Hello Sean! I tried to figure out how to compute the surface area of an object using the gqa's grid shooting method and I referred to your reply to one of my former mails -- you said that each hit point is a surface sample and I can imagine each as a circular patch. As density increases, SA gets refined.
Okay, so I look at the code and I see that each partition has an *in* and an *out* *"hit"* points. So, I would add the surface area given by those 2 for the present partition (SA = the cell area in which these 2 *"hit"* points are found -- I presume both would give the same cell area, as *out *can be assumed to be exactly behind *in*). Now, first of all, the surface can have some curvature and what I would be doing is simply adding the 2-D grid's cell area. Is this fine? (Maybe when gridSpacing is sufficiently refined, the curvature won't matter?) Okay, now the main problem -- what if my ray is parallel to the surface? Can I calculate the SA of that surface? (probably no, right?) I'm kind of stuck on this. Due to these doubts, I'm neither able to fix my termination criteria nor able to decide where to start. And I'm getting a feeling what I have understood is missing something. Please help me with this. With Regards, Kalpit Thakkar On Fri, Jul 31, 2015 at 11:18 AM, Christopher Sean Morrison <brl...@mac.com> wrote: > > Cool, thanks for the status update! Anyone else have a progress update to > share? > > Cheers! > Sean > > On Jul 30, 2015, at 3:50 PM, Kalpit Thakkar <ceasy...@gmail.com> wrote: > > Hello! > So, I have been working on the new API design that I sent in my last mail > in the thread with the Subject "Further in generic volume function for > libanalyze". > > I have successfully implemented (or rather ported from gqa.c) the volume > and centroid analysis for objects which don't have a primitive specific > method for them. > > Surface Area is still pending, but it's on the TODO-next list. I have > tested the code in MGED successfully. > > After completing the volume's and centroid's part of code, I updated the > API design doc[1] with the final design that I have followed and added some > comments about the pipeline with the patch[2] that I submitted. > > Next up is implementing the Caller function for volume and centroid in C++ > interface. > > With Regards, > Kalpit Thakkar > > [1] : > https://docs.google.com/document/d/1WKSH-6MtEAySWZdMa-iy9iVlof1mS4CzA5iehOg5NMw/edit?usp=sharing > [2] : https://sourceforge.net/p/brlcad/patches/385/#e47f > > PS : Please review the patch ASAP. Thank you :) > > > > > ------------------------------------------------------------------------------ > > _______________________________________________ > BRL-CAD Developer mailing list > brlcad-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/brlcad-devel > >
------------------------------------------------------------------------------
_______________________________________________ BRL-CAD Developer mailing list brlcad-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/brlcad-devel
