SightGlow: A Web Extension to Enhance Color Perception and Interaction for Vision Deficiency

TL;DR

SightGlow addresses accessibility for color-blind and low-vision users by implementing a web extension that combines real-time color Daltonization, selective zoom, and adjustable color controls. It uses a pipeline of $RGB -> LMS$ transformation and applies a $[4x5]$ Daltonizing matrix to generate perceptually distinguishable colors while preserving alpha. An iterative, user-centered development process guided by a pilot study with nine participants informs UI choices such as larger fonts and simplified controls. Results suggest enhanced readability and engagement for color-blind users, with future work to validate on color-vision users and extend PDF/document support.

Abstract

SightGlow is a web extension tailored to improve color perception accuracy for individuals with red-green color blindness. The research was focused on evaluating whether personalized color adjustment and selective zoom enhance user interaction and satisfaction for individuals with low vision and color vision impairment. The system was developed as an iterative process by conducting a pilot user survey. Existing web extensions were limited in addressing challenges faced by low vision and color blindness; hence this application provides additional features, including selective zoom and color controls, which make it unique. Most participants responded that the application's flexibility to adjust the color balance for any images or video graphic content enhanced their user experience, hence resulting in the effectiveness of the system.

Figures (8)

• Figure 1: User Interface SightGlow.
• Figure 2: Normalized responsivity spectra of human cone cells, S, M, and L types.
• Figure 3: System flow diagram.
• Figure 4: Redesigned interface.
• Figure 5: Top image represents raw colored image whereas button represents daltonized.