Land use and land cover (LULC) classification based on remote sensing images is one of the major research themes in terms of global environmental change. LULC information is essential for a variety of geospatial applications, such as urban planning, monitoring environmental changes, regional administration, and environmental management. Changes in LULC  is a significant contributor to a range of environmental problems such as deforestation, biodiversity loss, global warming, landslide and increase of natural disaster-flooding, etc. Therefore, available data on LULC changes can provide critical input to the decision-making of environmental management and planning the future. With the rapid development of remote sensing technology, LULC classification using remotely sensed image has become the most credible, fast and effective measure to monitor the condition and changing of LULC. Remotely sensed images help in tracking the land cover change by providing the repeatable observations of the same area and also of the areas that are inaccessible. 

Machines can do complex computations in short amounts of time by performing human-related tasks in an automated fashion. A combination of complex algorithms and massive computing power helps to create artificial neural networks that can recognize patterns in digital representations of sounds and images. These algorithms demonstrate high potential in identifying and characterizing LC and LU patterns from satellite images by observing the common patterns and details. Currently, lots of studies performed on LULC classification using machine learning algorithms, but few studies utilize environmental aspects in their studies. In this study, in addition to remote sensing data, environmental variables will be introduced to classification algorithms as inputs. 

By: Özge Yücel

Advisor: Prof.Dr.Ayşegül Aksoy

Date: 08.05.2019