Image Noise Removal via Average Pixel Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods fail to effectively separate signal and noise characteristics, leading to diminished noise elimination effects and reduced signal intensity.
Innovation Solution
An image processing device that produces an average image from multiple captured images, extracts noise pixels by comparing pixel values between individual images and the average image, and interpolates pixel values to eliminate noise while maintaining signal intensity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional noise elimination methods are applied to images with low signal intensity, then noise is reduced, but signal intensity is also reduced
Solution Approach 1:
The image processing is segmented into distinct functional units: signal intensity calculation unit, noise level calculation unit, and weighted composition unit. This segmentation allows independent optimization of each function, enabling noise reduction while preserving signal characteristics through separate calculation and combination steps
Solution Approach 2:
The invention dynamically changes the weighting parameter based on local signal intensity and noise level characteristics. By calculating signal intensity and noise level for each pixel region and using these to determine weights, the system adapts the noise reduction strength to local conditions, preventing uniform over-processing that would reduce signal intensity
2Object-affected harmful factors
If multiple images are processed to eliminate noise, then noise elimination effect is improved, but processing time increases
Solution Approach 1:
The system performs preliminary calculations of signal intensity and noise level for each pixel region before actual noise reduction processing. These preliminary assessments allow the weighted composition to be optimized in advance, reducing the computational burden during the final image generation phase and enabling faster processing of multiple images
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An average image producing means 52 produces an average image from all or some of a plurality of images captured at the same location. A noise extracting means 53 extracts a noise pixel on the basis of the result of a comparison between the pixel values of the pixels in the captured images and the pixel values of the pixels at the same position in the average image. An interpolating means 54 interpolates the pixel value of the noise pixel included in the captured images using the pixel values of other pixels to produce a noise-eliminated image.