Image Noise Reduction Using Color Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques struggle to effectively remove noise, particularly in regions with wide coverage or low-frequency noise, leading to inadequate image quality due to resolution degradation.
Innovation Solution
An image processing device and method that detect regions in certain colors, such as achromatic colors, and calculate feature amounts related to luminance, allowing for tailored noise reduction processes with varying strengths based on region characteristics, including stronger noise removal in achromatic and flat regions and weaker removal in border regions to prevent image quality degradation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If a strong noise removal process is applied to the entire image, then noise is effectively removed, but image quality degrades due to resolution loss
Solution Approach 1:
The patent applies different noise removal strengths to different regions of the image based on color characteristics. Achromatic regions (black, white, gray) receive stronger noise removal processing, while chromatic regions receive weaker processing to preserve color accuracy and prevent resolution degradation. This local differentiation resolves the contradiction by tailoring the noise removal intensity to the specific requirements of each region.
Solution Approach 2:
The image is segmented into different color regions (achromatic and chromatic) before applying noise removal. The color determination unit divides the image based on color characteristics, allowing the noise removal unit to apply appropriate processing strength to each segment. This segmentation enables effective noise removal in achromatic regions while preserving image quality in chromatic regions.
2Manufacturing precision
If noise removal is weakened in color regions to prevent resolution degradation, then image quality is preserved, but noise removal effectiveness is insufficient
Solution Approach 1:
The patent identifies achromatic regions where noise is more perceptible and applies stronger noise removal processing specifically to these areas. Since achromatic regions lack color information, they can tolerate stronger filtering without the same level of quality degradation as chromatic regions. This local quality approach ensures noise is effectively removed where it matters most while preserving color accuracy elsewhere.
3Device complexity
If uniform noise removal is applied across all regions, then processing is simple, but noise removal effectiveness varies by region
Solution Approach 1:
The patent segments the image into achromatic and chromatic regions using a color determination unit, then applies different noise removal processing to each segment. This segmentation approach adds minimal complexity while dramatically improving noise removal effectiveness by matching processing strength to regional characteristics. The segmentation enables the system to handle diverse noise patterns across different image regions.
Solution Approach 2:
The patent implements local quality by applying different noise removal strengths to different color regions. Achromatic regions receive stronger processing while chromatic regions receive weaker processing. This local differentiation improves overall noise removal effectiveness without requiring complex adaptive processing throughout the entire image.
Data Source
AI summary
There is provided an image processing device including a color determination unit configured to detect a region in a certain color from an input image, a feature amount calculating unit configured to calculate a feature amount related to luminance of the input image, and a noise reduction unit configured to perform noise removal on the input image on the basis of a result obtained by detecting the region in the certain color, and the feature amount.