Abstract | ||
---|---|---|
There is an increasing need to develop new adaptive technologies and new wayfinding assistance systems for blind and visually impaired persons in order to improve their daily lives. To address this need, we propose in this paper to develop a new deep learning-based indoor wayfinding assistance system consisting of detecting landmark indoor signs. Assistive technologies used for blind and sighted persons used to support daily activities to improve social inclusion are developing very fast. Training and testing experiments were performed on the proposed indoor signage dataset. Through the experiments conducted, we demonstrated the efficiency of the proposed indoor wayfinding aid system. We obtained 93.45% as a mean average precision (mAP) of the proposed indoor wayfinding and signage detection system. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1007/s11042-021-10999-6 | MULTIMEDIA TOOLS AND APPLICATIONS |
Keywords | DocType | Volume |
Indoor sign detection, Indoor wayfinding, Blind and visually impaired persons, Deep learning, Convolution neural network (CNN) | Journal | 80 |
Issue | ISSN | Citations |
18 | 1380-7501 | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mouna Afif | 1 | 9 | 3.28 |
Riadh Ayachi | 2 | 1 | 3.06 |
Yahia Said | 3 | 21 | 7.09 |
Mohamed Atri | 4 | 154 | 27.75 |