Patch-Based Image Sharpening Reducing Halo Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image sharpening techniques, such as unsharp masking and smart sharpen, often result in halo and ringing artifacts, and lack noise suppression capabilities, exacerbating noise and compression artifacts, especially when applied globally.
Innovation Solution
The method involves applying a blur kernel to an input image to create a blurred image, determining patch updates based on local spatial adaptation by searching for similar patches in the blurred image's neighborhood, and accumulating these updates to form a high-frequency image, which is then combined with the input image to sharpen it iteratively, reducing halo and ringing artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If unsharp masking or smart sharpen is applied globally to sharpen an image, then image sharpness and detail are improved, but halo and ringing artifacts are generated around edges
Solution Approach 1:
The patent applies local quality by transitioning from global sharpening operations to local patch-based operations. Each image patch is processed individually by finding similar patches in a local neighborhood and computing sharpening based on local differences, rather than applying the same global operation to the entire image. This localized approach prevents halo and ringing artifacts from propagating across edge regions while maintaining sharpness within local areas.
Solution Approach 2:
The patent segments the image into multiple patches and processes each patch independently. By dividing the image into local regions and handling each region separately through patch matching and local difference computation, the method avoids the artifact generation that occurs with global sharpening operations applied uniformly across the entire image.
2Measurement precision
If unsharp masking or smart sharpen is applied globally, then contrast enhancement is achieved, but noise and compression artifacts are amplified
Solution Approach 1:
The patent applies local quality by computing sharpening operations on local patches rather than globally. By restricting the sharpening computation to local neighborhoods and using patch matching to find corresponding regions, the method enhances local contrast only where needed while avoiding the amplification of noise and compression artifacts that occurs with global contrast enhancement.
Solution Approach 2:
The patent converts the harmful effect of noise amplification into a benefit by using local patch matching to identify reliable correspondence regions. By computing differences only between matched patches rather than all pixels globally, the method selectively enhances contrast in reliable regions while suppressing noise amplification in unreliable or textured regions where patch matching fails.
3Measurement precision
If iterative smart sharpen is used to improve accuracy, then sharpening precision is enhanced, but halo and ringing artifacts are exacerbated and noise increases
Solution Approach 1:
The patent applies local quality by performing iterative sharpening operations on local patches rather than globally. Each iteration computes local differences between matched patches and applies sharpening only to those local regions, preventing the cumulative buildup of halo and ringing artifacts that occurs with global iterative sharpening while maintaining accuracy through repeated local refinement.
Data Source
AI summary
Techniques for sharpening an image using local spatial adaptation and/or patch-based image processing. An image can be sharpened by creating a high-frequency image and then combining that high frequency image with the image. This process can be applied iteratively by using the output of one iteration, i.e., the sharpened image, as the input to the next iteration. Using local spatial adaptation and/or patch-based techniques can provide various advantages. How to change the intensity at a given position in the image can be calculated from more than just information about that same position in the input image and the blurred image. By using information about neighboring positions an improved high frequency image can be determined that, when combined with the input image, reduces ringing and halo artifacts, suppresses noise boosting, and/or generates results with sharper and cleaner edges and details.