I really didn't intend to move it off list. Not sure what happened there. I'm gonna stick with OCC for now, as I am fairly familiar with it.
Mark On Wed, Mar 4, 2009 at 2:38 AM, EBo <[email protected]> wrote: > Mark, > > I took a quick look at gdepth... It looks like extending that would probably > be the best way to go. I would suggest modifying a couple of the routines to > generate cut tool profiles from optical comparator images or some equivalent. > If my quick scan of the code read it correctly, this would allow you to load > a tool and either spin the tool or model it's geometry in a lathe type > operation. > > The stuff I was rattling on before would be good background info to know to > understand what is going on with all the transformations and flipping > projections, but a good bit of the basics are already provided with gdepth. > > Best of luck and look forward to the correspondence. > > EBo -- > > > EBo <[email protected]> said: > >> Mark, >> >> I see you moved this off list. Ok... >> >> YoW!. Ok... You can model the problem as moving the tool around the pice >> (ie >> milling) or spinning the piece around the tool (lathe). I can point you to >> the references to get started understanding what is going on with the >> projective geometry, etc., and give you a VERY old class project where I got >> all the basics working. The code will probably not compiler as it depends on >> stuff I had access to a decade ago, but all the basic math stuff is there for >> projecting between word, object coordinates, etc. The code is also likely a >> mess, but if it would help you then so be it. I will warn you though that >> going about it this way is many months of effort (depending on where you are >> with understanding 4D image transformations). My best suggestion is to take >> in addition to working on this a senior/graduate class in computer graphics. >> Some instructors will let you sit in so that you can wrap your head around >> the >> problems. Basically what you have set for yourself is the equivalent of a >> challenging undergraduate senior honors thesis or possibly even a masters > thesis. >> >> If you are really up for hurting yourself that bad I'll dig up the references >> in my library (and you are lucky I have it with me -- I'm 1000 miles from >> home >> at the moment), and I'll send you some old code. I can also suggest some >> interesting computational geometry packages which can do the mathematical >> morphology (ummm... think you have the tool sweep and you want to subtract >> that from the original object...). >> >> To get you started, take a look at the following: >> >> VTK: http://vtk.org/ >> GTS: http://gts.sourceforge.net/ >> >> I would strongly encourage you to use VTK (althought the learning curve is >> painful), but once you start wrapping your head around that as a tool you can >> run it on *nix, win*, and mac*. Also, it does more than you will ever want >> to, and is freely distributed... >> >> Anyway, if you really want to go there I'll dig suff up, but please know that >> I am really swamped and will be handing this out in dribs and drabs. Just >> give me a poke every now and again and I'll get stuff to you or point you >> down >> some other road... >> >> >> EBo -- >> >> Mark <[email protected]> said: >> >> > On Wed, Mar 4, 2009 at 12:12 AM, EBo <[email protected]> wrote: >> > > >> > > So are you wanting to do something like engrave something onto a > cylindrical? >> > > >> > > The times that I have had to do something like this it was in an advanced >> > > computer graphics class -- jumping between world and other projections >> > > and >> > > moving the view plane around graphics objects, but I doubt telling you > how to >> > > do that would be of any help. So, can you describe what you want to do a >> > > little more? >> > > >> > > EBo -- >> > >> > >> > Yes, it is computer graphics. >> > >> > I talked about what I'm doing in my second response to Alexy - >> > something similar to gdepth ( http://axis.unpy.net/01169521961 ) - >> > create an accurate solid model of the material that will be removed >> > (by sweeping the tool's cross-section along a line), then subtract >> > that from a block, creating an accurate model of the material that is >> > left. >> > >> > Hopefully this result can be compared with the solid model that was >> > used as input to a CAM program (HeeksCNC, or cam-occ IF I ever get >> > anywhere with it). >> > >> > I might just abandon cam-occ in favor of working on HeeksCNC, since >> > Dan is moving so quickly, but for right now I am still messing with >> > it. >> > >> > I don't know if it's your email or mine, but something keeps messing >> > with the replies. I see a bunch of HTML. I clicked "plain text" for >> > this one, maybe it will behave from now on. >> > Mark >> > >> > >> >> >> >> -- >> >> >> >> > > > > -- > > > > ------------------------------------------------------------------------------ Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise -Strategies to boost innovation and cut costs with open source participation -Receive a $600 discount off the registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
