Chromatic Aberration Correction Using Green Pixel Intermediary
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies struggle to accurately correct chromatic aberration of magnification in single-plate type color sensor cameras, particularly affecting the edges of red and blue colors due to limited pixel numbers and frequency limitations, leading to blurred edges and false colors.
Innovation Solution
The method involves calculating the aberration amount of different color component signals relative to a criterion color signal, generating a color difference signal, and correcting chromatic aberration using this signal, leveraging the higher number of green pixels to improve accuracy and prevent blurring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If up-sampling is performed on R or B using only single-color information, then the correction process can be simplified, but high frequency near the limit resolution of G cannot be reproduced, causing edges to be blurred or false colors to appear
Solution Approach 1:
The patent uses G pixels as an intermediary to assist in correcting R and B chromatic aberration. Since G pixels are more numerous and have better frequency characteristics, they serve as a mediator to provide accurate position information for correcting the less numerous R and B pixels, thereby maintaining high correction accuracy without relying solely on limited R or B pixel data
Solution Approach 2:
The patent combines information from multiple color channels (G pixels and R/B pixels) to perform chromatic aberration correction. By merging the abundant G pixel data with the R and B pixel data, the system achieves better frequency characteristics and more accurate edge correction than would be possible using single-color information alone
2Device complexity
If the number of R or B pixels is small in a Bayer arrangement, then the sensor structure is simplified, but the frequency that can be estimated by up-sampling is limited, reducing correction accuracy
Solution Approach 1:
G pixels serve as an intermediary data source to enhance the frequency estimation capability for R and B pixels. The abundant G pixel information compensates for the limited R and B pixel samples, enabling accurate frequency estimation and chromatic aberration correction despite the simplified sensor structure with fewer R and B pixels
3Ease of manufacture
If chromatic aberration correction is performed before color separation, then the correction can be applied to RAW data, but edges of R or B may be blurred due to insufficient high frequency information
Solution Approach 1:
The patent uses G pixel information as an intermediary to preserve high frequency edge information during chromatic aberration correction on RAW data. By leveraging the better frequency characteristics of G pixels, the system can perform correction before color separation while maintaining edge sharpness and avoiding blurring
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively corrects the edges of red and blue colors with high accuracy, preventing blurring and ensuring precise chromatic aberration correction in images from single-plate type color sensors.
Implementation Method 1
chromatic aberration of magnification is known to occur due to an expanded pointing image in light passing through a lens due to an influence of a point spread function (PSF) of each color. The chromatic aberration of magnification occurs due to a difference in the size of an image since the refractive index of a lens is different depending on a wavelength of the visible light
Data Source
AI summary
An image processing apparatus includes: an aberration amount generation unit that calculates an aberration amount of a different color component signal with reference to a position of a color component signal of a criterion color having pixels larger in number than a different color included in image data, based on a luminance value of a pixel signal included in the image data; and a correction unit that generates a color difference signal from the different color component signal and the color component signal of the criterion color and corrects chromatic aberration of the color component signal of the different color present between pixels of the criterion color using the color difference signal based on the aberration amount.