Integration of overlapping maps made by stereo vision using view field information

We propose a new method for combining two overlapping maps made by the stereo vision technique utilizing view field information. Degree of matching is evaluated for every matching candidate, where object point pairs and solitary points in the overlapping view fields contribute positive and negative...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on systems, man, and cybernetics man, and cybernetics, 1994-08, Vol.24 (8), p.1273-1279
Main Authors: Tateda, H., Naruse, H., Ide, A., Nobiki, A.
Format: Article
Language:English
Subjects:
Applied sciences
Area measurement
Artificial intelligence
Cameras
Computer science
control theory
systems
Costs
Exact sciences and technology
Failure analysis
Humans
Image analysis
Image recognition
Measurement standards
Pattern matching
Pattern recognition. Digital image processing. Computational geometry
Stereo vision
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose a new method for combining two overlapping maps made by the stereo vision technique utilizing view field information. Degree of matching is evaluated for every matching candidate, where object point pairs and solitary points in the overlapping view fields contribute positive and negative values, respectively. The threshold value used in judging pair is determined by considering the error magnitude, which depends on the distance from the view point. Various kinds of calculation are compared in terms of the freedom of threshold, which is defined as the ratio of the maximum and minimum threshold values, and with which the correct overlap is derived.< >
ISSN:0018-9472
2168-2909
DOI:10.1109/21.299709