Automatic Red Eye Removal Using Weighted Color Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for red eye removal in photographs require user intervention or have high computational complexity, and existing automated methods do not effectively distinguish between red eye regions and other image features.
Innovation Solution
A method that calculates weighted red values for each pixel using specific weightings for red, green, and blue color values and luminance, automatically selects red eye pixels, and corrects them by grouping contiguous pixels into regions, rejecting non-round pupils, regions too close to others, and those not proximate to facial or sclera features, ultimately replacing red eye pixels with black.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If automated red eye removal software is used, then user intervention is reduced, but computational complexity increases
Solution Approach 1:
The red eye removal process is divided into distinct stages: candidate region identification using color constraints, shape-based filtering to eliminate non-pupil candidates, and final verification. This segmentation allows each stage to focus on specific computational tasks, reducing overall complexity while maintaining automation.
Solution Approach 2:
The algorithm applies different processing strategies to different regions of the image. Candidate pixels are identified using local color properties (red dominance), then shape constraints are applied locally to each candidate region. This localized approach reduces computational complexity by avoiding global image analysis while maintaining effective red eye detection.
2Productivity
If simple color-based detection is used, then processing speed increases, but accuracy in distinguishing red eye from other features decreases
Solution Approach 1:
The detection process is segmented into multiple filtering stages. First, pixels are identified as candidates based on simple color constraints (red dominance). Then, shape constraints are applied to filter candidates. Finally, additional criteria eliminate false positives. This multi-stage segmentation maintains processing speed while progressively improving detection accuracy.
Solution Approach 2:
The algorithm uses multiple parameter thresholds to distinguish red eye from other features. Color constraints check for red dominance with specific threshold values. Shape constraints evaluate aspect ratio and circularity with defined thresholds. These parameter changes at different stages enable accurate discrimination while maintaining efficient processing.
3Measurement precision
If multiple processing stages are applied, then red eye detection accuracy improves, but computational complexity increases
Solution Approach 1:
The processing is divided into sequential stages with clear input-output relationships. Stage 1 identifies candidate pixels using color constraints. Stage 2 groups candidates into regions and applies shape constraints. Stage 3 applies additional filtering criteria. Each stage builds on previous results, improving accuracy while keeping individual stage complexity manageable through the segmentation structure.
Data Source
AI summary
A method for removing a red eye from an image includes (1) calculating a weighted red value for each pixel in the image from red, green, and blue color values and a luminance value of each pixel in the image, (2) selecting a plurality of pixels in the image having weighted red values greater than a threshold as red eye pixels, and (3) correcting some of the red eye pixels to remove the red eye from the image. The weighted red value for a pixel is calculated as follows:f=c1r+c2g+c3bY,wherein f is the weighted red value, r is the red color value, g is the green color value, b is the blue color value, c1 is a first weight given to the red color value, c2 is a second weigh given to the green color value, c3 is a third weight given to the blue color value, and Y is the luminance.