Detecting and Monitoring the Vegetal Cover of Karbala Province (Iraq) Using Change Detection Methods
Keywords:
Maximum likelihood, support vector machine classifier, matching filterAbstract
Karbala province was one of the most important areas in Iraq and considered an
economic resource of vegetation such as trees of fruits, sieve and other vegetation.
This research aimed to utilize change detection for investigating the current
vegetation cover at last three decay. The main objectives of this research are collect
a group of studied area (Karbala province) satellite images in sequence time for
the same area, these image captured by Landsat (TM 1995, ETM+ 2005 and
Landsat 8 OLI (Operational Land Imager) 2015. Preprocessing such as atmosphere
correction and rectification has been done. Mosaic model between the parts of
studied area was performing. Gap filling consider being very important step has
been implied on the defected image which captured in Landsat 2005. For monitoring
the changes in the studied area, many image processing such as supervised
classification using Maximum likelihood classifier and support vector machine
classifier have been applied. Target detection using matching filter and change
detection using subtractive method also have been used to detect the change in
vegetal cover of the studied area. Many histogram and statistical properties were
illustrated as well as the pixel count and the target area has been computed.