Alternating Upsampling Resizing Demosaicked Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods for mosaicked images face a trade-off between image quality and computational complexity, with simpler methods often resulting in inferior quality and more complex methods requiring excessive computation.
Innovation Solution
The method involves up-sampling and resizing a mosaicked image in alternating directions to generate a resized and partly demosaicked image, which is then processed to produce a demosaicked and improved image, reducing computational demand while enhancing image quality by ensuring at least one full-sampled color plane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simpler image processing methods are used, then computational complexity is reduced, but image quality deteriorates
Solution Approach 1:
The patent segments the image processing task into distinct stages: first performing demosaicking on individual color planes (separating the complex task), then performing sharpening and artifact removal only on the green channel (focusing computation where it matters most). This segmentation allows simpler overall processing while maintaining high image quality through targeted computation.
Solution Approach 2:
The patent applies different processing strategies to different parts of the image data. Specifically, it performs full demosaicking and sharpening on the green channel (which has the highest detail content and is most susceptible to artifacts), while using simpler processing for red and blue channels. This local quality approach reduces overall computational complexity while preserving image quality in critical regions.
2Manufacturing precision
If more complex image processing methods are used, then image quality is improved, but computational complexity increases
Solution Approach 1:
The patent applies partial demosaicking rather than complete demosaicking. It performs full demosaicking only for the green channel (which benefits most from it), while using simpler interpolation methods for red and blue channels. This partial action approach achieves sufficient image quality improvement without the excessive computational complexity of complete demosaicking across all channels.
3Productivity
If demosaicking is performed without proper resizing, then computational steps are reduced, but image artifacts increase
Solution Approach 1:
The patent performs preliminary resizing operations before the demosaicking process. It resizes the input image to an intermediate resolution, performs demosaicking on this resized image, then performs additional sharpening and artifact removal. This preliminary action ensures that the demosaicking operates on appropriately scaled data, preventing artifacts while maintaining processing efficiency.
Data Source
AI summary
An embodiment of the invention provides a method of processing a mosaicked image. First, the mosaicked image is up-sampled along a first direction to generate a first intermediate image. Then, the first intermediate image is resized along the first direction to generate a second intermediate image. Next, the second intermediate image is up-sampled along a second direction to generate a third intermediate image. Afterward, the third intermediate image is resized along the second direction to generate a resized and partly demosaicked image.