Deferred CCM Stages in Autonomous Pipelines for Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing pipelines in autonomous systems face issues with color correction matrices (CCM) being applied too early, leading to out-of-range values and artifacts due to constrained traditional color spaces, which degrade image quality and affect object recognition and tracking accuracy.
Innovation Solution
The CCM is split into two stages within the image processing pipeline, with a first matrix applied earlier and a deferred second matrix applied later, reducing out-of-range values and artifacts by optimizing color correction across multiple stages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the color correction matrix (CCM) is applied early in the image processing pipeline, then color correction is performed efficiently, but out-of-range values and artifacts are generated due to constrained traditional color spaces
Solution Approach 1:
The patent divides the color correction process into multiple stages by splitting the CCM into first and second matrices applied at different pipeline stages. The first matrix is applied after demosaicing, and the second matrix is applied after tone mapping and gamma correction. This segmentation allows each matrix to operate on appropriately prepared data, preventing out-of-range values while maintaining processing efficiency.
Solution Approach 2:
The patent performs preliminary actions by applying tone mapping and gamma correction before applying the second portion of the color correction matrix. This preliminary processing prepares the image data by normalizing tonal ranges and correcting gamma values, ensuring that the subsequent color correction operates on stable, in-range values and prevents artifact generation.
2Device complexity
If traditional color spaces are used for color correction, then processing is simplified, but colors from bright lights outside traditional color spaces cannot be represented accurately
Solution Approach 1:
The patent introduces dynamic adaptability by using two different color correction matrices with different strengths. The first matrix handles initial color correction with moderate strength, while the second matrix provides additional correction with adjusted strength based on the processed image data. This dynamic approach allows the system to adapt to colors outside traditional color spaces without increasing overall system complexity.
Solution Approach 2:
The patent changes the parameter of color correction strength by using matrices with different correction intensities. The first CCM matrix and second CCM matrix are designed with different strength characteristics, allowing the system to progressively adjust color correction intensity to match the dynamic range and color gamut of the input image, thereby representing bright lights and extreme colors accurately.
3Loss of time
If the full strength CCM is applied at an early stage, then color correction is completed quickly, but errors propagate through subsequent processing stages causing artifacts
Solution Approach 1:
The patent segments the color correction operation into two temporal stages separated by tone mapping and gamma correction. This time separation prevents error propagation because each stage operates on cleaned, normalized data from the previous stage, eliminating the harmful artifacts that would otherwise propagate through the entire pipeline.
Solution Approach 2:
The patent introduces tone mapping and gamma correction as intermediary processing stages between the two color correction matrix applications. These intermediary stages act as mediators that normalize the image data, remove artifacts, and prepare clean input for the second color correction stage, thereby preventing harmful factors from propagating through subsequent processing.
Data Source
AI summary
In various examples, image processing techniques are presented for reducing artifacts in images for autonomous or semi-autonomous systems and applications. Systems and methods are disclosed for deferring at least a portion of a strength associated with a color correction matrix (CCM) of an image processing pipeline to one or more other stages associated with the image processing pipeline. For instance, the CCM is used to determine a first CCM and a second, deferred CCM. The first CCM may be associated with an earlier stage of the image processing pipeline while the deferred CCM is associated with a later stage of the image processing pipeline. In some examples, the deferred CCM is combined with another matrix, such as a color space conversion matrix. By deferring part of the CCM, the image processing pipeline may reduce a number of artifacts and/or eliminate the artifacts that occur when processing image data.