Edge-Aware Variable Clipping Levels for CLAHE Noise Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing real-time sliding window CLAHE algorithms apply a uniform clipping level to all pixels, leading to noise amplification in areas without edge information and excessive contrast enhancement.
Innovation Solution
Calculate a pixel-specific clipping level based on surrounding edge information using a Sobel filter to enhance contrast only in areas with detail and reduce contrast in areas without detail, thereby preventing noise amplification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a uniform clipping level is applied to all pixels in the image, then the contrast enhancement is simplified and processing is faster, but noise is amplified in areas without edge information such as the sky
Solution Approach 1:
The patent applies different clipping levels to different regions of the image based on edge information. Specifically, pixels with edge information receive a first clipping level for contrast enhancement, while pixels without edge information receive a second clipping level to prevent noise amplification. This local differentiation resolves the contradiction by adapting the clipping level to the local characteristics of each pixel region.
Solution Approach 2:
The patent dynamically changes the clipping level parameter based on the presence or absence of edge information in different image regions. By detecting edge information and adjusting the clipping level accordingly, the system optimizes both contrast enhancement and noise suppression, resolving the contradiction between processing simplicity and noise control.
2Object-affected harmful factors
If a separate clipping level is calculated for each pixel based on edge information, then noise amplification is prevented in undetailed areas, but the computational complexity increases
Solution Approach 1:
The patent segments the image into regions with edge information and regions without edge information. By dividing the image based on edge detection results, the system can apply different clipping levels to each segment, preventing noise amplification in undetailed areas while maintaining manageable computational complexity through structured segmentation.
Solution Approach 2:
The patent performs edge information detection as a preliminary step before applying the CLAHE algorithm. By pre-identifying which pixels have edge information, the system prepares the necessary data structure to apply appropriate clipping levels, reducing the overall computational complexity compared to calculating clipping levels for all pixels without such preliminary filtering.
3Object-affected harmful factors
If the clipping level is reduced in areas without edge information, then noise amplification is prevented, but contrast enhancement is reduced in those areas
Solution Approach 1:
The patent applies different clipping levels to different regions: a first clipping level for pixels with edge information to maximize contrast enhancement, and a second clipping level for pixels without edge information to prevent noise amplification. This local quality approach ensures that contrast enhancement is optimized where needed while noise is suppressed where irrelevant details exist.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for calculating the pixel-specific clipping level parameter based on the edge information around each pixel for the real-time sliding window CLAHE algorithm.