Adaptive Image Correction for Color-Blind Users
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices struggle to effectively adjust images to be recognizable by color-blind users, often causing significant changes to the original image and losing its naturalness.
Innovation Solution
An image correction method that uses a target color adjustment strategy formed by different preset color adjustment strategies, adjusting the original image to make colors distinguishable by color-blind users while minimizing color change and preserving naturalness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If color replacement method is used to adjust images for color-blind users, then colors become distinguishable, but the original image's naturalness is lost
Solution Approach 1:
The patent transforms the color adjustment approach from discrete color replacement to continuous parameter transformation. By converting colors to HSV space and adjusting the Hue parameter within a rotation angle range, the system achieves color differentiation for color-blind users while maintaining the overall color distribution and natural appearance of the original image.
Solution Approach 2:
The patent introduces dynamic adjustment through rotation angles that can vary based on image content. Different pixels can have different rotation angles within a specified range, allowing the color adjustment to adapt to local image characteristics while maintaining global naturalness. This dynamic approach replaces the static color replacement method.
2Reliability
If single color adjustment strategy is used, then specific colors are corrected, but the overall image quality deteriorates
Solution Approach 1:
The patent applies partial action by selectively adjusting only the Hue parameter within the HSV color space while leaving Saturation and Value unchanged. This partial adjustment is sufficient to achieve color differentiation without excessively altering the image. Additionally, the rotation angle is applied within a controlled range to avoid over-correction.
Solution Approach 2:
The patent implements local quality by allowing different rotation angles for different pixels based on their local color characteristics. The color adjustment is tailored to each pixel's needs while maintaining consistency with the overall image style, thus improving local color recognition without compromising global image quality.
Data Source
AI summary
The present disclosure discloses an image correction method, an electronic device and a computer readable storage medium. The image correction method includes: obtaining an original image and a preset color adjustment strategy; obtaining pixel statistical information of the original image; determining a target color adjustment strategy based on the pixel statistical information and the preset color adjustment strategy; and adjusting the original image based on the target color adjustment strategy to obtain a target image with colors distinguishable by a user. The image correction method can adjust an original image through a target color adjustment strategy formed by different preset color adjustment strategies, so as to allow colors in the original image can be distinguished by a color-blind user. In addition, the color change in the original image is small, so as to improve the naturalness of the original image.