De-mosaicing via Intermediate Color Differences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for spatial interpolation of color images from cameras with Bayer Mask or similar sensor patterns often result in image sharpness issues and 'false color' effects, especially when objects have structures or textures similar to the color filter pattern, due to non-co-sited sampling of color components.
Innovation Solution
A method and apparatus for spatially interpolating pixel values by using second color component pixel values to derive intermediate values and combine them with co-sited first color component values, employing filters with specific apertures and coefficients to achieve co-sited pixel values, and utilizing a processor to generate fully-sampled pixel values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple bilinear interpolation of individual colour components is used, then the interpolation process is simple, but image sharpness deteriorates and aliasing occurs
Solution Approach 1:
The patent introduces intermediate values derived from the relationship between different color components as mediators in the interpolation process. Instead of directly interpolating each color component independently, the method uses intermediate calculations that leverage correlations between color components (e.g., using green component information to aid red and blue interpolation), thereby improving image sharpness while maintaining reasonable computational complexity
Solution Approach 2:
The patent changes the interpolation parameters by using adaptive filtering coefficients that vary based on local image characteristics and color component relationships. Rather than applying fixed bilinear interpolation weights, the method dynamically adjusts interpolation parameters based on the derived intermediate values and local variance, improving sharpness while managing complexity through localized parameter adaptation
2Manufacturing precision
If complex adaptive filtering schemes are used, then image sharpness is improved, but false colour effects occur when objects have structures similar to the colour filter pattern
Solution Approach 1:
The patent applies preliminary anti-action by using the derived intermediate values to predict and counteract potential false color artifacts before they manifest. The method calculates intermediate representations that encode local color relationships and uses these to guide the interpolation in a way that prevents the emergence of false color patterns, especially in regions with structures similar to the Bayer mask pattern
Solution Approach 2:
The patent incorporates feedback mechanisms where the derived intermediate values from one color component inform the interpolation of other color components. This cross-component feedback ensures that interpolated values are consistent with local color relationships, preventing false color effects while maintaining sharpness through adaptive adjustment based on local image characteristics
3Productivity
If colour components are spatially sub-sampled with non-co-sited patterns (e.g., Bayer Mask), then the sensor design is efficient, but accurate reconstruction of fully-sampled pixel values becomes difficult
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing the relationships between different color component samples during the interpolation process. The method derives intermediate values that capture the statistical relationships between non-co-sited color samples, enabling accurate reconstruction of fully-sampled pixel values without requiring complex real-time computations
Solution Approach 2:
The patent uses intermediate values as mediators that bridge the gap between non-co-sited color component samples. These intermediate representations encode the spatial and color relationships inherent in the sub-sampled data, enabling accurate reconstruction of fully-sampled pixel values by translating information from one color component to others through the intermediate stage
Data Source
AI summary
This invention concerns the spatial interpolation of color images and, in particular, the reconstruction, or “de-mosaicing” of data from a single sensor-array electronic camera.Electronic cameras typically have an image sensor comprising a matrix of individual pixel sensors, each sensor being responsive to a color component. In order to obtain color component information for all pixels, in accordance with the disclosed embodiment, green component pixel values (203) are interpolated to obtain green component pixel values for all pixels (209). A difference value (B−G) (212) is formed from the green color component values (207) and the original blue component values (204) at blue pixel locations, and the difference values (B−G) (212) are then interpolated to obtain difference values for all pixels (216). Blue component values for all pixels (225) can then be obtained from the difference values for all pixels (216) and the green component values for all pixels (209).