Image Correction Device Using Rotated Coordinate Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image correction methods fail to effectively address peripheral shading and local color-shift issues in images, as they cannot adjust gain distribution or resolve color-shift problems at specific image areas, and require complex calculations with many parameters, making them difficult to implement across various platforms.
Innovation Solution
An image correction device and method that uses coordinate transformation and parameter determination to adjust gain values, transforming pixel data coordinates and calculating correction values based on rotated coordinate systems to improve luminance evenness and address peripheral shading and color-shift issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If conventional lens shading correction method is used, then luminance unevenness of lens center and periphery is corrected, but color-shift problem in peripheral areas cannot be resolved
Solution Approach 1:
The patent applies local quality by dividing the image into multiple regions (central region and peripheral regions) and applying different correction methods to each. The central region uses conventional lens shading correction, while peripheral regions use coordinate transformation to calculate direction-specific correction parameters, allowing each area to receive tailored correction for its specific characteristics including color-shift issues.
Solution Approach 2:
The patent segments the correction process into multiple independent steps: first correcting the central region using conventional methods, then identifying peripheral regions with color-shift problems, and finally applying coordinate transformation only to those peripheral regions. This segmentation allows selective application of correction techniques based on local image characteristics.
2Illumination intensity
If quadratic function compensation method is used, then peripheral shading problem is addressed, but calculation complexity and memory space requirements increase greatly
Solution Approach 1:
The patent changes the approach by using coordinate transformation (rotation and translation) instead of complex quadratic functions. This parameter change simplifies the calculation by transforming the problem into a linear coordinate system rotation, reducing the number of calculation parameters from many quadratic coefficients to a few transformation parameters (rotation angle, translation amounts).
Solution Approach 2:
The patent substitutes the complex mathematical quadratic function compensation system with a coordinate transformation system based on linear algebra. This replacement reduces computational complexity while achieving the same peripheral shading correction effect, making the system more efficient and easier to implement.
3Object-generated harmful factors
If AWB function is used, then color-shift problem of whole image is corrected, but local color-shift problems at certain areas cannot be resolved
Solution Approach 1:
The patent applies local quality by detecting which specific regions of the image have color-shift problems and applying coordinate transformation only to those regions. The central region continues to use conventional correction, while only peripheral regions with color-shift issues undergo coordinate transformation, allowing localized correction without affecting the entire image uniformly.
Solution Approach 2:
The patent introduces dynamic selection of correction methods based on image characteristics. The system dynamically determines whether to apply coordinate transformation to peripheral regions based on detected color-shift problems, allowing flexible adaptation to different image conditions and enabling local correction when needed while maintaining simplicity when not needed.
Data Source
AI summary
An image correction device includes a coordinate transformation unit, a parameter value determination unit, and a correction value calculation unit. The coordinate transformation unit receives first coordinates of at least one pixel data of an illuminated pattern under a first coordinate system, and transforms the first coordinates into second coordinates under a second coordinate system, wherein the second coordinate system is rotated relative to the first coordinate system by an angle with respect to an optical center of the illuminated pattern. The parameter value determination unit determines a respective value of at least one correction parameter according to the position of each of the at least one pixel data in the second coordinate system. The correction value calculation unit calculates an image correction value of each of the at least one pixel data according to the second coordinates and the respective value of the at least one correction parameter.