Presentation 2002/10/11
Location and Pose Estimation for Active Vision Using Edge Histograms of Omni-directional Images
Caihua WANG, Hideki TANAHASHI, Yutaka SATO, Hidekazu HIRAYU, Yoshinori NIWA, Kazuhiko YAMAMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a new method for location and pose estimation of active vision using histograms of omni-directional images acquired by Stereo Omni-directional System (SOS). In this method, the omni-directional images are acquired at some reference points, and their edge histograms are computed and stored. When the location and pose of SOS are estimated, we first match the edge histograms of the omni-directional images obtained at present location and that obtained at reference points using a two-passes DP method, to find the edge shift in each horizontal angle. The edge shifts of different horizontal angles have the same periodic features as sin curve, where the constant shift item and the 0 phase of sin curve correspond to the relative rotation and movement direction of SOS, respectively. Then, the relative rotation and movement direction of SOS are estimated by independently fitting the constant shift item and the 0 phase of sin curve using the edges shifts. Finally, the location and pose of SOS are computed using the relative rotations and movement directions to two reference points which are closest to the current location of SOS. Experimental results in a real environment show the effectiveness and robustness of the proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) omni-directional image / location and pose estimation / edge histogram / DP matching
Paper # NC2002-62
Date of Issue

Conference Information
Committee NC
Conference Date 2002/10/11(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Location and Pose Estimation for Active Vision Using Edge Histograms of Omni-directional Images
Sub Title (in English)
Keyword(1) omni-directional image
Keyword(2) location and pose estimation
Keyword(3) edge histogram
Keyword(4) DP matching
1st Author's Name Caihua WANG
1st Author's Affiliation HOIP Project, Softopia Japan and JST()
2nd Author's Name Hideki TANAHASHI
2nd Author's Affiliation HOIP Project, Softopia Japan and JST
3rd Author's Name Yutaka SATO
3rd Author's Affiliation HOIP Project, Softopia Japan and JST
4th Author's Name Hidekazu HIRAYU
4th Author's Affiliation HOIP Project, Softopia Japan and JST
5th Author's Name Yoshinori NIWA
5th Author's Affiliation HOIP Project, Softopia Japan and JST
6th Author's Name Kazuhiko YAMAMOTO
6th Author's Affiliation Department of Information Science, Faculty of Engineering, Gifu University
Date 2002/10/11
Paper # NC2002-62
Volume (vol) vol.102
Number (no) 382
Page pp.pp.-
#Pages 6
Date of Issue