Erdvinio objekto paviršiaus atvaizdo rekonstravimas apšviečiant linijiniu šviesos pluoštu

Vilius Matiukas

Doctoral dissertation

Dissertations are not being sold



Technological sciences, electrical and electronic engineering (01T).

This dissertation investigates issues relevant to virtualization of a real 3D object – that is, producing a model of the object from its image data, and then visualizing this model as an image seen on the computer screen

The object of investigation is methods and algorythms for reconstruction of a complex geometric shape from a number of unorganized point sets in to electronic form. Unorganized point sets are obtained by scanning the 3d object from defferent points of view. The choise of the complex oject is made so that it can not be described by a simple mathematical expression.

To attain the aim, the following tasks were put forward: developing a model for a source of linear beam (scanner) and generating an unorganized point set that approximates the object scanned, filtering the unorganized point set and aggregating the unorganized point set to build an entire image of the object and reconstruct its surface.

The aim of this work was to reconstruct the surface image of a 3D object using linear beam. This aim was sought by modifying the existing methods, or proposing new methods, and evaluating the accuracy of the reconstruction using statistical techniques

The work consists of the general characteristic, four chapters, conclusions, list of literature and list of publications. The first section reviews the human visual system, computer vision and three-dimensional imaging technologies. The second section is addressed to a problem of linear beam’s centreline extraction in 2d image data and unorganised point set formation by rotating object around vertical axis as wel as moving plane of the beam. The third section deals with pre-processing and filtering the object’s unorganised point set using octree hierarchial data structure. Finaly the fourth section describes the merging of several unirganised point sets it to one unique representation and object’s 3d surface reconstruction.

The total scope of the dissertation – 94 pages, 33 equations 53 pictures and 6 tables. The dissertation is based on 102 references of other authors.

Main results of the dissertation were published in 7 scientific papers, 4 of which appeared in peer-reviewed scientific journals. Results were also presented in 8 scientific conferences in Lithuania and abroad.

Read electronic version of the book:


Book details

Data sheet

Imprint No:
116 p.
16 other books in the same category:

Follow us on Facebook