Edge Enhancement Using Luminance and Motion Intensity Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image enhancement methods fail to accurately distinguish and process small-sized, middle-sized, and large-sized edges, leading to increased noise and unnatural transitions between enhanced and unenhanced areas in images.
Innovation Solution
An imaging processing system that determines edge values and enhancement coefficients for each pixel based on luminance and motion intensity values, using filtering operations and threshold comparisons to differentiate between edge types and adjust luminance values accordingly, thereby enhancing edges with varying intensities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional edge enhancement methods use filtering to obtain high frequency information and superpose it onto the image, then edges are enhanced, but noise increases and small-sized edges cannot be distinguished from noise
Solution Approach 1:
The patent segments edges by size into three categories (small-sized, middle-sized, large-sized) using multiple thresholds applied to high frequency information. This segmentation allows different processing strategies for different edge types, enabling precise enhancement of small edges while suppressing noise that would otherwise be indistinguishable from small edges.
Solution Approach 2:
The patent applies local quality enhancement by using different enhancement coefficients for different edge sizes. Small-sized edges receive one level of enhancement, middle-sized edges receive another level, and large-sized edges receive a third level. This localized differentiation ensures that enhancement is applied appropriately to each edge type without uniformly amplifying noise across the entire image.
2Measurement precision
If thresholds are introduced to distinguish different edge sizes, then edge differentiation is improved, but transitions between enhanced and unenhanced areas become rough and unnatural
Solution Approach 1:
The patent implements dynamic enhancement coefficients that vary continuously based on the high frequency information magnitude. Instead of using fixed threshold-based step functions, the enhancement coefficient changes dynamically and smoothly across different regions, creating natural transitions between enhanced and unenhanced areas while still maintaining the ability to differentiate edge sizes.
Solution Approach 2:
The patent changes the enhancement parameter (enhancement coefficient) based on the magnitude of high frequency information. By adjusting this parameter continuously rather than using discrete threshold levels, the system achieves both precise edge size differentiation and smooth transitions in the enhanced image regions.
3Ease of operation
If global enhancement is applied to the entire image, then processing is simple, but small-sized edges cannot be distinguished from noise and transitions appear unnatural
Solution Approach 1:
The patent segments the image processing into distinct stages: first obtaining high frequency information through filtering, then segmenting edges by size using thresholds, and finally applying differentiated enhancement. This segmented approach maintains operational simplicity while achieving precise edge detail enhancement that global methods cannot provide.
Solution Approach 2:
The patent transitions from global enhancement to local quality enhancement by applying different enhancement coefficients to different regions based on their edge characteristics. This allows the system to maintain the simplicity of automated processing while achieving the precision of localized enhancement for different edge types.
Data Source
AI summary
A method configured to be implemented on at least one image processing device for enhancing edges in images includes obtaining, by the at least one imaging processing device, image data of an image, wherein the image includes a plurality of pixels, and each of the plurality of pixels has a luminance value and a motion intensity value. The method also includes performing at least one filtering operation to the image to obtain, by the at least one imaging processing device, one or more filtered values for each pixel. The method further includes performing a first logical operation to the one or more filtered values of each pixel in the image to obtain, by the at least one imaging processing device, an edge value and an edge enhancement coefficient for each pixel in the image. The method still further includes performing a second logical operation to the luminance value and the motion value of each pixel to obtain, by the at least one imaging processing device, an enhancement adjusting coefficient for each pixel in the image, wherein the enhancement adjusting coefficient is associated with the edge enhancement coefficient. The method still further includes adjusting, by the at least one imaging processing device, the luminance value of each pixel based on at least one of the type, the edge enhancement coefficient, and the enhancement adjusting coefficient of each pixel.