Super-Resolution Image Reconstruction Using Local Pixel Relationships
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current super-resolution image reconstruction methods fail to consider the inherent relationship between adjacent pixels, leading to jagged, blurred, and grainy displays, or have complex calculations with poor real-time performance.
Innovation Solution
A method that determines gray scale values for newly added pixels between adjacent pixels in a row and column pixel groups based on visible relationships, inserting these pixels to reconstruct a super-resolution image, considering the internal relationships between local pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If neighbor replacement algorithm, bilinear interpolation algorithm, or cubic interpolation algorithm is used to fill blank pixels, then the resolution is doubled to achieve higher display resolution, but the calculation process becomes complex and real-time performance deteriorates
Solution Approach 1:
The patent segments the image processing into two distinct phases: first doubling the resolution by inserting blank pixels, then filling those blank pixels by referencing only the four corner pixels of each 2x2 block. This segmentation separates the resolution enhancement from the pixel value calculation, simplifying the overall process while maintaining high resolution.
Solution Approach 2:
The patent applies different processing strategies to different regions: original pixels retain their full detail while newly inserted pixels are calculated using a simplified method based on corner pixel values. This local differentiation maintains image quality where it matters most while reducing computational burden for the newly added pixels.
2Manufacturing precision
If neighbor replacement algorithm, bilinear interpolation algorithm, or cubic interpolation algorithm is used to fill blank pixels, then the resolution is doubled to achieve higher display resolution, but the calculation efficiency decreases and real-time performance becomes poor
Solution Approach 1:
The patent extracts and utilizes only the essential information from corner pixels to determine the values of newly inserted pixels, discarding the need for complex calculations involving multiple neighboring pixels. This extraction of key information maintains resolution quality while dramatically improving calculation efficiency.
3Manufacturing precision
If neighbor replacement algorithm, bilinear interpolation algorithm, or cubic interpolation algorithm is used to fill blank pixels, then the target matrix of 8K resolution is achieved, but jagged, blurred, and grainy display problems occur due to not considering inherent relationship between neighboring pixels
Solution Approach 1:
The patent performs preliminary action by first doubling the resolution and inserting blank pixels before calculating their values. This preliminary structure establishment allows subsequent pixel value calculation to be based on a stable framework of corner pixels, ensuring consistent and high-quality display results without jagged or blurred artifacts.
Data Source
AI summary
The present application provides a super-resolution image reconstruction method and a device. Since the super-resolution image reconstruction method fully considers internal relationship between local pixels of an original image, detailed information of an image can be better maintained, while problems of aliasing, blurring, and graininess are significantly reduced, and display effect of the original image after stretching is improved.