CGAL users discussion list

[t doehring <>]

Hi,
well i have a couple ideas. first i DO think you can manage point clouds in cgal based upon my readings, but i'm not an expert at that...
here's another idea: if you have a point cloud, can you convert it to a voxel image? the idea is to generate a 3D logical matrix where you have 'ones' for each 'point' in the cloud, and zeros elsewhere (easy in matlab). then you can use any of a number of morphological operations (such as dilate)... with blurring, and/or (even more powerful) level-set methods to 'fill-in' the gaps in the cloud to create your solid object. Once you have the solid 'filled in' object, then you can isosurface in matlab, or use the image-based 3-D mesh generation methods in CGAL to create a surface or tet-mesh.
hope that helps!
best,
todd

-- 
oo-o--o---o----o-----o------o-------o------o-----o----o---o--o-oo
Todd C. Doehring, Ph.D.
Assistant Professor
School of Biomedical Engineering
Drexel University
oo-o--o---o----o-----o------o-------o------o-----o----o---o--o-oo

Sheikh Faridul Hasan wrote:

2010/3/19 Pavel Nyaga <>

Dear Experts,

 

I need to calculate volume of 3D scattered points cloud with  huge concave features and varying points density.

 

First of all I tried what is available in MATLAB:

1)      Volume inside of iso-surface of density function -  calculated as number of points per cubic element volume on rectangular grid;

2)      Volume inside of iso-surface of density function  - calculated as point per volume of Voronoi Region and interpolated on rectangular grid;

 

As above methods volume value depends on iso-surface value  and cell size, I decided to look at surface reconstruction from point clouds:

1)      ConvexHull  - overestimates the volume very much;

2)      Power Crust – seems not to be working on varying density clouds as many other techniques J

 

So now my questions are:

1)      Does any algorithm exist which will give me robust volume of the concave cloud without high sensitivity to input parameters?

a.       if yes, is there anything in CGAL which can help me?

                                                              I.      If no, could you recommend me any commercial code?

2)      Even in case I have triangulation of concave closed surface, how can I calculate volume inside? Is there constrained 3D Delaney tessellation in CGAL?

----------------------------------------------------------------------------- 

>>  constrained Delaunay Tetrahedralization does not always exists ; You can find some more info here [http://www.cs.cmu.edu/~jrs/jrspapers.html]

----------------------------------------------------------------------------- 

Thank you in advance.

 

With best regards,

 

Pavel Nyaga

Modeling Engineer