Image Noise Reduction Using Motion-Adaptive Temporal and Spatial Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional noise reduction technologies face challenges in effectively reducing noise in motion regions while minimizing motion blur, as they often prioritize temporal or spatial noise reduction over the other, leading to suboptimal performance in both regions.
Innovation Solution
An apparatus and method that adjust temporal and spatial noise reduction intensities based on pixel-based and block-based motion degrees, using a reference frame from which both temporal and spatial noise have been removed, to generate a final output frame with enhanced noise reduction and reduced motion blur.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If temporal noise reduction is increased in non-motion regions, then noise reduction effectiveness is improved, but motion blur increases in motion regions
Solution Approach 1:
The patent applies different noise reduction intensities to different regions of the image based on motion detection. Specifically, it calculates pixel-based motion degrees and block-based motion degrees to identify motion regions versus non-motion regions, then selectively applies temporal noise reduction with adjusted weights to each region type, achieving local optimization of noise reduction effectiveness while minimizing motion blur in motion areas
Solution Approach 2:
The patent dynamically adjusts the temporal noise reduction weight based on detected motion characteristics. The temporal noise reduction weight is calculated as a function of pixel-based motion degree and block-based motion degree, allowing the system to adaptively increase or decrease temporal noise reduction intensity in real-time according to the actual motion state of each region, thereby resolving the contradiction between noise reduction effectiveness and motion blur prevention
2Object-generated harmful factors
If spatial noise reduction is increased in motion regions, then motion blur is reduced, but noise reduction effectiveness decreases in non-motion regions
Solution Approach 1:
The patent applies different noise reduction strategies to different regions: in motion regions, it uses spatial noise reduction with adjusted weights to minimize motion blur, while in non-motion regions, it employs temporal noise reduction with higher weights to maximize noise reduction effectiveness. This localized approach ensures optimal performance in both region types simultaneously
Solution Approach 2:
The patent dynamically switches between temporal and spatial noise reduction modes based on motion detection results. By calculating motion degrees and comparing them against thresholds, the system adaptively determines whether to apply temporal or spatial noise reduction in each region, optimizing the balance between motion blur reduction and noise reduction effectiveness in real-time
3Reliability
If conventional temporal noise reduction is applied using previous frames, then noise reduction is achieved, but motion regions suffer from inadequate noise reduction and increased blur
Solution Approach 1:
The patent enhances conventional temporal noise reduction by introducing local quality assessment through pixel-based and block-based motion degree calculations. It identifies motion regions and applies adjusted temporal noise reduction weights specifically to these regions, ensuring that noise reduction is performed with appropriate intensity locally rather than uniformly across the entire image, thereby maintaining image quality in motion regions while achieving effective noise reduction
Solution Approach 2:
The patent incorporates feedback mechanisms by calculating motion degrees from the input frame and previous frame, then using these motion degree values to adjust the temporal noise reduction weight for the current frame. This closed-loop approach allows the system to continuously adapt the noise reduction process based on detected motion, improving both noise reduction effectiveness and image quality in motion regions compared to conventional open-loop temporal noise reduction
Data Source
AI summary
An apparatus and method for reducing noise of an image are provided. The apparatus includes: a temporal noise reduction unit configured to remove temporal noise from an input frame by adjusting a temporal noise reduction intensity of the input frame based on a pixel-based motion degree and a block-based motion degree that are detected from the input frame and a reference frame and generate a first output frame; and a spatial noise reduction unit configured to remove spatial noise from the first output frame by adjusting a spatial noise reduction intensity of the first output frame based on similarity between peripheral pixels of the first output frame and the pixel-based motion degree and the block-based motion degree, thereby to generate a final output frame.