Şebnem Düzgün

Unsupervised building detection in complex urban environments from multispectral satellite imagery

A generic algorithm is presented for automatic extraction of buildings and roads from complex urban environments in high-resolution satellite images where the extraction of both object types at the same time enhances the performance. The proposed approach exploits spectral properties in conjunction with spatial properties, both of which actually provide complementary information to each other. First, a high-resolution pansharpened colour image is obtained by merging the high-resolution panchromatic (PAN) and the low-resolution multispectral images yielding a colour image at the resolution of the PAN band. Natural and man-made regions are classified and segmented by the Normalized Difference Vegetation Index (NDVI). Shadow regions are detected by the chromaticity to intensity ratio in the YIQ colour space. After the classification of the vegetation and the shadow areas, the rest of the image consists of man-made areas only. The man-made areas are partitioned by mean shift segmentation where some resulting segments are irrelevant to buildings in terms of shape. These artefacts are eliminated in two steps: First, each segment is thinned using morphological operations and its length is compared to a threshold which is determined according to the empirical length of the buildings. As a result, long segments which most probably represent roads are masked out. Second, the erroneous thin artefacts which are classified by principal component analysis (PCA) are removed. In parallel to PCA, small artefacts are wiped out based on morphological processes as well. The resultant man-made mask image is overlaid on the ground-truth image, where the buildings are previously labelled, for the accuracy assessment of the methodology. The method is applied to Quickbird images (2.4 m multispectral R, G, B, near-infrared (NIR) bands and 0.6 m PAN band) of eight different urban regions, each of which includes different properties of surface objects. The images are extending from simple to complex urban area. The simple image type includes a regular urban area with low density and regular building pattern. The complex image type involves almost all kinds of challenges such as small and large buildings, regions with bare soil, vegetation areas, shadows and so on. Although the performance of the algorithm slightly changes for various urban complexity levels, it performs well for all types of urban areas.

Evaluation and comparison of landslide susceptibility mapping methods: a case study for the Ulus district, Bartın, northern Turkey

The purpose of this study was to investigate the capabilities of different landslide susceptibility methods by comparing their results statistically and spatially to select the best method that portrays the susceptibility zones for the Ulus district of the Bartın province (northern Turkey). Susceptibility maps based on spatial regression (SR), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), logistic regression (LR) method, and artificial neural network method (ANN) were generated, and the effect of each geomorphological parameter was determined. The landslide inventory map digitized from previous studies was used as a base map for landslide occurrence. All of the analyses were implemented with respect to landslides classified as rotational, active, and deeper than 5 m. Three different sets of data were used to produce nine explanatory variables (layers). The study area was divided into grids of 90 m × 90 m, and the ‘seed cell’ technique was applied to obtain statistically balanced population distribution over landslide inventory area. The constructed dataset was divided into two datasets as training and test. The initial assessment consisted of multicollinearity of explanatory variables. Empirical information entropy analysis was implemented to quantify the spatial distribution of the outcomes of these methods. Results of the analyses were validated by using success rate curve (SRC) and prediction rate curve (PRC) methods. Additionally, statistical and spatial comparisons of the results were performed to determine the most suitable susceptibility zonation method in this large-scale study area. In accordance with all these comparisons, it is concluded that ANN was the best method to represent landslide susceptibility throughout the study area with an acceptable processing time.

Landuse change detection in a surface coal mine area using multi-temporal high-resolution satellite images

Surface mining activities, exploitation of ore, and stripping and dumping of the overburden cause changes in the land cover and land use of a mine area. The area of land disturbance can be very large in the case of surface coal mining, due to the nature of the coal extraction process. Sustainable mining requires continuous monitoring of changes in land cover and land use induced by the mining activities. This is essentially important for identifying the long-term impacts of mining on theenvironment and on land cover in order to provide necessary mine closure andreclamation measures. In this sense, digital image classification provides apowerful tool for obtaining rigorous data, and reduces the cost of field measurements in time and money, particularly when dealing with large areas. Various remote sense data records and image classification techniques serve different features for numerous purposes. The selection of a suitable data and image classification method is significant for ensuring the effective use of information extracted from the satellite images, e.g. land-use classes. This paper proposes a methodology for identifying land-use change in surface coal mines using multi-temporal high-resolution satellite images. The methodology has been implemented for identifying, quantifying and analysing the spatial response of landscape to surface mining activities in the Goynuk, Bolu surface mine in Turkey, from 2004 to 2008. The methodology essentially entails (i) acquiring data, (ii) preprocessing the data, (iii)image classification using the maximum likelihood classification algorithm (iv) accuracy assessment and (v) change detection analysis depending on class-based approaches. The results show that the methodology can be utilised effectively in monitoring land-use change in surface coal mining areas. It also provides essential input for planning mine reclamation and closure activities.

