Adaptive Directional Spatial Filter for Edge-Enhanced NLM Denoising
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-local means (NLM) image denoising techniques suffer from unsmooth edges at low Signal-to-Noise Ratios (SNR), leading to poor edge smoothness as noise increases, which compromises image quality.
Innovation Solution
The implementation of an adaptive directional non-local means image denoising method that incorporates an adaptive directional spatial filter, where the spatial filter's directionality is adapted based on an edge metric, varying from uniform to anisotropic filter widths aligned with the spatial gradient direction, to enhance edge smoothness without compromising detail preservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If non-local means (NLM) image denoising is applied to reduce noise, then noise reduction is improved, but edge smoothness deteriorates at low Signal-to-Noise Ratios
Solution Approach 1:
The patent applies local quality by making the spatial filter properties location-dependent. The filter width is adapted locally based on the edge metric calculated at each target pixel location. In regions detected as edges (high edge metric), a larger filter width is used to preserve edge smoothness, while in non-edge regions, a smaller filter width maintains detail sharpness. This resolves the contradiction by allowing different filtering strengths in different local regions of the image.
Solution Approach 2:
The patent implements dynamics by making the spatial filter width adaptive rather than fixed. The filter width dynamically adjusts based on the calculated edge metric for each target pixel. This dynamic adaptation allows the filter to respond to local image characteristics, improving edge smoothness in edge regions while preserving details in non-edge regions, thus resolving the contradiction between noise reduction and edge smoothness.
2Shape
If directional spatial filter is applied to improve edge smoothness, then edge smoothness is improved, but detail preservation deteriorates
Solution Approach 1:
The patent applies local quality by making the filter width location-dependent based on edge detection. In edge regions, a larger filter width is used to smooth edges, while in non-edge regions, a smaller filter width preserves fine details. This selective application resolves the contradiction by applying strong smoothing only where needed (at edges) and minimal smoothing where details exist (in non-edge regions).
Data Source
AI summary
System, apparatus, method, and computer readable media for edge-enhanced non-local means (NLM) image denoising. In embodiments, edge detail is preserved in filtered image data by weighting of the noisy input target pixel value with other pixel values based on self-similarity and further informed by a data-driven directional spatial filter. Embodiments herein may denoise regions of an image lacking edge characteristics with a more uniform spatial filter than those having edge characteristics. In embodiments, directionality of a spatial filter function is modulated based on an edge metric to increase the weighting of pixel values along an edge when there is a greater probability the edge passes through the target pixel. In further embodiments, the adaptive spatial filter is elliptical and oriented relative to a spatial gradient direction with non-uniform filter widths that are based on the edge metric.