Hello,Yan Zhang. when I used compile_mex_codes to compile, I met the following errors:
C:\DOCUME~1\yanjpphd\LOCALS~1\Temp\mex_933260AC-1398-46B6-6CA4-9F6BBF3212F4\ac_tridiagonal_Thomas_dll.obj
LINK : fatal error LNK1104: cannot open file "ac_tridiagonal_Thomas_dll.mexw32"
C:\PROGRA~1\MATLAB\R2007B\BIN\MEX.PL: Error: Link of 'ac_tridiagonal_Thomas_dll.mexw32' failed.
??? Error using ==> mex at 208
Unable to complete successfully.
Error in ==> compile_mex_codes at 1
mex ac_tridiagonal_Thomas_dll.cpp
Yan:
I hope you can help with a compilation issue.
I am running (64bit) ubuntu 10.04 (vm on fusion 4.3 on a newish MacBook Pro), and I am having a problem compiling:
Upon typing "compile_mex_codes", I get:
mex: ac_tridiagonal_Thomas_dll.cpp not a normal file or does not exist.
Error using mex (line 206)
Unable to complete successfully.
Error in compile_mex_codes (line 1)
mex ac_tridiagonal_Thomas_dll.cpp
Note that I ran "mex -setup", though I have compiled mex files successfully before). It looks like (assuming is see my vm as a "glnxa64"). I have g++ selected, as you indicated in earlier post(s). I am not sure about the flags . . .
I am using "gcc/g++/gfortran" v. 4.4.3, while mex is apparently supported with 4.3.4, but I had no problems in previous mex compilations.(I am fairly new to Matlab, let alone mex files.)
Hi, Yan
Thanks very much for sharing.
I want to compile the code on Windows. However, it seems like the link given in "readme.txt" --http://ecson.org/resources/active_contour_segmentation.html doesn't work any more.
Could you tell me how to access the version for WINDOWS? Thanks.
Hi Yan, I tried to run the compile_mex_codes.m in matlab and I got this errors:
??? Error using ==> mex at 222
Unable to complete successfully.
Error in ==> compile_mex_codes at 1
mex ac_tridiagonal_Thomas_dll.cpp
Could you help me with this? Thanks.
Yan,
I am trying to use the ChanVese model to pick out a 3D image of data that does not have a sharp edge. In fact, the brightness value per 3D pixel just gradually increases toward the center of the image. ChanVese is supposed to be able to do this per the reference, but I have not had success. Any advice?
I have been trying to relate the smooth_weight and image_weight parameters to the parameters in the reference (u,v,lambda1,lambda2). The image_weight seems to correspond to the lamdas (where lamda1 = lamda2). In the reference by Chan & Vese typically a value of one is used, but matlab crashes with a standard exception error when I put in a value any greater than 0.01.
Does the smooth_weight relate to u an v somehow?
Thanks for any advice you can offer.
Brent
Comment only