Summary
International Workshop on Smart Info-Media Systems in Asia
2022
Session Number:RS2
Session:
Number:RS2-6
An Image Conversion Method for Color Discriminability Compensation of Colorblindness Using CycleGAN
Hideaki Orii, Koyuki Hatano, Hiromu Tanaka, Hideaki Kawano,
pp.124-127
Publication Date:2022/9/15
Online ISSN:2188-5079
PDF download
Summary:
In this paper, we propose a novel image conversion method to compensate color discriminability of colorblindness for natural images. People with colorblindness cannot differentiate between a specific set of colors. Colors that are hard to discretize by people with colorblindness are known as indiscriminable colors. The proposed method is an image transformation method that automatically recolors indiscriminable colors in natural images. In the proposed method, training dataset is generated from a dataset of natural images using simple rules without depending on manpower, and the image transformation is trained using CycleGAN. In the experiments, the proposed method was applied to images with various color combinations to confirm its effectiveness.