Iris Color Correction Using Skin Tone Reference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing iris color correction methods fail to maintain natural brightness and quality, often resulting in awkward corrections such as white-eye, especially when correcting for red-eye or gold-eye phenomena, as they lose original iris color and rely solely on iris color for judgment, leading to unnatural results.
Innovation Solution
An iris color correction apparatus that identifies skin color, estimates original iris color using peripheral region pixels, and adjusts brightness to prevent white-eye, using a combination of skin correction color and original iris color for correction, while automatically detecting face and iris regions to reduce user intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If existing iris color correction methods are used to correct red-eye or gold-eye phenomena, then the red-eye phenomenon is corrected, but the original brightness of the iris is lost and white-eye or unnatural appearance occurs
Solution Approach 1:
The patent applies preliminary action by estimating the original brightness of iris pixels before the red-eye phenomenon occurs. The brightness estimation unit calculates the original brightness value based on the assumption that red-eye affected pixels have elevated brightness, allowing the correction process to restore the iris to its pre-red-eye state by removing the excess brightness while preserving the original lighting conditions.
Solution Approach 2:
The patent changes the brightness parameter of affected pixels by calculating a corrected brightness value that removes the red-eye effect. The correction process modifies the brightness parameter specifically for red-eye affected pixels while leaving other pixels unchanged, thereby restoring natural appearance without affecting the overall image lighting.
2Object-affected harmful factors
If correction is carried out by identifying pixels where red-eye occurs and changing their color, then the red-eye color is corrected, but the original iris color information is lost making natural correction difficult
Solution Approach 1:
The patent introduces an intermediary approach by using the yellow component channel as a mediator to estimate the original iris color. Since the yellow channel contains less red-eye contamination compared to the red channel, it serves as a reference to reconstruct the original iris color information that was obscured by the red-eye phenomenon in the red channel.
Solution Approach 2:
The patent changes the color parameters of red-eye affected pixels by calculating corrected R, G, and B values. The correction process adjusts the red component to match the estimated original color while maintaining the brightness characteristics, thereby restoring natural iris color appearance.
3Device complexity
If only iris color is used for red-eye judgment, then the judgment process is simple, but irrational results occur when iris color is close to brown or under red illumination
Solution Approach 1:
The patent transitions from one-dimensional iris color analysis to multi-dimensional analysis by incorporating skin color information as an additional dimension. The judgment process compares iris color characteristics with skin color characteristics to determine whether red-eye is present, thereby resolving ambiguous cases where iris color alone is insufficient for accurate judgment.
Solution Approach 2:
The patent applies feedback by using skin color as a reference standard to verify iris color judgments. The system continuously compares iris region characteristics with skin region characteristics and adjusts the red-eye judgment accordingly, providing feedback-based correction for cases where initial iris-only judgment would be incorrect.
4Object-affected harmful factors
If redness removal correction is carried out based on skin color under red illumination or reddish face conditions, then skin color matching is achieved, but unnatural appearance occurs because the correction is based on incorrect assumptions
Solution Approach 1:
The patent applies preliminary action by performing red-eye judgment before carrying out color correction. The judgment process uses skin color as a reference to determine whether red-eye is actually present, thereby preventing incorrect correction in cases where the face or skin appears reddish due to illumination or natural complexion rather than red-eye phenomenon.
Solution Approach 2:
The patent uses feedback by continuously comparing iris region characteristics with skin region characteristics during the judgment process. This feedback mechanism allows the system to recognize when redness in the iris region is consistent with skin tone rather than red-eye, thereby preventing inappropriate correction and maintaining natural appearance.
Data Source
AI summary
An iris color correction apparatus has a skin color identification section which identifies the skin color of a photographed person in an inputted image, a skin correction color calculation section which obtains a skin correction color that is an iris color of said person that should be shown in said image based on the skin color identified by said skin color identification section, a brightness estimation section which estimates a brightness before a color tone abnormality occurs in pixels of an iris portion of said photographed person from a value of said pixels, and a correction section which corrects the iris color of said photographed person based on said estimated brightness and a correction reference color based on said skin correction color.