Image Color Cast Correction Using Pixel Luminance Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image correction technologies are limited in their ability to accurately and automatically correct color casts in images, often resulting in undesirable changes or incomplete corrections, particularly due to their reliance on single color corrections across varying luminance ranges and failure to account for regional differences in images.
Innovation Solution
A method that groups pixels by luminance ranges and determines individual neutral colors and color corrections for each group, applying tailored corrections while protecting specific pixels and using blended corrections to reduce discontinuities between groups, thereby providing high-quality, automated color cast correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single automated color correction is applied to the entire image, then the correction process is simple and fast, but the color correction accuracy is insufficient and may produce undesirable changes in different luminance regions
Solution Approach 1:
The image is divided into multiple luminance ranges (e.g., shadows, mid-tones, highlights), and separate color corrections are determined for each range. This segmentation allows each region to be corrected according to its specific characteristics, improving overall color accuracy while maintaining a systematic correction process.
Solution Approach 2:
Different color corrections are applied to different luminance regions of the image. Each luminance range receives a tailored correction based on its neutral color characteristics, ensuring that local color accuracy is optimized for each region rather than applying a uniform correction across the entire image.
2Measurement precision
If individual color corrections are determined for multiple luminance ranges, then color correction accuracy is improved, but the complexity of determining and applying corrections increases
Solution Approach 1:
By segmenting the image into distinct luminance ranges, the complex task of correcting the entire image is broken down into smaller, more manageable sub-tasks. Each luminance range can be processed independently and more efficiently, reducing the overall processing time while maintaining high accuracy.
Solution Approach 2:
The method determines neutral colors for each luminance range in advance, before applying the final color corrections. This preliminary analysis of neutral colors allows for more efficient correction application, as the correction parameters are pre-calculated based on the specific characteristics of each luminance range.
3Stability of the object's composition
If color corrections are applied to all pixels uniformly, then the correction process is simple, but regional color characteristics are not preserved and discontinuities may occur between different luminance ranges
Solution Approach 1:
Each luminance range receives a color correction tailored to its specific neutral color characteristics. This local quality approach ensures that regional color characteristics are preserved while maintaining consistency within each luminance range, avoiding the discontinuities that would result from uniform correction.
Solution Approach 2:
The method blends adjacent luminance ranges by determining neutral colors that account for both ranges. This merging approach ensures smooth transitions between luminance ranges, maintaining color consistency across the entire image while preserving regional characteristics through the blending process.
Data Source
AI summary
In some implementations, a method provides color correction in images and includes determining multiple input groups of pixels of an image based on one or more pixel attributes of the pixels. Each of the input groups includes different pixels of the image than the other input groups. The method determines an individual neutral color associated with each of the input groups of pixels based on the colors of the pixels in the associated input group. An individual color correction associated with each of a number of output groups of pixels is determined based on one or more neutral colors associated with the input groups. The method applies each individual color correction to its associated output group of pixels, including adjusting the color of pixels in each output group to reduce color casts on the pixels.