Availability and applicability of image processing techniques in performance analysis of urban infrastructure systems

In this thesis study, image classification techniques and Geographic Information Systems (GIS) were utilized to contribute to the monitoring of flood risk, which places significant pressure on one of the key urban infrastructure systems—stormwater drainage. The study area was selected as the surroundings of the Gökpınar Dam in Denizli Province, and the urbanization process in the region was analyzed using Sentinel-2 satellite images from different years. The satellite images were processed using both supervised and unsupervised classification methods in ArcMap and ENVI software, and land use changes in the region were identified. Within the scope of the thesis, the performance of classification methods in both software programs was compared, and their advantages and disadvantages were evaluated based on criteria such as user experience, processing time, and classification accuracy. Based on the findings, it was concluded that image classification techniques and GIS are effective and applicable methods for monitoring flood risk and supporting urban infrastructure planning processes. In addition, shortcomings in this area were identified, and recommendations for future studies were provided.