Image Processor Correcting Color Misregistration

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional image processing technologies face challenges in accurately correcting magnification chromatic aberration for raw data, particularly due to low pixel density of R and B color components and positional shifts affecting G components, leading to loss of fine structures and destruction of chromatic color structures.

Innovation Solution

An image processor that captures images with multiple color components, acquires positional shift information, and corrects positional shifts using a color misregistration suppression unit, which includes a positional shift correction unit, image structure extraction unit, and image structure compensation unit, utilizing higher pixel density reference color components to interpolate and compensate for lost structures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If simple interpolation is performed for R and B color components to correct magnification chromatic aberration, then color misregistration is corrected, but fine structures of R and B color components are quickly lost due to low pixel density

Engineering Contradiction:

Improvecolor misregistration correction accuracyVSAvoidfine structures of color components

Core Design Contradiction:

Manufacturing precisionVSLoss of information

Solution Approach 1:

The patent performs preliminary color interpolation to generate all color components (R, G, B) for all pixels before magnification chromatic aberration correction. This preliminary action ensures that R and B components have sufficient pixel density (through interpolation from neighboring pixels) to preserve fine structures during subsequent correction operations, rather than correcting the sparse original R and B samples directly

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces G components as an intermediary with high pixel density to guide and support the correction process. By using G components (which have full pixel coverage) to calculate correction amounts and as a reference for preserving fine structures, the method indirectly protects R and B fine structures even though they have low original pixel density

Inventive Principle:

Principle #24Intermediary (Mediator)

2Manufacturing precision

If G components are scaled in accordance with R and B components shifted due to magnification chromatic aberration, then color difference components are produced, but positional shifts between G components and color difference components cannot be neglected in chromatic color structure areas

Engineering Contradiction:

Improvecolor misregistration correctionVSAvoidpositional accuracy of color components

Core Design Contradiction:

Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The patent performs preliminary color interpolation to generate all color components for all pixels before magnification chromatic aberration correction. This ensures that when G components are scaled and color difference components are calculated, all components are already interpolated to the same pixel grid, eliminating positional misalignment issues in chromatic color structure areas

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the processing sequence and parameters: instead of scaling G components based on shifted R/B positions, it first interpolates all components to a common grid, then applies correction. This parameter change (processing order and interpolation step) resolves the positional accuracy problem

Inventive Principle:

Principle #35Parameter changes

3Manufacturing precision

If interpolation is performed for R and B color components in raw data with Bayer array, then color misregistration correction is achieved, but pixel density of R and B components remains low at one-quarter of total image pixels

Engineering Contradiction:

Improvemagnification chromatic aberration correctionVSAvoidpixel density of color components

Core Design Contradiction:

Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent performs preliminary color interpolation to generate all color components (R, G, B) for all pixels in the image before magnification chromatic aberration correction. This preliminary interpolation increases the effective pixel density of R and B components from 1/4 to 100% of total pixels, ensuring sufficient data points for accurate correction while preserving fine structures

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses the high-density G components (which are already present at all pixel positions in Bayer array) to serve the low-density R and B components. By calculating correction amounts using G components and applying them to interpolated R and B components, the method allows R and B to benefit from G's high pixel density without requiring additional physical R and B sensors

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS7945091B2Image processor correcting color misregistration, image processing program, image processing method, and electronic camera

Publication Date: 2011.05.17 NIKON CORP

AI summary

An image processor captures image data which include at least correction object color components and reference color components and has one kind of color component per pixel. The image processor acquires or detects information about positional shifts of the correction object color components and corrects positional shifts of correction object color components. In this correction, image structures lost from correction object color components are compensated for by image structures extracted from reference color components.