BYU ScholarsArchive

Title

Figures in biological journal articles are often unfriendly to people with color vision deficiencies

Authors

Harlan Stevens, Brigham Young University - ProvoFollow
Arwen Oakley, Brigham Young University - Provo
Stephen Piccolo, Brigham Young UniversityFollow

Files

Keywords

Colorblind, Machine Learning, CNN, Accessibility

Abstract

8% of men have red-green color vision deficiencies (CVD).

We focused on deuteranomaly and deuteranopia (deficiency in seeing green) since it is by far the most common CVD.

Colorblind unfriendly figures hinder equity in research and discourage individuals with CVD from pursuing science.

To determine how often researchers use colorblind-unfriendly figures, we classified and labeled 5000 images.

Our annotated dataset will be freely available in the hope that it will prove helpful to other researchers.

We created a computer vision model using a Convolutional Neural Network (CNN) to classify images as colorblind-friendly or not.

BYU ScholarsArchive Citation

Stevens, Harlan; Oakley, Arwen; and Piccolo, Stephen, "Figures in biological journal articles are often unfriendly to people with color vision deficiencies" (2023). Library/Life Sciences Undergraduate Poster Competition 2023. 45.
https://scholarsarchive.byu.edu/library_studentposters_2023/45

Document Type

Poster

Publication Date

2023-03-03

Language

English

College

Life Sciences

Department

Microbiology and Molecular Biology

University Standing at Time of Publication

Junior

Copyright Use Information

https://lib.byu.edu/about/copyright/