I installed opencv2.0 and I am using MSV c++ 2008 v9. I tested some code online that return the inverted version of a singel static image, but I would like to see the FaceDetect code working as well. I tried following the instructions in the wiki but I get a memory alocation error coming from a file I am not supposed to edit. I wanted to know if anyone wants to or has tried doing this and could share some insight into how to accomplish the test of this already provided code. My main problem at the moment is that there is a lot of junk out there about working with openCv1.0~ and other versions of MVS, but I want specifics to my version.
Thanks for the help in advance,
Bc
Assuming that you have already installed OpenCV2.0 and compiled it using MSV C++ 2008 successfully. Now to execute already provided code you can either directly execute the facedetect.exe command located at "C:\OpenCV2.0\vs2008\bin\Debug\" assuming you have installed OpenCV2.0 in it's default location.
Syntax : facedetect --cascade="C:\OpenCV2.0\data\haarcascades\haarcascade_frontalface_alt.xml" "C:\OpenCV2.0\samples\c\lena.jpg"
OR
You can execute facedetect.exe from "C:\OpenCV2.0\samples\c\" path. specify the cascade classifier name and image.
Syntax : facedetect --cascade="C:\OpenCV2.0\data\haarcascades\haarcascade_frontalface_alt.xml" lena.jpg
Related
iam actually trying to get opencv running on my Computer. I allready configured the environment variables on Windows and added the Path for the includes and the libraries in a propertiesheet.
Now when i want to add additional dependencies i watched into my directory (opencv\build\x64\vs12\lib)and there are only two files. opencv_world310.lib and opencv_world310d.lib. So they are on my HDD.
I think these are the files i have to add?
I did this and then i got the error LNK1104 could not open "opencv_world310d.lib. Same with opencv_world310.lib.
Iam trying to build a example-code from opencv so there should be everything fine with it. I saw some Tutorials they did not have any issue. But they used OpenCv3.0
Is there any workaround?
kind regards
Yes. By default, opencv_world310.lib is the only file you should link with.
Or if you're building with Debug configuration, you should use opencv_world310d.lib instead.
Since the error code is LNK1104, I believe your additional library path is something wrong.
And then please check you are building x64 code not win32 code.
I have been having this issue with OpenCV 2.4.2 that it doesn't really get installed properly with Microsoft VS 2010 Express; It is always missing a dll file or two doesn't matter what I follow by Googling. Also, Visual Studio 2010 Express is only valid for a few days, so it is acceptable that VS will not behave well with OpenCV at that point. However, when I changed my IDE to CodeBlocks, it is even stranger because now I have missing dll files and also "Missing Entry Point" error because of a duplicate .dll file. I found this guide and followed what it told me to do:
http://conanhung.wordpress.com/2012/05/23/opencv-2-2c-codeblocks-and-mingw-got-it-working-on-windows/
But it still doesn't work!! What I am after is a complete and definitive solution to integrating OpenCV (preferrably 2.4.2 version) on a Windows 7 64-bit machine. I have been turning Stackoverflow inside out but no solution is working for me. The problems are
WHEN USING VISUAL STUDIO:
Missing dll file (e.g. libopencv_core242d.dll, etc.)
WHEN USING CODE BLOCKS:
First it complains that libstdc++-6.dll is missing. When I download it to my C:\MinGW\libexec\gcc\mingw32\4.7.0 folder, it stops complaining about this. But now it complains that there is any entry point problem and some other .dll file is missing!!!
If someone has managed to make it work by hook or crook, I will be more than happy to know what it is, IN DEPTH. I don't believe that such an important library can be so stupidly distributed so that developers have to struggle night and day to make this work.
FYI, I have tried even the OPenCV website and frankly their instructions are a bit useless. Also, all the necessary information e.g. adding PATH, adding includes and libraries, etc. Please someone point me to the right direction as it has been two days and I cannot get anywhere. It is literally stopping my job now :(
As an answer, but probably a very simple one, I did the following and got rid of the problem:
1) Uninstalled code::blocks completely from the PC and downloaded the IDE-only version (i.e. withouth MinGW compiler), then installed it.
2) Downloaded the latest stable version of MinGW, installed, and added C:\MinGW\bin to PATH
3) Followed rest of the instructions from [link] http://conanhung.wordpress.com/2012/05/23/opencv-2-2c-codeblocks-and-mingw-got-it-working-on-windows/
4) Just to be sure, copied the opencv .dll files from the build folder (i.e. the one that you have created "make" and "install" files using CMake) to the debug (i.e. the executables') folder.
This will fix the problems. In case, it doesn't it might be that you need to add some environment variables.
I've looked into it through older questions and by using google, but I can't find an asnwe to my problem.
I have been trying to get devcpp to work with opencv2.1. I configured it with cmake and generated files. Then I compiled it with Mingw32-make command through a windows 7 cmd. Whichever way I tried to link stuff from devcpp, it's never worked. It keeps giving me cv::freecast(void) error.
I have seen some people talking about setting things to the system path. What is that system path exactly ?
Thanks in advance.
You have to tell Windows where to look for the openCV libraries when asked. There are two options:
Not to tell windows where to look for the .dll and put them in the same folder as the project
Tell windows where are the .dll by adding them to the system path
Here in "Compile using Visual Studio" point 3 you can see how to do it.
I'm not sure if this is the right place for asking these kind of question but I have no better place in mind.
After LOTS of coding in OpenCV, I'm trying to migrate to new C++ interface of openCV which has been introduced in OpenCV 2.0. but I keep getting strange errors.
For example:
trying to use pre-built dlls and libs, I got exceptions in "cv::warpAffine" and "findContours" cv::function at runtime. I compiled the openCV, used my own build of libs and dlls and everything got fixed!
the same problem happened in linux.
linking the "Debug" built dlls with "Release" built of program cause some unexpected runtime errors, and vice versa.
and some other errors.
can anyone helping me with what I'm doing wrong? (please keep in mind that I coded alot with older version of openCV).
When I migrated to C++ interface I couldnt make openCV work until I compiled the source by myself. Probably DLL's and Libs that are installed with opencv are good only to a specific machine.
Another thing. Try to do the following if you use Visual Studio
go to Project Properties->Click the C/C++ folder->Code Generation->Runtime Library and change it to /MDd in Debug configuration and /MD in release
It solves a problem with esceptions related to STL
Some of the functions ave moved out to new libs (especially some of the stereo functions are in calib3d) and the header file structure is completely changed in 2.2
There are some guides to the configuration eg for visual studio
I am interested in trying to get a program ported to 64-bit and would like to know if it's even a good candidate for porting. I am a lighting director and have built a SUSE 11.1 Linux box for a program called MagicQ made by Chamsys (http://www.chamsys.be/download.html). I have been working on this for about 6 months now and have all hardware recognised. I am still working on stage visualizers, and I have a separate CPU/board generating the DMX512 code via PoE. I don't think getting it to run in SUSE will be a problem "it was natively built for Ubuntu".
Any help or direction is greatly appreciated!!
Unbuntu and Suse are subtlely different in how things are laid out for file sytems, home directories and such. Usually when you try to install a package on either on you need to use their own package manager programs so that all dependincies are handled and you don't need to manually try to find package 'x' version 'y' and package 'a' version 'b' just to get something working.
If you know that you have all the dependencies covered, and if you have the raw source code, you should be able to just run a compilier against the source code and have it compilied for a 64-bit processor.
Here is a link to the GCC, the GNU Compiler Collection for your reference.
Good luck with your porting project.