HDR Image Anti-Aliasing with Local Brightness Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image anti-aliasing methods, such as Fast Approximate Anti-Aliasing (FXAA), struggle to balance the clarity of image details with the anti-aliasing effect in highlight areas, particularly when using a sub-pixel anti-aliasing quality parameter, leading to either poor anti-aliasing in highlights or blurred details.
Innovation Solution
An image anti-aliasing method that generates a customized scaling factor for each edge pixel in HDR space, adjusting color ratios based on local brightness environments to achieve balanced clarity and anti-aliasing effects by dynamically mixing pixel colors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the sub-pixel anti-aliasing quality parameter is set to 1 to maximize anti-aliasing effect, then the anti-aliasing effect in highlight area is optimized, but the image details become blurred
Solution Approach 1:
The patent applies local quality by using a scaling factor that varies across different regions of the image. The scaling factor is determined based on local brightness environment, allowing different parts of the image to have different anti-aliasing intensities. Specifically, highlight areas with high brightness values receive stronger anti-aliasing effects, while other areas maintain their original clarity, thus resolving the contradiction between anti-aliasing effect and detail clarity
Solution Approach 2:
The patent changes the parameter of anti-aliasing intensity dynamically based on local brightness. By introducing a scaling factor that is calculated from the brightness values of surrounding pixels, the system adjusts the anti-aliasing strength adaptively. This parameter change allows the system to apply maximum anti-aliasing (parameter=1) only where needed (highlight areas), while using lower parameters elsewhere to preserve detail clarity
2Manufacturing precision
If the sub-pixel anti-aliasing quality parameter is set to 0 to maintain clarity of image details, then the image details remain clear, but distinct aliasing exists in the highlight area
Solution Approach 1:
The patent implements local quality by making the anti-aliasing effect location-dependent through the scaling factor. Each pixel's anti-aliasing treatment is determined by its local brightness environment, allowing clear preservation of details in most areas while providing targeted anti-aliasing correction in highlight areas where it is most needed
3Adaptability or versatility
If continuously adjusting the sub-pixel anti-aliasing quality parameter is performed, then a balance between clarity and anti-aliasing effect is attempted, but the balance cannot be achieved effectively
Solution Approach 1:
The patent applies dynamics by making the anti-aliasing parameter dynamic rather than static. The scaling factor is calculated dynamically for each pixel based on its local brightness environment, allowing the system to adapt to different regions automatically. This dynamic approach replaces the ineffective continuous manual adjustment with an automated, context-aware parameter determination system
Solution Approach 2:
The system performs self-service by automatically determining the appropriate anti-aliasing parameter for each pixel based on local brightness characteristics. The scaling factor calculation uses the brightness values of surrounding pixels to self-determine the optimal anti-aliasing intensity, eliminating the need for external manual adjustment and achieving effective balance automatically
Data Source
AI summary
A method, apparatus, and a non-transitory computer-readable medium storing program code for anti-aliasing images may be provided. The method may include obtaining a first edge pixel of an image in a high-dynamic range (HDR) space and determining a first scaling factor based on a local brightness environment of the first edge pixel. The method may also include determining a color ratio corresponding with the first edge pixel and scaling the color ratio to obtain a first color ratio. Then the first color value is obtained based on a color value of the first edge pixel, a color value of a target pixel, and the first color ratio, wherein the target pixel has a large color value difference from the first edge pixel among a plurality of surrounding pixels of the first edge pixel; and updating the first color value to the color value of the first edge pixel.