Image Tone Transformation Gain Control for Color Saturation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques for tone transformation in luminance/color difference spaces often lead to color saturation issues, particularly in areas where the luminance component is low, as they do not effectively balance transformation gains across color components.
Innovation Solution
An image processing apparatus and method that generates a signal Z based on luminance and color differences, using this signal to calculate a transformation gain for tone transformation, thereby reducing color saturation by accurately reflecting depth and luminance in the image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If the transformation gain K is determined based on the luminance component Y, then the hue change is prevented, but the color saturation occurs
Solution Approach 1:
The patent applies local quality by determining transformation gain differently for different color components based on their individual characteristics. Instead of using a uniform gain based on luminance, the system calculates separate transformation gains for R, G, and B components, adjusting each according to its specific saturation properties and contribution to luminance. This allows localized optimization of gain for each color channel to prevent saturation while maintaining hue consistency.
Solution Approach 2:
The patent changes the parameter used for gain determination from luminance component Y to a combination of luminance and color difference components. By introducing additional parameters (color differences) into the gain calculation, the system can adjust transformation gains to account for saturation tendencies of individual color components, thereby preventing color saturation while maintaining hue consistency.
2Object-generated harmful factors
If the maximum values of R, G, and B are used to determine correction coefficient, then color saturation of red and blue is prevented, but the processing does not work optimally in luminance/color difference space
Solution Approach 1:
The patent segments the gain determination process by calculating separate transformation gains for each color component (R, G, B) rather than using a single unified gain. This segmentation allows the system to optimize gain for each color channel independently, taking into account their individual saturation characteristics and luminance contributions, thereby achieving both color saturation prevention and optimal performance in luminance/color difference space.
Solution Approach 2:
The patent introduces color difference components as intermediary elements in the gain calculation. By using color differences (such as R-Y, G-Y, B-Y) as intermediaries between luminance and color information, the system can bridge the gap between luminance-based processing and color saturation control, enabling optimal transformation in luminance/color difference space while preventing color saturation.
Data Source
AI summary
An image processing apparatus of the present invention receives a luminance signal indicating luminance or an amount of brightness and input signals indicating a plurality of color differences, and it includes: a signal generator generating a signal Z depending on the input signals indicating the plural color differences; a gain generator generating a transformation gain k(Z) by using the signal Z; and a tone transformation unit which multiplies the luminance signal by the transformation gain k(Z) to calculate a luminance signal after tone transformation.