Spatial Domain Detail Control for Image Noise and Sharpness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processors do not consider spatial factors when performing image processing techniques like noise reduction, sharpness control, color saturation, and color interpolation, leading to compromised image quality in the center portion to improve corner portions, which suffer from noise due to lack of illumination.
Innovation Solution
A global spatial domain detail controlling method that adjusts detail parameters for each pixel based on its space position within the image, allowing separate processing of the center and corner portions by using bivariate polynomials to determine appropriate spatial values for noise reduction, sharpness, color saturation, and color interpolation controls.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If a single specific weighting factor is used for noise reduction across the entire image, then the corner portion noise is reduced, but the center portion becomes blurred
Solution Approach 1:
The patent applies different weighting factors for noise reduction based on the spatial position of pixels. Corner portions use higher weighting factors to reduce noise, while center portions use lower weighting factors to preserve sharpness. This local differentiation resolves the contradiction by allowing each region to be processed according to its specific quality requirements.
Solution Approach 2:
The image is segmented into different regions (center and corner portions) with distinct processing parameters. By dividing the image processing into spatial zones, the patent enables independent optimization of noise reduction for corners while maintaining center portion quality, thus resolving the contradiction between noise reduction and sharpness preservation.
2Manufacturing precision
If the edge gain is increased to sharpen the corner portion, then the sharpness of corner portion is improved, but the center portion becomes over-sharpened and distorted
Solution Approach 1:
The patent implements spatially varying edge gain values where corner portions receive higher gain to enhance sharpness, while center portions receive lower or zero gain to avoid over-sharpening and distortion. This local quality approach ensures each region receives appropriate sharpening treatment without adversely affecting other regions.
Solution Approach 2:
The image is divided into different processing zones with distinct edge gain parameters. By segmenting the sharpening operation into corner-specific and center-specific processing, the patent achieves improved corner sharpness while preserving center portion natural appearance without distortion.
3Manufacturing precision
If the color saturation gain is decreased to clarify corner portion texture, then the image texture of corner portion is clear, but the image color of center portion deteriorates
Solution Approach 1:
The patent applies different color saturation gains to different spatial regions. Corner portions use lower saturation gains to reduce noise-induced color errors and clarify texture, while center portions use higher saturation gains to enhance color vibrancy and prevent color information loss. This resolves the contradiction by allowing region-specific color processing.
4Measurement precision
If the color interpolation threshold is increased to prevent incorrect direction determination in corner portion, then the corner portion color interpolation is accurate, but the center portion direction determination becomes unclear
Solution Approach 1:
The patent implements spatially varying color interpolation thresholds where corner portions use higher thresholds to filter out noise-induced false edge detections, while center portions use lower thresholds to maintain sensitivity for accurate direction determination. This local differentiation resolves the contradiction between noise robustness and edge sensitivity.
Data Source
AI summary
A global spatial domain detail controlling method for an image processor includes adjusting at least one detail parameter corresponding to each pixel during an image processing according to each space position of the each pixel in an image; and performing the each pixel with the image processing according to the at least one detail parameter of the each pixel.