Chromatic Aberration Correction via In-Camera Sharpening Reversal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-resolution digital images often suffer from chromatic aberration due to inadequate optical quality in digital cameras, with existing correction techniques failing to effectively account for spatially-varying defocus and intensity distortions, and being limited in their applicability across different images.
Innovation Solution
A technique that models chromatic aberration by reversing sampling, defocusing, and magnification steps, using edge pixels and a blurring kernel to produce un-sharpened intensities, and applying a CA removal function with unknown parameters to correct for chromatic aberration, while also considering the effects of in-camera sharpening.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If chromatic aberration correction is performed by warping color channels to match edges, then chromatic aberration is reduced, but spatially-varying defocus and intensity distortion effects are not accounted for
Solution Approach 1:
The patent applies a preliminary defocus model and intensity distortion model before performing chromatic aberration correction. By pre-characterizing the spatially-varying defocus and intensity effects using calibration data, the system prepares correction parameters in advance that account for these effects, allowing them to be compensated during the CA correction process rather than losing this information.
Solution Approach 2:
The patent introduces calibration images and derived models as intermediaries between the raw image and the final corrected output. These calibration-based models serve as mediators that capture and represent the spatially-varying defocus and intensity distortion characteristics, enabling their effects to be systematically accounted for during chromatic aberration correction.
2Measurement precision
If displacement maps are computed based on edge density and distribution, then chromatic aberration correction is applied, but the extracted maps cannot be used to correct different images due to image-specific edge characteristics
Solution Approach 1:
The patent creates universal calibration data by processing multiple calibration images with varying content and edge distributions. The resulting displacement maps and correction models are derived from aggregated statistics across these diverse images, making them applicable to a wide range of different images rather than being specific to any single image's edge characteristics.
Solution Approach 2:
The patent uses calibration images as copies or proxies for typical scene content. By deriving correction parameters from these representative calibration images rather than from the actual target image, the system creates displacement maps that generalize across different images while avoiding overfitting to image-specific edge patterns.
3Manufacturing precision
If in-camera sharpening is applied to enhance image quality, then image sharpness is improved, but it complicates the chromatic aberration correction process by modifying intensity values
Solution Approach 1:
The patent performs chromatic aberration correction on the sharpened image output directly, treating the sharpening operation as a preliminary step that has already been applied. The correction process works with the intensity values as they exist after sharpening, using calibration-based models to account for the sharpening effects rather than attempting to reverse or complicate the process by undoing sharpening first.
Data Source
AI summary
A chromatic aberration (CA) correction technique is presented that substantially removes CA from an image captured by a digital camera. In general, the effects of any in-camera sharpening are reversed by applying a blurring kernel. The image is then super-sampled to approximate its state prior to the application of in-camera sampling. One of the color channels is designated as a reference channel, and an objective function is established for each of the non-reference channels. The reference color channel is assumed to be CA-free, while the objective functions are used to compute the unknown CA parameters for each non-reference channel. These sets are used in a CA removal function to substantially remove the CA associated with each of the non-reference channels. The image is then sampled to return it to its original resolution, and a sharpening filter is applied if needed to undo the effects of the previously applied blurring kernel.