the main objective of this research is to evaluate change detection techniques to monitoring land-cover
changes that occurred between 2016 and 2017 in the study area located in new administrative capital
region in Cairo Governorate, Egypt. The Study area is 77.350 km
2
Keywords:
Change detection
Post classification
Independent component analysis
Principal component analysis
. Image data is two Satellite images covering
the study area, both are by Landsat_8 OLI/TIRS satellite with a low spatial resolution 30 m * 30 m.
Reference data is two Satellite images covering the study area, both are by Nanosat satellite with a high
spatial resolution 3 m * 3 m. The Study area was classified into roads, sand, rocks, bare soil, routes and
buildings using Maximum Likelihood Classifier. Three change detection techniques namely; postclassification,
independent component analysis and principal component analysis were applied.
Quantitative evaluations for the results of these techniques were performed to determine the most
appropriate change detection technique which will provide the highest accuracy for identifying the nature
and extent of land-cover changes in New capital. The results indicated that the post classification
change detection technique provided the highest accuracy while the principal component analysis technique
gave the least accuracy. And the results show that the post classification change detection technique
is straightforward because it is direct and its accuracy is mainly dependent upon the accuracy of
the initial classifications of the two images. In the other two change detection techniques, more processing
stages are required such as transformation stage
|
