Computer vision-based door detection for accessibility of unfamiliar environments to blind persons

Authors:
Yingli Tian;Xiaodong Yang;Aries Arditi
Affiliations:
Department of Electrical Engineering, The City College, City University of New York, New York, NY;Department of Electrical Engineering, The City College, City University of New York, New York, NY;Arlene R Gordon Research Institute, Lighthouse International, New York, NY
Venue:
ICCHP'10 Proceedings of the 12th international conference on Computers helping people with special needs
Year:
2010

Citing 2
Cited 1

A Computational Approach to Edge Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Visual door detection integrating appearance and shape cues

Robotics and Autonomous Systems

RGB-D image-based detection of stairs, pedestrian crosswalks and traffic signs

Journal of Visual Communication and Image Representation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Doors are significant landmarks for indoor wayfinding and navigation to assist blind people accessing unfamiliar environments. Most camera-based door detection algorithms are limited to familiar environments where doors demonstrate known and similar appearance features. In this paper, we present a robust image-based door detection algorithm based on doors' general and stable features (edges and corners) instead of appearance features (color, texture, etc). A generic geometric door model is built to detect doors by combining edges and corners. Furthermore, additional geometric information is employed to distinguish doors from other objects with similar size and shape (e.g. bookshelf, cabinet, etc). The robustness and generalizability of the proposed detection algorithm are evaluated against a challenging database of doors collected from a variety of environments over a wide range of colors, textures, occlusions, illuminations, scale, and views.