Adaptive Image Sharpening with Dual Overshoot Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image sharpening technologies often produce halo effects and noise magnification when attempting to restore blurred images, and existing halo suppression methods are not entirely satisfactory.
Innovation Solution
The method involves identifying different types of pixels in an image, applying a first level of overshoot control for non-line pixels to reduce halos and a second level of overshoot control for line pixels to maintain local contrast, with the first level being significantly stronger than the second to address specific areas of the image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional sharpening software amplifies high frequency areas, then image sharpness is improved, but halo effect and noise magnification occur
Solution Approach 1:
The patent applies different levels of overshoot control to different pixel types: strong overshoot control for non-line pixels (to suppress halos) and weak overshoot control for line pixels (to preserve local contrast and avoid painterly appearance). This localized differentiation resolves the contradiction by allowing sharpness enhancement where needed while suppressing harmful artifacts in specific regions.
Solution Approach 2:
The patent dynamically adjusts the overshoot control parameter based on pixel classification. By changing the overshoot control level from strong to weak depending on whether a pixel is classified as non-line or line type, the system achieves both sharpness improvement and artifact suppression in different image regions.
2Object-generated harmful factors
If halo suppression is applied to reduce halos, then halo effect is reduced, but local contrast and fine feature sharpness deteriorate
Solution Approach 1:
The patent selectively applies strong overshoot control only to non-line pixels where halos occur, while applying weak overshoot control to line pixels where local contrast is critical. This localized approach suppresses halos in appropriate regions while preserving fine feature sharpness in line regions.
Solution Approach 2:
The patent segments pixels into two categories: non-line pixels and line pixels. This segmentation allows different overshoot control strategies to be applied to different segments, resolving the contradiction by suppressing halos in non-line regions while maintaining local contrast in line regions.
3Device complexity
If uniform overshoot control is applied to all pixels, then processing simplicity is maintained, but image quality deteriorates due to either excessive halos or loss of fine details
Solution Approach 1:
The patent implements pixel-type identification and classification, which adds moderate processing complexity but enables differentiated overshoot control strategies. This results in superior image quality by suppressing halos in non-line regions while preserving fine details in line regions, outweighing the additional processing complexity.
Solution Approach 2:
The patent segments the image processing into two distinct pathways based on pixel classification. This segmentation adds processing steps but enables optimized quality outcomes by applying appropriate overshoot control levels to different pixel types, resolving the trade-off between complexity and quality.
Data Source
AI summary
A method for sharpening a captured image includes the steps of (i) identifying a plurality of edge pixels in the captured image, (ii) reviewing the plurality of edge pixels to identify one or more line pixels, and one or more non-line pixels in the captured image (354), and (iii) sharpening the captured image utilizing a first level of overshoot control for the non-line pixels (362), and utilizing a second level of overshoot control for the line pixels (360) The method also includes the steps of (i) identifying an intensity value for each of a plurality of neighboring pixels that are positioned near a selected pixel in a predetermined pixel window that have the highest intensity values and the lowest intensity values.