Smartphone Color Search Using Reference Chart Illuminant Transformation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for identifying colors in images taken under unknown illuminants are prone to errors due to simple matrix transformations, which can lead to poor performance, especially when the difference between the unknown and known illuminants is significant.

Innovation Solution

A method that captures images of a sample and a reference color chart under multiple unknown illuminants, using a processor to generate transformation matrices based on known illuminant measurement values, allowing for the adjustment of color values to match a known illuminant database, thereby improving color identification accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a simple matrix transformation is used to convert images taken under unknown illuminant to known illuminant, then the need for controlled illumination is eliminated, but errors become very large and system performance becomes poor

Engineering Contradiction:
ImproveAbility to work under unknown illuminantsVSAvoidColor identification accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system performs preliminary actions by capturing images under multiple different illuminants before color transformation. By taking images under several known illuminants in advance, the system creates multiple reference datasets that enable more accurate color conversion later, reducing the errors inherent in simple single-illuminant transformations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes the illuminant parameter by capturing images under multiple different lighting conditions rather than relying on a single illuminant. This multi-illuminant approach provides multiple parameter sets for transformation, allowing the system to select or combine the most appropriate transformation parameters to minimize color accuracy errors

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If multiple images under different illuminants are captured and processed, then color identification accuracy is improved, but the complexity of the system increases

Engineering Contradiction:
ImproveColor identification accuracyVSAvoidSystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system uses the smartphone's existing camera and processing capabilities to capture and analyze multiple illuminant images, making the device self-sufficient. The smartphone itself provides the computational power needed to handle the increased complexity of multi-illuminant processing without requiring external specialized equipment

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system makes the smartphone camera multi-functional by enabling it to serve both as a simple point-and-shoot device and as a sophisticated color measurement instrument. The same camera hardware performs multiple functions: capturing images under different illuminants, storing reference data, and executing color transformation algorithms

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS10748304B2Color search using a smartphone and a reference color chart
Publication Date: 2020.08.18 DATACOLOR
  • US10748304B2 patent drawing
  • US10748304B2 patent drawing
  • US10748304B2 patent drawing

AI summary

A method is provided for identifying colors across illuminants using a processor configured by code executing therein, the method comprising capturing a sample image with an image recording device, the image including pixels resenting a color sample under analysis and a color reference chart. The processor is also configured to execute comparing the color values of pixels representing the color reference chart of the sample image to the pixel values corresponding to a plurality of entries within a plurality of color reference datasets to determine the color reference having the closest match of sample captured in the image. The reference identifier of the color that is the closest match to the sample is then output for further use.