Pixel Interpolation Device Preventing False Colors in Bayer Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pixel interpolation methods, such as those described in JP2008-98971A and JP2006-261789A, generate false colors when capturing objects with high-frequency vertical or horizontal stripes, particularly in white-and-black patterns, due to inadequate handling of pixel mixing in Bayer array-based color images.
Innovation Solution
A pixel interpolation device that separates and mixes pixels of the same color in specific patterns within a Bayer array, using a correlation determination device to identify the correlation direction in the image, thereby reducing the influence of certain pixels and generating interpolation pixels that prevent false color generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If pixel mixture is performed by adjusting pixel addition ratio or amplification amount to arrange centers of gravity at equal intervals, then image reduction is achieved, but false colors are generated when capturing high-frequency striped patterns
Solution Approach 1:
The patent segments the pixel mixture process by direction. It separates horizontal pixel mixing from vertical pixel mixing, performing correlation calculation and pixel addition independently for each direction. This segmentation allows the system to handle high-frequency patterns in specific directions without generating false colors, while maintaining image reduction capability.
Solution Approach 2:
The patent applies different processing characteristics to different spatial locations and directions. By calculating correlation values separately for horizontal and vertical directions, and selectively performing pixel addition based on local pattern characteristics, the system adapts the mixing process to local image content, preventing false colors in striped regions while maintaining reduction elsewhere.
2Device complexity
If pixels of the same color are mixed without direction-specific processing, then processing simplicity is maintained, but false colors appear in high-frequency vertical or horizontal stripe patterns
Solution Approach 1:
The patent introduces dynamic, conditional processing into the pixel mixing operation. Instead of a fixed mixing approach, the system dynamically determines whether to perform horizontal or vertical pixel addition based on correlation calculations. This dynamic adaptation allows the processing complexity to increase only when necessary (in striped regions), while maintaining simplicity in other areas.
Solution Approach 2:
The patent changes the processing parameters (mixing direction, correlation threshold) based on local image characteristics. By adjusting which pixels are mixed and in what direction based on calculated correlation values, the system optimizes color interpolation accuracy for high-frequency patterns without requiring a complete redesign of the mixing architecture.
Data Source
AI summary
The generation of a false color is prevented. In a pixel mixture block, pixels of the same color are mixed. For example, red pixels are mixed to obtain a red mixed pixel and blue pixels are mixed to obtain a blue mixed pixel. Green pixels in an odd-numbered row and pixels in an even-numbered row are separately mixed to obtain a green mixed pixel and a green mixed pixel. A correlation direction in a partial color image 1 is detected from these mixed pixels. Interpolation pixels are generated from the mixed pixels according to the correlation direction such that a false color is not generated.