Enhancing Visual Accessibility of Indoor Spaces

Project Summary/Abstract The visual accessibility of a space refers to the effectiveness with which vision can be used to travel safely and pursue intended activities in the space. Visual accessibility of a space reduces significantly for impaired vision, leading to higher risks of encountering hazardous situations, falling, and being disoriented in navigation. It is

possible to improve the environmental accessibility by enhancing the visibility of informative or hazardous features through cost-effective modifications of contrast, illumination, and materials. However, it is difficult for people with normal vision, even those with rehab expertise, to judge whether specific objects or features would

be visible for individuals with different levels of vision impairment, or to examine the efficacy of modifications. This research aims to develop an objective tool for evaluating the accessibility of indoor spaces to complement the current observational practice in visual environmental evaluation. This tool will be

implemented as a smartphone app for easy use by rehab specialists and the general public. Using computer vision algorithms, RGB camera and LiDAR sensing technology, the tool will 1) visualize an environment for a specified level of vision impairment, 2) flag hazardous features, such as edges of stairs and chairs that are not

visible for this level of vision, and 3) generate visibility metrics that quantify the visibility for a given object to the level of vision. The development and validation of this tool will be conducted in environments with different levels of realism including high-dynamic range images on digital displays, controlled lab spaces simulating real

environments, and real clinical and home environments. The engineering approaches will be established through rigorous sensor testing in real environments. The computational algorithms will be developed using a large database of visibility ratings by low vision participants. The app will be first validated in controlled

laboratory environments and then implemented in complex real environments including eye clinics and homes. This research strives to support the highest level of independence for people with vision impairment before they need to rely on assistive technology or human assistance. This endeavor is supported by a team

with expertise in low vision research and rehabilitation, computer vision and modeling, architecture lighting and design, and assistive technology development. The proposed tool will facilitate the services provided by rehab specialists, caregivers, and facility managers in environmental evaluation and modification which will in turn

enhance safety and independence in people with vision impairment. It will also facilitate telehealth by allowing easy sharing of home evaluation results. A tool that provides quantitative measures will contribute to the consciousness-raising of visual accessibility among the public. The engineering and computational

approaches, once established, can be extended to platforms other than the smartphone models in the current development phase and can generalize to broader environmental contexts.