作者: A. Aladren , G. Lopez-Nicolas , Luis Puig , Josechu J. Guerrero
DOI: 10.1109/JSYST.2014.2320639
关键词: Ranging 、 Image segmentation 、 Image processing 、 Segmentation 、 Visualization 、 Artificial intelligence 、 RGB color model 、 Robustness (computer science) 、 Search engine 、 Computer vision 、 Engineering
摘要: Navigation assistance for visually impaired (NAVI) refers to systems that are able assist or guide people with vision loss, ranging from partially sighted totally blind, by means of sound commands. In this paper, a new system NAVI is presented based on visual and range information. Instead using several sensors, we choose one device, consumer RGB-D camera, take advantage both particular, the main contribution combination depth information image intensities, resulting in robust expansion range-based floor segmentation. On hand, information, which reliable but limited short range, enhanced long-range other difficult prone-to-error processing eased improved The proposed detects classifies structural elements scene providing user obstacle-free paths order navigate safely across unknown scenarios. has been tested wide variety scenarios data sets, giving successful results showing works challenging indoor environments.