OpenCV Tutorial

Part II
Loading Images and Using Histograms

29 November 2005

Gavin S Page [email protected]

du
 Tasks

The first step after establishing

a working environment is to
begin manipulating images.
This tutorial will give an Steps Performed
introduction to the usage of Load an Image
some basic functions. Calculate Histogram Values
Calculate Basic Statistics Using
Histogram Information

For explanations on any

functions used here see the
OpenCV documentat.

29 November 2005 Gavin S Page [email protected] 2

du
 Loading the Image
OpenCV makes it relatively
easy to load images. There are
several syntax variations that //the name of the image being loaded
simply take in the path/file char* imageName = "Critters_00005.JPG";
//Load the image and make sure that it loads correctly
name. One is presented here. IplImage* im = cvLoadImage(imageName, -1);
if( im == 0 ) {
//Drop out if the image isn't found
std::cerr << "Failed to load: " << imageName << std::endl;
return 1;
}
Load the File
Specify a File to be Loaded

Use the cvLoadImage function to

assign the image to an IplImage pointer
OpenCV uses an IplImage to
represent image internally.

29 November 2005 Gavin S Page [email protected] 3

du
 Specifying a Working Region
In order to work with a
histogram the image will have //Create a single planed image of the same size as the original
to converted to a single plane. IplImage* grayImage = cvCreateImage(cvSize(im->width,im->height),
IPL_DEPTH_8U, 1);

//convert the original image to gray

cvCvtColor(im, grayImage, CV_BGR2GRAY);
//create a rectangular area to evaluate
Create the Grayscale Image CvRect rect = cvRect(0, 0, 500, 600 );
//apply the rectangle to the image and establish a region of interest
cvSetImageROI(grayImage, rect);
Create an Image of a Single Plane

The cvCvtColor function can

Convert the Image to Gray be used to convert images to
one of several color spaces.
Specify a Rectangular Region of
Interest (ROI) and apply it to the image
To restore the region of
interest to the whole image the
function cvResetImageROI is
used
29 November 2005 Gavin S Page [email protected] 4
du
 Perform Initial Histogram
Calculations
OpenCV provides built-in
//create an image to hold the histogram
functions to work with IplImage* histImage = cvCreateImage(cvSize(320,200), 8, 1);
//create a histogram to store the information from the image
histograms. CvHistogram* hist =
cvCreateHist(1, &hist_size, CV_HIST_ARRAY, ranges, 1);
//calculate the histogram and apply to hist
cvCalcHist( &grayImage, hist, 0, NULL );

Create the Histogram Data //grab the min and max values and their indeces
cvGetMinMaxHistValue( hist, &min_value, &max_value, &min_idx, &max_idx);
Create a Histogram Image and a //scale the bin values so that they will fit in the image representation
cvScale( hist->bins, hist->bins, ((double)histImage->height)/max_value, 0 );
Histogram
//set all histogram values to 255
cvSet( histImage, cvScalarAll(255), 0 );
//create a factor for scaling along the width
bin_w = cvRound((double)histImage->width/hist_size);
Calculate the Histogram

Grab Min/Max Values

Set Up Factors For Visualization

29 November 2005 Gavin S Page [email protected] 5

du
 Prepare Visualization/Perform
Calculations for( i = 0; i < hist_size; i++ ) {
Here we will iterate across the //draw the histogram data onto the histogram image
cvRectangle( histImage, cvPoint(i*bin_w, histImage->height),
histogram bins and apply the cvPoint((i+1)*bin_w,
values to the image while histImage->height - cvRound(cvGetReal1D(hist->bins,i))),
cvScalarAll(0), -1, 8, 0 );
calculating the statistics. //get the value at the current histogram bucket
float* bins = cvGetHistValue_1D(hist,i);
//increment the mean value
mean += bins[0];
Draw Values on Image }
//finish mean calculation
mean /= hist_size;
Use cvRectangle to draw. //go back through now that mean has been calculated in order to calculate variance
for( i = 0; i < hist_size; i++ ) {
float* bins = cvGetHistValue_1D(hist,i);
variance += pow((bins[0] - mean),2);
}
Get Values/Perform Calculations //finish variance calculation
variance /= hist_size;

29 November 2005 Gavin S Page [email protected] 6

du
 Display Results
std::cout << "Histogram Mean: " << mean << std::endl;
std::cout << "Variance: " << variance << std::endl;
This segment displays the std::cout << "Standard Deviation: " << sqrt(variance) << std::endl;

visual and textural results.

//display the 3 images
cvNamedWindow("Original", 0);
cvShowImage("Original", im );
Display
cvNamedWindow("Gray", 0);
Output Statistics cvShowImage("Gray", grayImage );

cvNamedWindow("Histogram", 0);
cvShowImage("Histogram", histImage );
Show Images //hold the images until a key is pressed
cvWaitKey(0);
cvNamedWindow creates a container.
The first parameter is the name and the
second declares if the container is to
expand to fit the contents.

Hold For Input. Passing the parameter

“0” waits for a keypress.

29 November 2005 Gavin S Page [email protected] 7

du
 Cleaning Up
//clean up images
cvReleaseImage(&histImage);
Release Images cvReleaseImage(&grayImage);
cvReleaseImage(&im);

//remove windows
cvDestroyWindow("Original");
Destroy Containers cvDestroyWindow("Gray");
cvDestroyWindow("Histogram");

It is very important to perform

clean-up functions. It is easy
for memory utilization to go out
of control when multiple
images are involved.

29 November 2005 Gavin S Page [email protected] 8

du
 Results

Here are the original image,

the grayscale region, and the
histogram of that region.

29 November 2005 Gavin S Page [email protected] 9

du
 Other Histogram Functions
OpenCV has several other
functions for working with
histograms. These include:
• cvNormalizeHist
• cvThreshHist
•cvCompareHist
For more information about
usage of these functions see
the OpenCV documentation

29 November 2005 Gavin S Page [email protected] 10

du
 Revision History

Initial Creation: 28 November 2005

29 November 2005 Gavin S Page [email protected] 11

du