Super-Resolution Image Synthesis Using Motion Estimation Reliability Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image resolution enhancement techniques suffer from noise artifacts in high resolution images due to motion estimation errors, as they do not adequately evaluate the reliability of motion estimation results.
Innovation Solution
An apparatus and method that includes a motion estimation reliability evaluating component to assess the reliability of motion estimation results, using corresponding relationship and analogy degree evaluations, and applies these evaluations to weight pixels in low resolution images for synthesis, thereby minimizing the impact of unreliable pixels in the high resolution image generation process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If motion estimation is performed to align low resolution images for super-resolution processing, then image alignment and resolution enhancement are achieved, but motion estimation errors introduce noise artifacts in the generated high resolution image
Solution Approach 1:
The patent changes the parameter of reliability evaluation from binary (correct/incorrect) to continuous (degree of reliability), allowing the system to weight pixels according to their motion estimation reliability. This resolves the contradiction by enabling precise alignment while filtering out noisy regions through reliability-based weighting in the synthesis process
Solution Approach 2:
The patent introduces reliability evaluation values as an intermediary between motion estimation and image synthesis. These values act as a mediator that identifies and weights reliable pixels, preventing noise propagation from erroneous motion estimates while preserving the alignment benefits
2Measurement precision
If sub-pixel accuracy motion estimation is used to achieve high precision alignment, then image registration quality improves, but estimation errors and noise become more apparent in the synthesized image
Solution Approach 1:
The patent implements feedback by evaluating the reliability of motion estimation results and using this evaluation to adjust the weighting of pixels during synthesis. The reliability assessment feeds back into the synthesis process, allowing the system to correct for estimation errors by down-weighting unreliable regions while maintaining the precision benefits
3Quantity of substance
If all pixels from low resolution images are used in synthesis, then image data completeness is maximized, but unreliable pixels from erroneous motion estimation degrade the final image quality
Solution Approach 1:
The patent applies local quality by assigning different reliability weights to different pixels based on their individual motion estimation quality. Instead of uniformly treating all pixels, the system identifies reliable regions and weights them higher, while down-weighting noisy regions, thus maintaining data completeness while filtering out harmful noise
Data Source
AI summary
In super-resolution processing in which a plurality of low resolution images are synthesized to generate a high resolution image, a high-quality high-resolution image is to be generated with suppression of the noise ascribable to motion estimation error. Motion estimating means 11 estimates motion of pixels between a basis image selected out of plural low resolution images and remaining reference images, and outputs a result of motion estimation. Motion estimation reliability evaluating means 12 evaluates reliability of the result of motion estimation output from the motion estimating means 11, based on such as analogy in luminance of pixels, correlated by the results of the motion estimation, and outputs a motion estimation reliability value indicating the degree of reliability. High resolution image estimating means 16 synthesizes the respective pixels of the input low resolution images with weighting conforming to the motion estimation reliability value output from the motion estimation reliability evaluating means 12 to generate a high resolution image.