College of Engineering Technology – Kirkuk, Northern Technical University discussed a master dissertation on the topographic analysis of the terrain classification of Kirkuk Governorate via using geographic information systems (GIS) by the postgraduate student, Mr. Jalal Abdul Rahman Khider.

The dissertation reviewed a study of terrain classification using modern models and algorithms, the most important of which was the CNN algorithm in order to determine and prepare DEM-based geomorphometric parameters for terrain classification and the development of a machine learning model based on a convolutional neural network for terrain classification.

The dissertation highlighted that the CNN model was the most effective in classifying terrain and achieved an overall accuracy (OA) of 88.91% and a kappa coefficient of 0.883, while SVM was the second most effective model, reaching 79.81% and a kappa coefficient of 0.781, while TPI was the least effective model, reaching 0.781. 67.12% OA and kappa coefficient 0.658.