Facial Skin Tone High-Frequency Luma Superposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image beautification technologies fail to provide a stereoscopic and moistened effect for face images, often resulting in yellowish or reddish skin tones.
Innovation Solution
An image processing method that identifies the facial skin tone area, filters the image to obtain a smooth image, extracts high-frequency details, and superimposes a luma channel signal of the facial skin tone high-frequency image onto the original image to enhance the luma of facial skin details, thereby improving the stereoscopic and highlighted display effect.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If existing image beautification technologies are used, then image processing can be performed, but the skin tone appears yellowish or reddish and lacks stereoscopic effect
Solution Approach 1:
The patent segments the image processing by separating high-frequency details from the smooth image, and further isolates facial skin tone regions using masking. This allows targeted processing of skin areas without affecting other parts of the image, resolving the color distortion issue while enhancing stereoscopic effect.
Solution Approach 2:
The patent applies different processing strategies to different regions: high-frequency enhancement is applied specifically to facial skin tone areas identified by the mask, while other regions remain unchanged. This localised approach prevents global color distortion and maintains natural skin tones in processed areas.
2Manufacturing precision
If high-frequency enhancement is applied to the entire image, then detail sharpness improves, but color distortion occurs in skin tone areas
Solution Approach 1:
The patent extracts the high-frequency component from the original image and separately applies it only to the masked facial skin tone regions. By taking out the high-frequency enhancement effect and applying it selectively rather than globally, the patent achieves detail sharpness improvement without the color distortion that would result from full-image processing.
3Productivity
If simple filtering is applied, then processing speed is maintained, but stereoscopic and moistened effect is not achieved
Solution Approach 1:
The patent performs preliminary smoothing filtering on the entire image before extracting high-frequency components. This preliminary action creates a clean smooth image that, when combined with the extracted high-frequency details and applied selectively through masking, achieves both the stereoscopic effect and maintains processing efficiency by avoiding repeated complex operations.
Data Source
AI summary
An image processing method includes: obtaining a facial skin tone area in an image to be processed; filtering the image to be processed to obtain a filtered smooth image; obtaining a high-frequency image based on the smooth image and the image to be processed; obtaining a facial skin tone high-frequency image based on the high-frequency image and a facial skin tone mask; and superimposing the high-frequency image and the image to be processed based on the facial skin tone mask and preset first superimposition strength in a luma channel, and superimposing a luma channel signal of the facial skin tone high-frequency image onto a luma channel signal of the image to be processed, to obtain a first image.