Communicating Photograph Content Through Tactile Images to People With Visual Impairments

Bibliographic Details
Title: Communicating Photograph Content Through Tactile Images to People With Visual Impairments
Authors: Karolina Pakėnaitė, Petar Nedelev, Eirini Kamperou, Michael J. Proulx, Peter M. Hall
Source: Frontiers in Computer Science, Vol 3 (2022)
Publisher Information: Frontiers Media S.A., 2022.
Publication Year: 2022
Collection: LCC:Electronic computers. Computer science
Subject Terms: communicating photograph content, visual impairments, accessibility, tactile image recognition, icons, tactile representations, Electronic computers. Computer science, QA75.5-76.95
More Details: Millions of people with a visual impairment across the world are denied access to visual images. They are unable to enjoy the simple pleasures of viewing family photographs, those in textbooks or tourist brochures and the pictorial embellishment of news stories etc. We propose a simple, inexpensive but effective approach, to make content accessible via touch. We use state-of-the-art algorithms to automatically process an input photograph into a collage of icons, that depict the most important semantic aspects of a scene. This collage is then printed onto swell paper. Our experiments show that people can recognise content with an accuracy exceeding 70% and create plausible narratives to explain it. This means that people can understand image content via touch. Communicating scene foreground is a step forward, but there are many other steps needed to provide the visually impaired with the fullest possible access to visual content.
Document Type: article
File Description: electronic resource
Language: English
ISSN: 2624-9898
Relation: https://www.frontiersin.org/articles/10.3389/fcomp.2021.787735/full; https://doaj.org/toc/2624-9898
DOI: 10.3389/fcomp.2021.787735
Access URL: https://doaj.org/article/dc7a9503904d4b7fa12da2411dcfb873
Accession Number: edsdoj.7a9503904d4b7fa12da2411dcfb873
Database: Directory of Open Access Journals
More Details
ISSN:26249898
DOI:10.3389/fcomp.2021.787735
Published in:Frontiers in Computer Science
Language:English