Modeling, Measuring, and Compensating Color Weak Vision.

Area of Science:

• Color science
• Computational geometry
• Visual perception

Background:

• Color perception varies significantly between color-normal and color-weak individuals.
• Existing methods for color image compensation often involve approximations and can introduce distortions.

Purpose of the Study:

• To develop a novel method for simulating color-weak perception and compensating color images using Riemann geometry.
• To create a global color-weak compensation map that preserves perceived color differences accurately.

Main Methods:

• Applied Riemann geometry to model transformations between color spaces of different observers.
• Characterized color space metrics using ellipsoids based on just-noticeable-differences from color-matching experiments.
• Introduced normal coordinates in Riemann spaces to construct a global compensation map, avoiding local linearization errors.

Main Results:

• Developed isometries in Riemann spaces that preserve perceived color differences for both observer types.
• The normal coordinates approach provides a global map free from approximation errors and origin shifting issues.
• Analyzed variations in Riemann metrics for different observers and tested three method variations.

Conclusions:

• Riemann geometry offers a robust framework for accurate color space transformations and image compensation.
• The normal coordinates method represents a significant improvement over previous approximation-based techniques.
• The developed methods have potential applications in image processing and assistive technologies for color vision deficiency.