Adaptive Slope Tone Mapping for Image Sharpening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image sharpening techniques, such as tone mapping, often result in posterized images due to inadequate contrast correction, failing to effectively address blur caused by camera movement or out-of-focus issues.
Innovation Solution
A processor-based method that identifies and processes pixels differently based on their intensity values, using adaptive contrast correction functions with varying slopes to selectively sharpen images, applying more sharpening to step-like edges and less to smooth areas, thereby reducing posterization and enhancing visual appeal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If existing tone mapping algorithms are used for image sharpening, then edge sharpness is improved, but the image becomes posterized and visually unpleasant
Solution Approach 1:
The patent applies different contrast correction functions to different regions of the image based on edge detection. Step-like edges receive one type of correction while fine lines and smooth areas receive another, allowing localized optimization that prevents posterization in delicate regions while maintaining sharpness in prominent edges
Solution Approach 2:
The patent changes the slope parameter of the contrast correction function adaptively. By adjusting the slope value based on the type of edge detected (steeper slope for step-like edges, gentler slope for fine lines), the system optimizes sharpness while avoiding excessive contrast adjustment that causes posterization
2Manufacturing precision
If uniform sharpening is applied across the entire image, then overall edge clarity is improved, but noise is introduced in smooth areas and fine details are degraded
Solution Approach 1:
The system detects the type of edge at each pixel location and applies different sharpening strengths accordingly. Smooth areas and fine lines receive reduced sharpening to avoid noise amplification, while step-like edges receive enhanced sharpening for maximum clarity
Solution Approach 2:
The image is segmented into different edge types (step-like edges, fine lines, smooth areas) based on local pixel analysis. Each segment is then processed with an appropriate contrast correction function, allowing differentiated optimization without uniform treatment
Data Source
AI summary
A method for sharpening a captured image (14) includes (i) selecting a pixel (240) in the captured image (14); (ii) selecting a selected high intensity value (302); (iii) selecting a selected low intensity value (302); (iv) normalizing the intensity value to establish a normalized intensity value using the selected high intensity value and the selected low intensity value (304); (v) determining an adjusted normalized intensity value for the normalized intensity value using a contrast correction function (306); and (vi) scaling the adjusted normalized intensity value to get a transformed intensity value (308). Subsequently, the adjusted image (16) can be generated using the transformed intensity value for each pixel (240). The contrast correction function can be selected that provides the desired amount of sharpening. Thus, the amount of sharpening that is applied to the image (14) can be specifically selected.