Light Source Estimation for Lens Shading Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Miniaturization of digital cameras leads to light attenuation and color distortions due to the close proximity of lenses and sensors, resulting in inaccurate lens shading correction and automatic white balancing, especially when the light source is unknown or misestimated.

Innovation Solution

A computational method for light source estimation that maps image data to a chromaticity space, using linear transformations and principal component analysis to determine the correct light source by measuring compactness and likelihood, improving lens shading correction and automatic white balancing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Volume of moving object

If the distance between lens and sensor is reduced to miniaturize the camera, then the form factor is reduced, but light attenuation and color distortions increase

Engineering Contradiction:

Improvecamera form factorVSAvoidlight attenuation and color distortions

Core Design Contradiction:

Volume of moving objectVSObject-affected harmful factors

Solution Approach 1:

The system performs preliminary light source estimation and selects appropriate lens shading correction tables before final image processing. By predicting the light source type early in the processing pipeline and pre-selecting correction parameters, the system compensates for the optical defects caused by miniaturization before they fully manifest in the final image.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system dynamically changes correction parameters based on detected light source characteristics. Different light sources (incandescent, fluorescent, LED, natural light) have different spectral properties, and the system adjusts the lens shading correction tables and white balancing parameters accordingly to maintain color accuracy despite the fixed short distance between lens and sensor.

Inventive Principle:

Principle #35Parameter changes

2Device complexity

If conventional lens shading correction is used without accurate light source identification, then processing is simpler, but color restoration accuracy deteriorates

Engineering Contradiction:

Improveprocessing complexityVSAvoidcolor restoration accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The system performs self-identification of the light source by analyzing the spectral characteristics of the captured image. The light source estimation module automatically determines whether the scene is illuminated by incandescent, fluorescent, LED, or natural light without external input, and autonomously selects the appropriate correction parameters to maintain color accuracy.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces complex manual or hardware-based light source identification mechanisms with computational analysis. By using image processing algorithms to analyze spectral signatures and chromaticity values, the system substitutes physical measurement devices with software-based detection, achieving accurate light source identification while maintaining relatively simple device architecture.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If multiple light source correction tables are maintained for different illuminants, then color restoration improves, but device complexity increases

Engineering Contradiction:

Improvecolor restoration accuracyVSAvoiddata structure complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the correction data into separate lookup tables for different light source types (incandescent, fluorescent, LED, natural light). Each table contains pre-computed lens shading correction parameters optimized for its specific illuminant category. This segmentation allows the system to store multiple correction sets in an organized manner and quickly retrieve the appropriate table based on light source estimation, managing complexity through structured data organization.

Inventive Principle:

Principle #1Segmentation

4Measurement precision

If light source estimation is performed using spectral analysis, then accuracy improves, but computational complexity increases

Engineering Contradiction:

Improvelight source identification accuracyVSAvoidcomputational power

Core Design Contradiction:

Measurement precisionVSPower

Solution Approach 1:

The system extracts only the most discriminative spectral features from the captured image for light source identification. Instead of analyzing the complete spectral profile, the algorithm focuses on key chromaticity values and color ratios that are most indicative of the light source type. This extraction approach maintains high identification accuracy while significantly reducing the computational burden compared to full spectral analysis.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS10542243B2Method and system of light source estimation for image processing

Publication Date: 2020.01.21 INTEL CORP

AI summary

A system, article, and method to perform light source estimation for image processing includes measuring a compactness of the distribution of the image data to select a light source.