Page 194 - Contributed Paper Session (CPS) - Volume 1

CPS1277 Amal E.S
In the study I used supervised classification which is based on the training
samples collected directly on the image, and then the software classifies the
rest of the pixels in the image according to theses samples.

2.4.1 Collecting training samples
In supervised classification, training samples are created to identify classes
by draw on the image for each class and calculate their signatures as shown
in figure 3.

2.4.2 Evaluating training samples
Examine the Histograms for each class on all the bands to make sure that
the classes represented by the training samples are distinguishable and not
overlap as shown in figure 4.

2.4.3 Creating the signature file
After determine the training sample which represent each class and
examine the Histogram, a signature file was created to classify the image.

2.4.4 Applying classification
The Maximum Likelihood Classification method used to classify the image;
it assigns each pixel to one of the different classes based on the means and
variances of the class signatures (stored in a signature file).

183 | I S I W S C 2 0 1 9