Super-resolution Image Processing via Correlation-Based Compositing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional super-resolution processing for generating high-resolution images from multiple low-resolution images is computationally costly and often results in artifacts like ghost images due to moving objects, degrading the output image quality.
Innovation Solution
An image processing apparatus and method that detects movement between low-resolution images, performs position alignment, calculates correlation information, and adjusts the compositing ratio between high-resolution and lower-resolution images based on correlation, reducing the proportion of high-resolution image in areas with low correlation and increasing it in areas with high correlation to minimize artifacts and improve resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If reconfigurable super-resolution processing is performed with repeated estimation of high-resolution images, then image resolution is improved, but computational cost increases
Solution Approach 1:
The patent divides the image processing into multiple passes: first generating a high-resolution image through pixel arrangement, then performing compositing processing on specific regions (areas with moving objects) separately from other regions. This segmentation allows selective application of computationally intensive operations only where needed, reducing overall computational cost while maintaining resolution improvement benefits in static regions.
Solution Approach 2:
The patent applies different processing strategies to different regions of the image based on local characteristics. Areas with moving objects (low correlation) receive different compositing treatment compared to static areas (high correlation). This local quality approach optimizes computational resources by applying intensive processing only where necessary while using simpler processing elsewhere.
2Power
If the number of repetitions is limited to reduce computational cost, then processing time is reduced, but artifacts such as ghost images increase
Solution Approach 1:
The patent calculates correlation information for different regions and applies different compositing ratios based on local characteristics. Regions with moving objects (low correlation) use compositing ratios that suppress ghost images, while static regions (high correlation) use ratios that maximize resolution improvement. This localized approach reduces artifacts without requiring excessive repetitions throughout the entire image.
Solution Approach 2:
The patent dynamically adjusts compositing ratios based on correlation information calculated from the images. The compositing ratio is not fixed but varies according to the detected movement and correlation in different regions, allowing the system to adaptively reduce artifacts in moving regions while maintaining resolution enhancement in static regions.
3Manufacturing precision
If compositing processing is performed to reduce artifacts, then image quality is improved, but device complexity increases
Solution Approach 1:
The patent segments the image processing into distinct functional units: movement detection, correlation calculation, region classification, and compositing processing. Each unit handles a specific aspect of the problem, making the overall complex process more manageable and implementable through modular design while achieving improved image quality.
Data Source
AI summary
An image processing apparatus including: an image obtaining unit that obtains low-resolution (LR) images acquired in time series; a position alignment unit that aligns the LR images on the basis of a movement between the LR images to generate a high-resolution (HR) image; a correlation calculating unit that calculates correlation information between areas of the LR images, the areas corresponding to partial area in the HR image; a compositing-ratio calculating unit that calculates, for each partial area, a ratio between the HR image and an image to be composited which is generated from the LR or HR images and has a lower resolution than the HR image, such that the proportion of the HR image becomes smaller as the correlation of the area decreases; and an image compositing unit that composites the HR image and the image to be composited according to the ratio.