A quick guide to setting up and installing OpenCV for using in the Netbeans integrated development environment in Linux.
Step 1: Download and extract OpenCV for Linux
Versions of OpenCV can be downloaded from here:
http://opencv.org/downloads.html
Save it to the location of your choice. Open a command prompt, navigate to the download location and unzip:
Step 2: Create an OpenCV build directory
Navigate to the OpenCV directory, and use
mkdir
to create the build directory:
Step 3: Use CMake to build OpenCV
Navigate to the ‘build’ directory created and run the following CMake command:
cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local ..
And then run
make
command, which can take a while to complete:
Followed by the
sudo make install
command:
Step 4: Create a Netbeans Project
Open the NetBeans IDE and select File > New > Project. Select Empty Project:
Click Next and give the project a name:
And press Finish.
Once the project is created, select File > New > Empty File, to add the main.cpp file to the project. Copy and paste the following code example to main.cpp:
// DetectBlobs.cpp : Defines the entry point for the console application.
//
#include <cv.h>
#include <cxcore.h>
#include <highgui.h>
int main()
{
// Initialise
//std::string filepath = "spots.bmp";
std::string filepath = "spots2.jpg";
int num_blobs = 0;
// Load grayscale version of coloured input image
IplImage* original = cvLoadImage( filepath.c_str() );
IplImage* grayscale = cvLoadImage( filepath.c_str(),
CV_LOAD_IMAGE_GRAYSCALE );
// Check bitmap image exists
assert( grayscale );
// Create IplImage struct for a black and
// white (binary) image
IplImage* img_bw = cvCreateImage( cvGetSize( grayscale ),
IPL_DEPTH_8U,
1 );
// Use thresholding to convert grayscale image
// into binary
cvThreshold( grayscale, // source image
img_bw, // destination image
40, // threhold val.
255, // max. val
CV_THRESH_BINARY ); // binary
// Create IplImage struct for inverted black
// and white image
IplImage* img_bw_inv = cvCloneImage( img_bw );
IplImage* img_bw_cpy = cvCloneImage( img_bw );
// Find connected components using OpenCV
CvSeq* seq;
CvMemStorage* storage = cvCreateMemStorage( 0 );
cvClearMemStorage( storage );
// cvFindContours the 12 + 1 extra object for
// white backgrounds and black spots, hence
// subtract 1
num_blobs = cvFindContours( img_bw,
storage,
&seq,
sizeof( CvContour ),
CV_RETR_LIST,
CV_CHAIN_APPROX_NONE,
cvPoint( 0, 0 ) ) - 1;
// Display the input / output windows and images
cvNamedWindow( "original" );
cvShowImage( "original", original );
cvNamedWindow( "grayscale" );
cvShowImage( "grayscale", grayscale );
cvNamedWindow( "black_and_white" );
cvShowImage( "black_and_white",
img_bw_cpy );
// Wait for user key press and then tidy up
cvWaitKey(0);
cvReleaseImage( &original );
cvReleaseImage( &grayscale );
cvReleaseImage( &img_bw );
cvReleaseImage( &img_bw_inv );
cvReleaseImage( &img_bw_cpy );
cvDestroyWindow( "greyscale" );
cvDestroyWindow( "black_and_white" );
cvDestroyWindow( "inverted" );
return 0;
}
Step 5: Configure project settings
Right click the project folder and select Properties.
Select C++ Compiler > Include Directories. Set this to “../../opencv-2.4.9/include/”
Then select Linker > Additional Library Directories to set the location of the OpenCV libraries, in the build/lib folder that was created using CMake earlier:
Select Linker > Libraries > Add Library. Set the libraries that you will need to include, in this example these are core, highgui and imgproc:
Click Apply and OK. These steps once completed should be sufficient to enable the project to build correctly.
Step 6: Try it!
This is the “spots2.jpg’ image used in this test program (obviously, save the file in a place where your NetBeans project can find it):
And this screenshot shows the results of running the program, giving us the original colour image, the gray-scale equivalent image and the thresholded black and white image (and hence detect the number of spots in the jpg image):
