Thresholding with an OpenCV Function

Thresholding is very very Important aspect in any field and not just in Open Computer Vision..but in Image processing you need to specify some conditions which is very frequently Thresholding and this is what is the main concern of this post.Here you would get to know about what is Thresholding and also how to use the same and when its required.In almost every program you would specify to extract a pixel above a certain value which is nothing but thresholding …..

Friends I had previously put up a code which spoke about thresholding but without using thresholding functions in OpenCV.Now here is a blog post program which uses cvThreshold to threshold Images have a look..Beside is the picture lena.jpg which is frequently used for image processing functions…_Even i have used this picture in all of my Image processing functions..!

#include "cv.h"
#include "highgui.h"
#include "stdio.h"
#include "conio.h"

int main() {

int height,width,step,step_mono,channels;
/*Here i have declared step_mono
for handling the widthstep member of a monochrome image*/
uchar *data,*data_mono;/*similarly data mono for handling the
 data of monochrome image*/
int i,j,k;

IplImage* frame=cvLoadImage("lena.jpg",1);

if(frame==NULL){ /*If image not found then exit.
This is the function of this loop */
printf("the file doesnot exist");
exit(0);
}
IplImage* mono_thres=cvCreateImage( cvGetSize(frame), 8, 1 );
height = frame -> height;/*height is a member of IPLIMAGE structure
and hence it comes handy like this in such situations,
and same goes with below four statements*/
width = frame -> width;
step = frame -> widthStep;
step_mono = mono_thres -> widthStep;
channels = frame -> nChannels;/*Number of channels in the image*/
data = (uchar *)frame -> imageData;/*Image is treated as as unsigned char data
hence we use an unsigned char pointer to point to the same*/
cvNamedWindow("My Window", CV_WINDOW_AUTOSIZE );
data_mono = (uchar *)mono_thres -> imageData;/*data of mono image is handled by the data_mono*/

for(i=0;i < height;i++) for(j=0;j < width;j++)
/*I am copying the first channel from the image in "frame" in
the monochrome image with the help of this line below..*/
data_mono[i*step_mono+j*1+0]=data[i*step+j*channels+0];

cvThreshold(mono_thres,mono_thres,70,/*70 is the lower cut off*/
150,/*this is the higher cut off*/
CV_THRESH_BINARY/*The type of thresholding,
more description in the documentation*/);

/*Here if the pixel is above the threshold(200) is moved to highest value
and the pixels below the threshold are left as it is*/

cvShowImage("My Window",mono_thres);
cvSaveImage("con_img2.jpg",mono_thres);
/*THis function will save the image */
 /*Destroying the initialized Windows */
cvWaitKey();/*sometimes getch is not compatible with opencv I
 suggest you use cvwaitkey ..this time it worked with me..*/
cvDestroyWindow( "My Window" );
/*as we can see name of the image is followed by the extension..
and the second argument "frame" is the image which will
be saved .This applies similarly in the above statement
 cvShowImage as well.Be careful regarding this.*/

return 0;/*This statement is compiler specific and might change.
As i already said that I am using DEV C++ as my compiler*/
}

Lena

This is the picture of the Thresholded image .See the Result.

Thresholded Result Image