Super-Resolution Image Processing Apparatus with Quality Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing super-resolution processing techniques face challenges in generating high-resolution images from low-resolution images due to uneven pixel density, making it difficult for users to determine if an optimal low-resolution image set is acquired for accurate high-resolutionization.
Innovation Solution
An image processing apparatus that selects a basis image and reference images, computes displacement amounts, generates deformed images, and computes image characteristic amounts such as filling rate, high-frequency components, and edge amounts to aid users in acquiring an optimal low-resolution image set by displaying these characteristics, allowing for user feedback to adjust parameters and improve image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a plurality of low-resolution images are used for super-resolution processing, then high-resolution image generation is enabled, but pixel density on the high-resolution image space becomes uneven due to motion estimation and registration
Solution Approach 1:
The patent performs preliminary actions by selecting a basis image and reference images before actual super-resolution processing. The displacement amount is computed in advance based on the basis image and reference images, allowing the system to prepare registration parameters before processing the full image set, thereby controlling pixel density distribution.
Solution Approach 2:
The patent applies local quality by differentiating the roles of different images: one image is selected as the basis image while others serve as reference images. This creates a hierarchical structure where the basis image provides the primary framework and reference images provide additional information, allowing localized optimization of image contribution to achieve more uniform pixel density.
2Manufacturing precision
If interpolation processing is performed to fill lacking pixels, then high-resolution image generation is completed, but image accuracy deteriorates due to estimation errors
Solution Approach 1:
The patent implements feedback by computing image characteristic amounts (filling rate, high-frequency component, edge amount) and displaying them to users. This allows users to see the impact of different image selections and adjustment parameters on image quality, enabling iterative optimization to minimize interpolation errors and maximize accuracy.
Solution Approach 2:
The patent applies parameter changes by allowing users to adjust the number of reference images, selection criteria, and other processing parameters based on the displayed image characteristic amounts. This enables optimization of the super-resolution process by modifying parameters to achieve better balance between resolution and accuracy.
3Manufacturing precision
If users acquire low-resolution images with sub-pixel displacements for super-resolution, then high-resolution image quality is improved, but users cannot determine whether optimal images are sufficiently obtained
Solution Approach 1:
The patent provides feedback to users by computing and displaying image characteristic amounts (filling rate, high-frequency component, edge amount) that quantify the quality of the low-resolution image set. This enables users to objectively assess whether their acquired images are optimal for super-resolution processing without requiring complex manual evaluation.
4Manufacturing precision
If the number of reference images is increased for better super-resolution, then image quality is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent applies partial action by allowing users to select a manageable number of reference images (e.g., 2-4 images) rather than requiring exhaustive processing of all available images. The system computes image characteristic amounts that enable quality assessment with a limited set of reference images, achieving good image quality without excessive computational time.
Data Source
AI summary
An image processing apparatus comprises a processing unit for computing displacement amounts between a basis image and each reference image, a processing unit for generating multiple deformed images based on the displacement amounts, the basis image and multiple reference images, a processing unit for setting a threshold of a parameter, a processing unit for selecting image information from the reference image by threshold, a processing unit for generating composed images and weighted images based on the basis image, the displacement amounts and the image information, a processing unit for generating high-resolution grid images by dividing the composed image by the weighted image, a processing unit for generating simplified interpolation images based on high-resolution grid images, a processing unit for generating an image characteristic amount, a display unit for displaying the image characteristic amount and a control unit that controls the necessary processing as necessary.