Page 195 - Contributed Paper Session (CPS) - Volume 1

CPS1277 Amal E.S

Caio & New Cairo Caio & New Cairo

2017 before 2017 after
classification classification

2.4.5 Classification accuracy assessment
In the context of information extraction by image analysis, accuracy
“measures the agreement between a standard assumed to be correct and a
classified image of unknown quality.” (Campbell, 2007). Accuracy assessment
is an important part of any classification project. It compares the classified
image to ground truth data which are derived from interpreting high-
resolution imagery. The most common way to assess the accuracy of a
classified map is to create a set of random points from the ground truth data
and compare that to the classified data in a confusion matrix.
The following diagram shows the steps of accuracy assessment process.

And below is the confusion matrix of one of the cities in the study, this
matrix shows that the accuracy rates range from 96.4 to 100 percent accuracy
for each class.

Class N.CAIRO_Urban N.CAIRO_GreenArea N.CAIRO_Desert N.CAIRO_Water
Urban 100 0 3.571428571 0
Green 0 100 0 0
Area
Desert 0 0 96.42857143 0

Water 0 0 0 100
184 | I S I W S C 2 0 1 9