Development of a visual-to-tactile conversion system for automating tactile graphic generation process

PROJECT SUMMARY/ABSTRACT There are an estimated 23.7 million people who are blind or visually-impaired (BVI) in the U.S. and 285 million globally. Of this population, 30% do not travel independently outside of their home, only ~11% have a bachelor’s degree, and more than 70% are unemployed. The goal of this SBIR effort is to develop a novel system, which performs principled down-sampling and translation of visual information from digital documents into tactile equivalents. Timely access to information is one of the biggest challenges for BVI people. While access to textual information has largely been solved via screen reading software (e.g., JAWS or VoiceOver), very little progress has been made in making graphical information accessible. Although few assistive technology (AT) devices aim provide non-visual graphical access, they suffer from several shortcomings including high cost, limited portability, lack of multi-purpose, and inability to present information in a real-time context. Importantly, a common underlying problem across all extant approaches is that they require intensive human effort for producing or authoring tactile (and/or multimodal) graphics, which leads to high production costs and significant delays in the time between when the accessible materials are needed, and when they are actually delivered, adversely impacting BVI individuals in K-12 schools, colleges, and workplace settings. To address this long-standing problem, UNAR Labs aims to develop a novel system, which will automatically down-sample and translate visual graphical information into an intuitive tactile equivalent that can be used in tactile embossers. Building upon eight years of empirical research, this Phase I SBIR effort will prove the technical feasibility and functional viability of a prototype system for automating visual-to-tactile graphic conversion process and using the output in embossers. Two specific aims will guide this Phase I project: (1) to develop a prototype of an automated system for performing visual-to-tactile conversion without human intervention, and (2) to assess the technical feasibility and functional utility of the system through a rigorous human study. Success in this effort will provide a robust automated system for tactile graphic generation and promote empowerment of millions of BVI individuals by supporting increased educational attainment, proliferation of vocational opportunities, and enhancing overall quality of life for BVI people.PROJECT NARRATIVE Lack of equitable and timely access to information among persons who are blind or visually impaired (BVI) is key to realizing an inclusive world for all as it alleviates a known impediment that is hugely detrimental to their success in activities affecting quality of life and socio-economic status. The proposed innovation presents a first- of-its kind on-demand visual-to-tactile translation system, which will fully automate the tactile graphic generation process using bio-inspired sensory substitution rules and will instantly deliver the translated information for use in tactile embossers. Successful completion of this project will significantly reduce tactile graphic production costs and preparation time, and will promote empowerment of millions of BVI individuals by supporting increased educational attainment, vocational opportunities, and overall better quality of life.