Image Processing Apparatus Noise Detection Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods cannot accurately distinguish between noise interference and object movement, leading to erroneous noise determination and potential loss of image fidelity.
Innovation Solution
An image processing apparatus and method that calculates pixel differences between present and previous images, using a pixel difference calculator, summing unit, determining unit, and output unit to determine noise levels by analyzing positive and negative pixel differences and applying weighted blending values to generate an output image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional MAD technique is used to determine noise presence, then noise detection capability is provided, but object movement cannot be distinguished from noise interference
Solution Approach 1:
The patent segments the pixel difference values into positive differences and negative differences, calculating separate sums for each category. This segmentation allows the system to analyze the distribution pattern of differences, where noise typically produces roughly equal positive and negative sums, while object movement creates asymmetric patterns. This resolves the contradiction by enabling reliable distinction between noise and movement while maintaining noise detection capability.
Solution Approach 2:
Instead of directly comparing absolute pixel differences as in conventional MAD, the patent inverts the approach by separately summing positive and negative differences and comparing their relationship. This inversion reveals that noise affects pixels in both directions equally, while object movement creates systematic directional changes, thereby achieving reliable determination that was impossible with conventional methods.
2Object-generated harmful factors
If mean processing is performed to suppress noise based on MAD values, then noise suppression is achieved, but image streaking occurs due to erroneous suppression of moving objects
Solution Approach 1:
The patent segments pixel differences into positive and negative categories and calculates separate sums. By comparing the relationship between these summed values, the system can identify whether changes are due to noise (balanced positive/negative sums) or object movement (asymmetric sums). This enables selective noise suppression without erroneously processing moving objects, thereby maintaining image fidelity while reducing noise.
Solution Approach 2:
The patent implements a feedback mechanism where the comparison results of positive and negative difference sums are used to determine whether noise suppression should be applied. This feedback loop prevents erroneous suppression of moving objects by continuously monitoring the pattern of pixel differences and adjusting the noise suppression decision accordingly, thus maintaining image quality.
3Manufacturing precision
If high-resolution display apparatuses process video content, then picture quality is improved, but sensitivity to noise interference increases
Solution Approach 1:
The patent segments pixel differences into positive and negative categories, allowing high-resolution display apparatuses to process video content with enhanced noise discrimination capability. By analyzing the segmented difference patterns, the system can distinguish noise from genuine image details, enabling effective noise suppression that preserves the high picture quality expected from high-resolution displays while reducing noise sensitivity.
Solution Approach 2:
The patent implements feedback control where the relationship between summed positive and negative differences determines noise suppression actions. This feedback mechanism allows high-resolution display apparatuses to adaptively process video content, maintaining high picture quality while compensating for increased noise sensitivity through intelligent, context-aware noise reduction.
Data Source
AI summary
An image processing apparatus includes a pixel difference calculator, a summing unit, a determining unit, and an output unit. The pixel difference calculator receives a present image having first pixels and a previous image having second pixels, calculates pixel differences between corresponding first and second pixels, and outputs positive and negative pixel difference values. The summing unit obtains a first output value by adding up those of the positive pixel difference values and a second output value by adding up those of the negative pixel difference values. The determining unit determines a noise level of the present image from the first and second output values, and outputs a blended value. The output unit adds together weights of pixels at the same positions of the present and previous images according to the blended value to generate an output image. An image processing method is also disclosed.