Abstract

Working technologically with the limitations imposed by the lack of vision is a challenge when the objective is to guarantee an improvement in the lives of people with visual impairment. Trivial tasks for people with vision become challenges for those who cannot see obstacles, resulting in accidents. Thanks to technology, these tasks may be alleviated for people with visual impairment by detecting safe paths and obstacles. However, there may be different needs and preferences among people with visual impairment, and this issue lacks exploration in the literature, since most of the existing approaches offer little or no flexibility in their functioning, not allowing visually impaired users to self-customize them. Thus, this article presents a research effort towards the autonomous, interactive, and multimodal customization of navigation systems for visually impaired people with total vision loss or severe low vision by introducing an abstract model for interactively customizable multimodal navigation systems to serve as a basis for developing this type of system as well as the development of a vision-based multimodal, embedded, and interactively customizable navigation system with input from 14 members of a local Non-Governmental Organization (NGO) for visually impaired people. The developed system was embedded in a wearable prototype based on a System on Chip, validated, and then submitted to evaluations in two experiments carried out with the NGO members, who also co-designed the wearable aspects of the prototype. The evaluation results show that by allowing a navigation system to be fully customized, different needs and preferences can be satisfied, making it easier to be adopted in the daily lives of visually impaired people and contributing to the state of the art of this type of system’s customizability.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call