Recognizing signs and doors for Indoor Wayfinding for Blind and Visually Impaired Persons

Indoor signage plays an essential component to find destination for blind and visually impaired people. In this paper, we propose an indoor signage and doors detection system in order to help blind and partially sighted persons accessing unfamiliar indoor environments. Our indoor signage and doors recognizer is builded based on deep learning algorithms. We developed an indoor signage detection system especially used for detecting four types of signage: exit, we, disabled exit and confidence zone. Experiment results demonstrates the effectiveness and the high precision of the proposed recognition system. We obtained 99.8% as a recognition rate.