Image Signal Skeleton Component Separation for Noise Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing techniques for digital cameras often result in noise amplification during tone conversion, which degrades image quality, as they lack effective methods to separate and manage signal and noise components adaptively.
Innovation Solution
An image processing device and method that separates the original image signal into a skeleton component and a noise component, allowing for adaptive tone conversion and noise reduction by setting tone conversion coefficients and noise reduction parameters based on signal levels, and applying soft-thresholding processing to reduce noise effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If tone conversion processing is performed on the original image signal, then the bit-width is matched to the output system, but noise component amplification occurs
Solution Approach 1:
The original image signal is segmented into a skeleton component (preserving edges and structures) and a remaining component (containing noise and texture). Tone conversion is applied selectively to the skeleton component, preventing noise amplification while maintaining output bit-width compatibility.
Solution Approach 2:
The skeleton component is extracted from the original image signal using edge detection or gradient-based methods. By isolating and processing only the skeleton component for tone conversion, the harmful noise amplification is avoided while achieving proper bit-width matching for the output system.
2Manufacturing precision
If adaptive tone conversion is performed on each region, then image quality is improved, but noise amplification occurs more prominently
Solution Approach 1:
The image is divided into regions, and within each region, the skeleton component is separated from the remaining component. Adaptive tone conversion is applied to the skeleton component based on regional characteristics, improving image quality without amplifying noise in the remaining component.
Solution Approach 2:
Different processing strategies are applied to different components: the skeleton component receives adaptive tone conversion tailored to each region's characteristics, while the remaining component is preserved with minimal processing to avoid noise amplification, achieving local optimization of image quality.
3Shape
If tone conversion coefficient is increased to improve contrast, then contrast is enhanced, but noise component becomes more conspicuous
Solution Approach 1:
The image signal is segmented into skeleton and remaining components. Tone conversion with enhanced contrast coefficients is applied only to the skeleton component, which contains edges and structures. The remaining component, containing noise, is not subjected to aggressive contrast enhancement, preventing noise from becoming conspicuous while still achieving desired contrast improvement.
Data Source
AI summary
An image processing device separates an original image signal into a plurality of components including a first component serving as a skeleton component and a second component obtained after the first component is separated from the original image signal, obtains a signal level of the first component or the original image signal, sets a tone conversion coefficient to be applied during tone conversion based on the signal level of the first component or the original image signal, performs tone conversion processing on the first component using the tone conversion coefficient, obtains the signal level of the first component, sets a noise reduction processing parameter on the basis of the signal level of the first component, and reduces a noise of the second component using the noise reduction processing parameter and the tone conversion coefficient.