Image Processing - MATLAB Recipes for Earth Sciences - page 354

h e following example for object segmentation illustrates the segmentation,
measuring, and counting of objects using the watershed segmentation
algorithm (Fig. 8.9). We i rst read an image of coarse lithic grains of dif erent
sizes and store it in the variable I1 . h e size of the image is 284-by-367 pixels
and, since the width is 3 cm, the height is 3 cm∙284/367=2.32 cm.
clear
I1 = imread('grainsize.tif');
ix = 3; iy = 284 * 3 / 367;
imshow(I1,'XData',[0 ix],'YData',[0 iy])
title('Original Image')
Here, ix and iy denote the coordinate axes used to calibrate the image I1 to
a centimeter scale. h e true number of objects counted in this image is 236,
including three grains that overlap the borders of the image and will therefore
be ignored in the following exercise. We reject the color information of the
image and convert I1 to grayscale using the function rgb2gray .
I2 = rgb2gray(I1);
imshow(I2,'XData',[0 ix],'YData',[0 iy])
title('Grayscale Image')
h is grayscale image I2 has a relatively low level of contrast. We therefore
use the function imadjust to adjust the image intensity values. h e function
imadjust maps the values in the intensity image I2 to new values in I3 , such
that 1% of the data is saturated at low and high intensities. h is increases the
contrast in the new image I3 .
I3 = imadjust(I2);
imshow(I3,'XData',[0 ix],'YData',[0 iy])
title('Adjusted Intensity Values')
We next determine the background of the I3 image, which means basically
the texture of the black foil on which the grains are located. h e function
imopen(im,se) determines objects in an image im with a specii c shape se
(a l at structuring element such as a circular disk) and size (expressed as a
specii c number of pixels), as dei ned by the function strel . We then produce
a background-free image, I4 .
I4 = imopen(I3,strel('disk',1));
imshow(I4,'XData',[0 ix],'YData',[0 iy])
title('No Background')
We subtract the background-free image I4 from the original grayscale image
I3 to observe the background I5 that has been eliminated.