Spatial Analyses of Electricity Supply and Consumption in Turkey for Effective Energy Management and Policy-making

Over the past decades, both public and private institutions have extensively analyzed energy supply and demand on national and international bases (WEC-TNC 2004; Munasinghe and Meier 1993). The main focus of these analyses has mainly been the security of the energy supply and the determination of income and price elasticity of energy consumption (Zachariadis and Pashourtidou 2006). In addition, concerns about climate change are considered as another motivation for these analyses (Kuik 2003). This kind of information is useful for making inferences about energy policy implications. Moreover, energy plays a vital role in the economic, social and political development of any nation (Surrey 1996; Varian 2002; ECN 2006). Therefore, no modern society can seriously address development issues if the consideration is not based on a foundation of an adequate, sustainable, and affordable energy supply (Akinbami and Lawal 2009).

Spatial analysis of morphometric indices: the case of Bolu pull-apart basin, western section of North Anatolian Fault System, Turkey

Advances in quality, acquisition and processing of high-resolution digital topography over the past decade have enabled geomorphologists to interpret topography in terms of tectonic processes in a quantitative and more objective way. This study defines the tectonic impact of the North Anatolian fault system (NAFS) on the evaluation and activity of the Bolu pull-apart basin using morphometric indices based on topographic maps. In this study, three different indices including valley floor width to height ratio (Vf), asymmetry factor (Af), and Stream Length Gradient Index (SL) are applied to the tectonic basin area. To detect the deviation from spatial randomness of applied indices, the weighted correlation coefficient Moran’s I is used, and results are interpreted at 99% confidence interval. According to the results, the tectonic impact of NAFS on the northern and western sides of the basin distinctly differs from one another. Clustered distribution of Vf and Af values on the southern side of the basin margin reveals that this side of the basin margin is tectonically more deformed compared with the northern side. The applied methodology reveals that the spatial analysis of the morphometric indices proved to be the effective tool in analyzing tectonic influence of the fault system on the opposite side of the basin margins.

Framework for Systemic Socio-economic Vulnerability and Loss Assessment

A unified approach for modeling shelter needs and health impacts caused by earthquake damage which integrates social vulnerability into the physical systems modeling approaches has been developed. The shelter needs and health impact models discussed here bring together the state-of-the-art casualty and displaced population estimation models into a comprehensive modeling approach based on multi-criteria decision support, which provides decision makers with a dynamic platform to capture post-disaster emergency shelter demand and health impact decisions. The focus in the shelter needs model is to obtain shelter demand as a consequence of building usability, building habitability and social vulnerability of the affected population rather than building damage alone. The shelter model simulates households’ decision-making and considers physical, socio-economic, climatic, spatial and temporal factors in addition to modeled building damage states. The health impact model combines a new semi-empirical methodology for casualty estimation with models of health impact vulnerability, and transportation accessibility to obtain a holistic assessment of health impacts in the emergency period after earthquakes.

System Reliability Investigation of Draglines Using Fault Tree Analysis

Functionality and reliability of a system can be investigated by fault tree analysis (FTA) which is a structured and deductive procedure frequently exploited in complex systems. This analysis technique is related to Boolean algebra and probability theory in capable of qualifying and quantifying the system failure behavior by utilizing basic symbols and probabilistic relations. In this manner, internal and external causes of failures which come out in the working elements of draglines and their occurrence probabilities can be also analyzed by FTA. This research study examines and evaluates the connections between prior and superior failures in dragline components which eventuate in different sub-systems of dragline and their contributions to overall system reliability using FTA. The study utilizes real-time failure data of two different draglines gathered by the technical staff in a coal mine site owned by Turkish Coal Enterprise (TKI).

Risk analysis based on spatial analysis of chronic obstructive pulmonary disease (COPD) with respect to provinces in Turkey

The goal of this study is to analyze and understand the risks of chronic obstructive pulmonary disease (COPD) with respect to the provinces of Turkey according to the results of spatial analysis. The insurance sector needs this kind of analysis to achieve more precise pricing in insurance products. COPD prevalence may exhibit spatial autocorrelation due to the spatial proximity of the provinces. Hence understanding of spatial patterns of COPD prevalence may provide better actuarial decisions. In this research, the common risk factors for COPD are considered to be tobacco sales, air pollution, urbanization, gross schooling rate, life expectancy, median age and GDP per capita for the provinces. The spatial patterns of these factors in Turkey as well as their correlations with COPD prevalence are explored. It has been observed that the parameters show spatial autocorrelation. On the basis of the autocorrelations, financial risk assessment calculations are made with respect to the provinces. Since the analysis could not have been made on the basis of individuals, and financial burdens of morbidities for insurance companies are not given clearly, it is not possible to calculate any health insurance product premium. However, the results obtained provide prior information for the calculations for health insurance products.