Image Processing Apparatus Motion Noise Differentiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods cannot differentiate between pixel differences caused by noise interference and object motion, leading to potential loss of image fidelity and image streaking due to incorrect classification.
Innovation Solution
An image processing apparatus and method that calculates pixel differences between frames, determines a motion level using positive and negative pixel difference densities, and adjusts blending values to distinguish between noise and motion, ensuring accurate image processing by weighting current and previous image pixels accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional MAD-based pixel difference comparison is used to determine motion, then the method is simple and widely applicable, but it cannot differentiate between noise interference and object motion, leading to misclassification
Solution Approach 1:
The patent divides the pixel difference analysis into multiple segments: positive pixel differences, negative pixel differences, and their respective counts within a sampling window. This segmentation allows the system to analyze the distribution characteristics of pixel differences rather than relying on a single aggregate MAD value, enabling differentiation between motion and noise based on the pattern of differences.
Solution Approach 2:
The patent transitions from one-dimensional MAD value comparison to a two-dimensional analysis by introducing both positive and negative pixel difference counts. This dimensional expansion provides additional information about the nature of pixel changes, allowing the system to determine motion levels based on the combined characteristics of positive and negative differences rather than relying on a single magnitude metric.
2Object-affected harmful factors
If pixel differences are treated as noise and mean processing is performed to cancel noise, then noise interference is reduced, but image streaking occurs due to incorrect classification of motion as noise
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors the ratio of positive to negative pixel differences and adjusts the motion level determination accordingly. This feedback loop allows the system to learn from the distribution patterns of pixel differences and make more accurate classifications, preventing the misidentification of motion as noise while still effectively reducing actual noise interference.
Solution Approach 2:
The patent changes the parameter used for motion detection from a single MAD value to a composite parameter based on the counts of positive and negative pixel differences. This parameter transformation enables the system to capture the directional information of pixel changes, allowing accurate distinction between motion (which produces systematic positive/negative patterns) and noise (which produces random patterns).
Data Source
AI summary
An image processing apparatus includes: a pixel difference calculator for calculating a difference value between each first pixel of a previous image and a second pixel of a present image and at a position corresponding to said each first pixel, and outputting a plurality of pixel differences; a counter counting a number of positive pixel differences and a number of negative pixel differences in the pixel differences of a sampling window; a motion level determining unit calculating a motion level of a pixel in the sampling window according to the numbers of the positive and negative pixel differences; a blending value determining unit determining a blending value according to the motion level; and an output unit adding together weights of the present and previous images according to the blending value to generate and output an output image. An image processing method is also disclosed.