Presentation 2002/1/11
Distance Estimation and Detection of Stairs from An Image
Kenzo YAMANAKA, Haiyuan WU, Tadayoshi SHIOYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a method for estimating the distance and detecting the position of stairs from images. In our method, we detect candidate regions of stairs in an input image with the histogram of edge. And we extract the region of stairs using invariants by calculating geometrical features of stairs. Then, we extract the lines of pillars and steps with the edge information and Hough transformation in the region. We obtain 3-D vector with direction of step edge and a surface normal of floor from these edge lines. We estimate the distance from the blind to the nearest point of stairs based on projective geometry and vector analysis.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Vector analysis / Image sensor / Stairs / Invariants
Paper # 2001-PRMU-210
Date of Issue

Conference Information
Committee PRMU
Conference Date 2002/1/11(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Distance Estimation and Detection of Stairs from An Image
Sub Title (in English)
Keyword(1) Vector analysis
Keyword(2) Image sensor
Keyword(3) Stairs
Keyword(4) Invariants
1st Author's Name Kenzo YAMANAKA
1st Author's Affiliation Graduate School of Science and Technology Kyoto Institute of Technology()
2nd Author's Name Haiyuan WU
2nd Author's Affiliation Graduate School of Science and Technology Kyoto Institute of Technology
3rd Author's Name Tadayoshi SHIOYAMA
3rd Author's Affiliation Graduate School of Science and Technology Kyoto Institute of Technology
Date 2002/1/11
Paper # 2001-PRMU-210
Volume (vol) vol.101
Number (no) 569
Page pp.pp.-
#Pages 8
Date of Issue