to accurately identify and process the light source.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If white balance processing is performed based on light source color temperature detection, then the white balance processing can be automated, but objects appear grayish when the light source color is identical to the object's natural color
Solution Approach 1:
The patent introduces an object color estimation module as an intermediary to provide a reference natural color when the light source color matches the object color. This intermediary component helps distinguish between actual object color and color influenced by lighting, preventing the grayish effect while maintaining automated processing.
Solution Approach 2:
The patent implements dynamic switching between different white balance processing paths based on the comparison result between light source color and object color. The system dynamically adjusts its processing mode: using light source color temperature when they differ, and using object color estimation when they match, thereby maintaining color accuracy across different scenarios.
2Productivity
If the light source color is used directly for white balance processing, then the processing is simple and fast, but the object's color in the image deviates from its real-world color when light source color matches object natural color
Solution Approach 1:
The system dynamically selects the processing path based on the comparison between light source color and object color. When they match, it switches to the object color estimation path; when they differ, it uses the direct light source color path. This dynamic adaptation maintains both processing efficiency and color accuracy.
Solution Approach 2:
The patent implements a feedback mechanism where the system compares the detected light source color with the estimated object natural color, and based on this feedback, determines whether to apply light source-based or object color-based white balance processing. This feedback loop ensures color accuracy without significantly impacting processing speed.
3Device complexity
If white balance processing is performed without distinguishing between light source color and object natural color, then the processing method is simple, but the white balance effect is poor when colors are identical
Solution Approach 1:
The patent segments the white balance processing into distinct modules: light source color detection module, object color estimation module, comparison module, and processing execution module. This segmentation allows the system to implement complex color comparison and decision logic while keeping each individual module relatively simple, thus improving reliability without excessive overall complexity.
Solution Approach 2:
The system dynamically adjusts its processing complexity based on the scene characteristics. When light source color matches object color, it activates the more complex object color estimation path; otherwise, it uses the simpler direct light source color path. This dynamic approach improves reliability only when necessary, minimizing the impact on overall system complexity.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
The present disclosure provides a white balance processing method including processing (S110) an image to obtain a color of a light source contained in a scene corresponding to the image; determining (S120) whether the color of the light source is identical to a natural color of an object contained in the scene; when the color of the light source is identical to the natural color of the object, determining (S130) a type of the scene and performing a white balance processing on the image according to the type of the scene; and when the color of the light source is different from the natural color of the object, determining (S140) color temperature of the light source contained in the scene according to the color of the light source and performing the white balance processing on the image according to the color temperature of the light source.