A Dual-Field Sensing Scheme for a Guidance System for the Blind
- Authors
- Lin, Qing; Han, Youngjoon
- Issue Date
- May-2016
- Publisher
- MDPI AG
- Keywords
- electronic travel aids; guidance systems for blind people; sensor fusion; scene interpretation; probabilistic graphical model; context awareness
- Citation
- SENSORS, v.16, no.5
- Journal Title
- SENSORS
- Volume
- 16
- Number
- 5
- URI
- http://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/5555
- DOI
- 10.3390/s16050667
- ISSN
- 1424-8220
- Abstract
- An electronic guidance system is very helpful in improving blind people's perceptions in a local environment. In our previous work "Lin, Q.; Han, Y. A Context-Aware-Based Audio Guidance System for Blind People Using a Multimodal Profile Model. Sensors 2014, 14, 18670-18700", a context-aware guidance system using a combination of a laser scanner and a camera was proposed. By using a near-field graphical model, the proposed system could interpret a near-field scene in very high resolution. In this paper, our work is extended by adding a far-field graphical model. The integration of the near-field and the far-field models constitutes a dual-field sensing scheme. In the near-field range, reliable inference of the ground and object status is obtained by fusing range data and image data using the near-field graphical model. In the far-field range, which only the camera can cover, the far-field graphical model is proposed to interpret far-field image data based on appearance and spatial prototypes built using the near-field interpreted data. The dual-field sensing scheme provides a solution for the guidance systems to optimise their scene interpretation capability using simple sensor configurations. Experiments under various local conditions were conducted to show the efficiency of the proposed scheme in improving blind people's perceptions in urban environments.
- Files in This Item
-
Go to Link
- Appears in
Collections - College of Information Technology > Department of Smart Systems Software > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.