Composite Still Image Generation via Pixel-Wise Alignment Accuracy Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating high-quality still images from dynamic images often fail to accurately align frames, leading to suboptimal weighting and compromised image quality due to reliance on image movement amounts rather than position alignment accuracy.
Innovation Solution
A method involving position alignment, accuracy calculation, and pixel-wise weighting of frame images to generate a composite still image, where the composition ratio between frames is adjusted based on the calculated accuracy of alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If weighting is determined based on image movement amount, then the processing is simple, but the image quality and resolution of the composite still image deteriorate
Solution Approach 1:
The patent changes the parameter used for weighting from image movement amount to position alignment accuracy. This parameter change resolves the contradiction by providing a more accurate basis for weighting that directly reflects the quality of alignment, thereby improving image quality without excessive complexity increase
Solution Approach 2:
The patent replaces the simple movement amount calculation with a more sophisticated position alignment accuracy calculation system. This substitution introduces a new mechanism that accurately measures alignment quality, resolving the contradiction between simplicity and image quality
2Manufacturing precision
If position alignment accuracy is calculated for each pixel, then the image quality improves, but the processing complexity increases
Solution Approach 1:
The patent segments the image into individual pixels and calculates position alignment accuracy for each pixel independently. This segmentation allows for precise local weighting that improves image quality while distributing the computational load in a manageable way
Solution Approach 2:
The patent applies local quality by determining weighting values for each pixel based on its specific position alignment accuracy. This allows different regions of the image to be weighted differently according to their individual alignment quality, improving overall image quality while maintaining processing efficiency
Data Source
AI summary
A non-transitory recording medium storing a computer readable program that causes a computer to perform position aligning, accuracy calculating, and image combining. The position aligning is performing position alignment of other frame images with respect to one reference frame image selected from among multiple frame images that form a dynamic image. The accuracy calculating is calculating an accuracy of the position alignment for each pixel of each of the other frame images. The image combining is combining the reference frame image with the other frame images after the position alignment to generate one composite still image. In the image combining, the computer performs weighting and generates the one composite still image, the weighting being processing of changing a composition ratio between the other frame images for each pixel according to the accuracy calculated by the accuracy calculating.