Article

O'Map: An assistive solution for identifying and localizing objects in a semi-structured environment

A system capable of detection and localization of objects of interest (OI) in a semi-structured environment will enhance the quality of life of people who are blind or visually impaired. Towards building such a system, this paper presents a personalized real-time system called O'Map that runs on Android devices. Main feature of the system is to locate misplaced/moved personal items in a familiar setting. The system provides auditory feedback about the presence and relative position of the query item with respect to known landmarks. First, we adopted participatory design approach to identify users need, functionalities of the system and personalization (i.e., user profile and object map of the environment) in collaboration with representative users. Second, we used the concept from system thinking to develop a real-time object recognition engine that was optimized to run on low form factor devices. Finally, concepts from design thinking were adopted to implement the feedback and user interface to interact with the system. Quantitative evaluation demonstrates that O'Map identifies object of interest with an average F-measure of 0.9650.

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.