White Balance Correction for Face Area in Electronic Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing auto white balance functions in electronic devices adjust white balance based on the background rather than the face, leading to inaccurate skin color representation for individuals under different light sources, particularly affecting those with varying skin tones.
Innovation Solution
An electronic device that identifies the face area and background in an image, determines the skin tone attribute, and adjusts white balance accordingly to provide a more accurate representation of skin color by using specific correction levels based on the identified attribute.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing auto white balance function applies white balance to the entire image based on background, then white balance correction can be applied uniformly, but skin color accuracy deteriorates especially for faces under different light sources
Solution Approach 1:
The image is segmented into face area and non-face area. The processor identifies the face area separately and applies different white balance correction strategies: for face areas, white balance is corrected based on skin tone attributes detected from the face region, while for non-face areas, traditional background-based white balance correction is applied. This segmentation resolves the contradiction by allowing specialized handling of skin tones without compromising overall image white balance.
Solution Approach 2:
Different white balance correction qualities are applied to different regions of the image. Specifically, face regions receive localized white balance correction optimized for skin tone accuracy, while other regions receive standard white balance correction. This local quality approach ensures that skin color accuracy is prioritized in face areas without sacrificing the overall white balance adaptation to different light sources in the entire image.
2Device complexity
If existing auto white balance function identically applies white balance to all skin tones, then processing is simplified, but skin color representation accuracy deteriorates for individuals with different skin colors
Solution Approach 1:
The white balance correction parameters are changed based on the detected skin tone attribute. The processor identifies skin tone characteristics from the face area and adjusts the white balance correction parameters accordingly. Different skin tones receive different correction parameters to achieve accurate skin color representation. This parameter adaptation resolves the contradiction by making the processing more accurate without requiring fundamentally complex additional hardware or processing architectures.
Data Source
AI summary
Disclosed in various embodiments of the present invention are a method and a device for correcting white balance during image capturing in an electronic device. According to the various embodiments of the present invention, a display and a processor functionally connected with the display are comprised, wherein the processor can be configured so as to: acquire an image including one or more objects; determine a first area corresponding to a face among the one or more objects in the image; determine a second area corresponding to the background among the one or more objects; confirm the ratio between color information of the first area and color information of the second area; correct, on the basis of a first level, white balance corresponding to the first area if the ratio satisfies a set first condition; correct, on the basis of a second level, the white balance corresponding to the first area if the ratio satisfies a set second condition; and display, through the display, an image of which the white balance corresponding to the first area has been corrected on the basis of corresponding one level between the first level and the second level. Various embodiments are possible.