Digital Image Artifact Correction via Difference in Gaussians
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing methods lack consistency in correcting artifacts, particularly in matching blemish colors with surrounding areas, leading to subjective improvements rather than objective comparisons.
Innovation Solution
A processor-implemented method and system that receive a digital image with artifacts, perform blurring functions using different encompassment measures, calculate differences of Gaussians, and apply compositing equations to generate modified images with scaled pixel values, making artifact regions more or less differentiable to the human eye.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual image touch-up software is used to remove blemishes by matching colors, then the image quality is subjectively improved, but the processing consistency across multiple images deteriorates
Solution Approach 1:
The patent replaces manual mechanical image editing operations with an automated computer-based system that performs artifact detection, classification, and correction through algorithmic processing. The system substitutes human subjective judgment with objective computational analysis, achieving consistent results across multiple images without requiring manual intervention.
Solution Approach 2:
The system enables images to correct their own artifacts autonomously through self-service processing. The computer automatically detects artifacts, determines appropriate correction methods, and applies corrections without external human intervention, allowing each image to be processed independently and consistently according to predefined algorithms.
2Measurement precision
If visual distinction methods are used for feature selection in image processing, then the processing is easier to perform, but the processing precision and consistency deteriorate
Solution Approach 1:
The patent segments the image processing task into distinct automated stages: artifact detection, classification, correction method selection, and application. Each stage is handled by specific computational algorithms that analyze image data objectively, replacing subjective visual inspection with precise measurement-based processing.
Solution Approach 2:
The system changes the fundamental parameters of image processing from subjective visual assessment to objective computational measurement. It uses numerical algorithms to detect artifacts based on pixel value variations, applies mathematical models for correction, and measures results quantitatively, thereby improving precision while managing complexity through automation.
3Manufacturing precision
If automated artifact correction is applied to make artifacts less differentiable, then the image quality is improved, but the ability to detect and measure remaining artifacts deteriorates
Solution Approach 1:
The system applies partial correction rather than complete elimination of artifacts. It adjusts pixel values to reduce artifact differentiability to a controlled extent, preserving enough artifact visibility for verification while achieving sufficient correction quality. This prevents over-correction that would completely mask remaining artifacts.
Solution Approach 2:
The patent implements feedback mechanisms that allow verification of correction results. The system compares original and corrected images, measures remaining artifact differentiability, and adjusts processing parameters accordingly. This feedback loop ensures corrections achieve desired quality while maintaining detectability for quality control purposes.
Data Source
AI summary
A processor-implemented method and system of this disclosure are configured to correct or resolve artifacts in a received digital image, using a selection of one or more encompassment measures, each encompassing the largest artifact and an optional second smaller artifact. The processor herein is configured to calculate difference in Gaussians using blurred versions of the input digital image and optionally, the input digital image itself; to composite the resulting difference in Gaussians and the input digital image; and to determine pixels with properties of invariant values, also referred to as invariant pixels. The values in the properties of the invariant pixels are then applied to the artifacts regions to correct these artifacts, thereby generating the modified digital image in which the artifact regions are more or less differentiable to a human eye when compared with the digital image.