Title :
Turning unorganized points into contours
Author :
Zhou, Yong ; Toga, Arthur W.
Author_Institution :
Lab. of Neuro Imaging, California Univ., Los Angeles, CA, USA
Abstract :
There are a variety of research applications that require the reconstruction of objects from unorganized points. In our implementation, we accomplish this task in two steps: first by connecting points into contours and then contours into objects. We focus on the first step in this paper. We present a voxel-coding algorithm which assembles unorganized points into contours in a straightforward and efficient way. First, the points are converted into a binary volumetric object using simple 3D voxel-coding starting with voxels that include the sample points. Then contours are interpreted as center-lines of cross-sections. The center-lines are obtained by using a series of 2D voxel-coding operations. Input points are sampled from an unknown object satisfying certain sampling criteria. No additional details about the input data are needed. Output contours pass through or approximate the sample points. The algorithm is tested with several data sets, showing its efficiency
Keywords :
computational geometry; computer graphics; encoding; sampling methods; surface fitting; 2D voxel-coding operations; binary volumetric object; contour connections; cross-section centre-lines; input point sampling; object reconstruction; point connections; sample points; sampling criteria; unknown object; unorganized points; voxel-coding algorithm; Assembly; Image reconstruction; Interpolation; Joining processes; Laboratories; Neuroimaging; Sampling methods; Shape; Testing; Turning;
Conference_Titel :
Computer Graphics and Applications, 2000. Proceedings. The Eighth Pacific Conference on
Conference_Location :
Hong Kong
Print_ISBN :
0-7695-0868-5
DOI :
10.1109/PCCGA.2000.883947