Home | About IR | Author | Year | Title | My Account | Help | Contact      

EPrints@IIT Delhi >
Faculty Research Publicatons  >
Electrical Engineering >

Please use this identifier to cite or link to this item: http://eprint.iitd.ac.in/handle/2074/2427

Title: Quadrics-based matching technique for 3D object recognition
Authors: Hanmandlu, M
Rangaiah, C
Biswas, K K
Keywords: object recognition
feature matching
tree search
Issue Date: 1992
Citation: Image and Vision Computing, 10(9), 578-588p.
Abstract: We propose a new method for the recognition of objects from range data by matching the features of the observed surfaces in the scene with the features of the model surfaces. The surfaces are represented as quadrics in which planes are treated as a special case. The parameters of quadrics are converted into features which include Euler parameters accounting for the orientation of the surface. The matching process consists of finding a transformation of the model surfaces and devising a measure of consistency, such that the mean square error between the feature space is a minimum. The matching involves a tree search procedure with backtracking whenever error exceeds the limit.
URI: http://eprint.iitd.ac.in/dspace/handle/2074/2427
Appears in Collections:Electrical Engineering

Files in This Item:

File Description SizeFormat
hanmandluqua92.pdf423.63 kBAdobe PDFView/Open
View Statistics

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback