Please use this identifier to cite or link to this item: http://hdl.handle.net/2440/60144
Type: Thesis
Title: Surface reconstruction from three dimensional range data.
Author: Myers, Andrew
Issue Date: 2005
School/Discipline: School of Computer Science
Abstract: This thesis looks at the problem of reconstructing a single surface representation from multiple range images acquired from a terrestrial laser scanner. A solution to this problem is important to industries such as mining, where accurate spatial measurement is required for mapping and volumetric calculations. Laser scanners for 3D measurement are now commercially available and software for deriving useful information from the data these devices generate is essential. A reconstruction technique based on an implicit surface representation of the range images and a polygonisation algorithm called marching triangles has been implemented in software and its performance investigated. This work improves upon the existing techniques in that it takes into account the particular differences of terrestrial range data as compared with data from small scale laser scanners. The implementation is robust with respect to noisy data and environments and requires minimal user input. A new approach to 3D spatial indexing is also developed to allow rapid evaluation of the true closest point to a surface which is the basis of the signed distance function implicit surface representation. A new technique for locating step discontinuities in the range image is presented, which caters for the varying sampling densities of terrestrial range images. The algorithm is demonstrated using representative range images acquired for surface erosion monitoring and for underground mine surveying. The results indicate that this reconstruction technique represents an improvement over current techniques for this type of range data.
Advisor: Brooks, Michael John
Dissertation Note: Thesis (Ph.D.) -- University of Adelaide, School of Computer Science, 2005
Keywords: spatial indexing, marching triangles algorithm
Provenance: Copyright material removed from digital thesis. See print copy in University of Adelaide Library for full text.
Appears in Collections:Research Theses

Files in This Item:
File Description SizeFormat 
01front.pdfFront matter71.74 kBAdobe PDFView/Open
02chapters1-5.pdfChapters 1-52.38 MBAdobe PDFView/Open
03chapters6-10.pdfChapters 6-102.62 MBAdobe PDFView/Open
04append_refs.pdfAppendices-References94.54 kBAdobe PDFView/Open
Supplement.zipSupplement information336.04 MBUnknownView/Open


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