jump to navigation

Configuring Dev-Cpp for compiling C programs using the OpenCV May 8, 2012

Posted by viboon in : C , add a comment

Before configuring the Dev-Cpp, we need:
- Dev-Cpp
- MinGW
- CMake
- OpenCV

1. Download and extract or install Dev-Cpp, MinGW, CMake and OpenCV; also add C:\Dev-Cpp\bin\;C:\MinGW\bin\ into the system path.

2. Use CMake to build OpenCV binaries from the source code.

In CMake, fill in the source code and the binary generated directories.

Where is the source code: C:/OpenCVSource
Where to build the binaries: C:/OpenCV

Click “Configure” and select “MinGW Makefiles” as the generator.

Select “Specify native compilers”.

The C and C++ compilers are specified as follows:

C: C:/MinGW/bin/gcc.exe
C++: C:/MinGW/bin/g++.exe

Click “Finish”.

In the configuration screen type in “RELEASE” (or “DEBUG” if you want to build a debug version) for “CMAKE_BUILD_TYPE”.

Click “Configure” and “Generate”, respectively.

Go to the command prompt and inside the directory “C:\OpenCV” type:



mingw32-make install

3. Configure the Dev-Cpp so it knows where to find OpenCV for your projects.

Go to “Tools -> Compiler Options”

Click “+” sign to add a new compiler named here, OpenCV.

On the section “Add these commands to the linker command line”
-lopencv_core231 -lopencv_highgui231 -lopencv_imgproc231 -lopencv_features2d231 -lopencv_calib3d231

In “Directories” tap, add the following into “Binaries”, “Libraries”, “C Includes” and “C++ Includes”.
*** Binaries ***
*** Libraries ***
*** C Includes ***
*** C++ Includes ***

- http://opencv.willowgarage.com/wiki/InstallGuide
- http://opencv.willowgarage.com/wiki/MinGW
- http://opencv.willowgarage.com/wiki/DevCpp

Tags: , , , , ,

Related posts:

A very brief introduction to C implementation in Matlab August 23, 2009

Posted by viboon in : C, Matlab , add a comment

1. Create a file in C format which a heading file of mex.h has to be added in order to interface with Matlab. Instead of staring from main() function, we use mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[]) to acquire and return the variables between C function and Matlab.

#include "math.h"
#include "mex.h"
void mexFunction(int nlhs, mxArray *plhs[], int nrhs,
const mxArray *prhs[])
    //Declare the input variable in pointer format
    mxArray *xInput;
    double *xMat, *xAns;
    int row, col;
    //Copy to the pointer
    xInput = prhs[0];
    //Get the matrix
    xMat = mxGetPr(xInput);
    row = mxGetM(xInput);
    col = mxGetN(xInput);
    //Allocate the memory for output value
    plhs[0] = mxCreateDoubleMatrix(row, col, mxREAL);
    //Apply the pointer to the reserved memory
    xAns = mxGetPr(plhs[0]);
    //Start your coding here

2. Compile the C file by using a command below:

mex mycfun.c

3. Call your compiled C function in the M-file.

[xAns] = mycfun(xInput);

4. Enjoy!

Tags: , , , ,

Related posts:

Writing C functions in Matlab August 17, 2009

Posted by viboon in : C, Matlab , add a comment

I face a long computing time problem when I work on Matlab to compute the large and complicated matrices with many loops inside. To reduce the calculating time particularly the loops, the external function written in C language is an effective method to be applied in the MEX-file format. By implementing C function, the computing time can be faster significantly.

Good how-to sources:
- MathWorks: Calling C and Fortran Programs from MATLAB Command Line
- Writing C Functions in MATLAB (MEX-Files)

Tags: , , , ,

Related posts:

How to compile C/C++ using GNU compiler in Linux July 25, 2009

Posted by viboon in : C, Ubuntu , add a comment

To display the installed GNU compiler version

$ gcc -v

Compile the code

$ gcc test.c

Using this command the compiler will make an output file named “a.out”. In order to change it to the different name, an additional argument has to be added.

$ gcc -o outputoftest test.c

To run the compiled program when the current directory is not included in the environmental variable.

$ ./outputoftest

Tags: , , , , ,

Related posts: