Booth Id:
ROBO005
Category:
Robotics and Intelligent Machines
Year:
2019
Finalist Names:
Etteib, Bacem (School: Pioneer Prep School Medenine)
Abstract:
Developing dexterous assistive devices has gained momentum over the last few years and has become one of the most important topics which attracts scientists, business angels and investors.
Notwithstanding the remarkable advances that have been made in developing intelligent and portable devices for visually impaired people, barriers remain and existing solutions still exhibit several problems.
In this context, we propose SmartCap; a wearable device for visually impaired people whose goal is to provide a rich description of the surrounding environment of its users. First, we developed a deep learning model for real-time object detection and localization. Second, we implemented the trained algorithm on a microcontroller to detect objects in an online fashion. Thereafter, a text to speech algorithm was implemented and used to convert detected objects into a synthesized speech in order to help visually impaired people better interact with the surrounding environment.
Overall, promising results were obtained when testing the system on seen and unseen objects. Hence, the developed system could chart a route ahead for developing a new generation of dexterous and low cost assistive devices for visually impaired people.