Image Processing Gain Control for False Contour Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image processing apparatuses face challenges in tone compression processing, leading to the generation of false contours in gradation areas, such as optically blurred portions, when attempting to reproduce wide dynamic range scenes.
Innovation Solution
An image processing apparatus that detects gradation areas and applies different gain characteristics based on luminance values, using a first gain characteristic for non-gradation areas and a second gain characteristic with reduced gain change for gradation areas to perform gain processing, thereby reducing the occurrence of false contours.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If tone compression processing is performed on an image to reproduce wide dynamic range scene characteristics, then the dynamic range reproduction is improved, but false contours are generated in gradation areas
Solution Approach 1:
The patent applies different gain characteristics to different regions of the image based on gradient magnitude. In gradation areas (low gradient regions), a gain characteristic with smaller changes is used to prevent false contours, while in non-gradation areas (high gradient regions), a gain characteristic with larger changes is used to maintain tone compression effectiveness. This local differentiation resolves the contradiction by adapting the processing strength to the local image characteristics.
Solution Approach 2:
The patent dynamically selects gain characteristics based on the local gradient magnitude at each pixel position. The gain characteristic changes dynamically according to the gradient conditions, using a first gain characteristic for non-gradation areas and a second gain characteristic with reduced changes for gradation areas. This dynamic adaptation allows the system to maintain tone compression while preventing false contours in appropriate regions.
2Productivity
If gain processing with a gain characteristic having large gain changes is applied to enhance tone characteristics, then tone compression effectiveness is improved, but false contours are generated in gradation areas
Solution Approach 1:
The patent applies different gain characteristics to different regions of the image based on gradient magnitude. In gradation areas (low gradient regions), a gain characteristic with smaller changes is used to prevent false contours, while in non-gradation areas (high gradient regions), a gain characteristic with larger changes is used to maintain tone compression effectiveness. This local differentiation resolves the contradiction by adapting the processing strength to the local image characteristics.
Solution Approach 2:
The patent dynamically selects gain characteristics based on the local gradient magnitude at each pixel position. The gain characteristic changes dynamically according to the gradient conditions, using a first gain characteristic for non-gradation areas and a second gain characteristic with reduced changes for gradation areas. This dynamic adaptation allows the system to maintain tone compression while preventing false contours in appropriate regions.
Data Source
AI summary
A method includes detecting a gradation area in an input image, and performing, based on a detection result acquired by the detecting, for an area that is not the gradation area, gain processing on the input image by using a gain based on a first gain characteristic in which different gains are set depending on luminance values, whereas performing, for the gradation area, gain processing on the input image by using a gain in which an amount of change in gain with respect to a change in luminance value is more reduced than the first gain characteristic.