Abstract
Urbanization in Turkey and in the world continues is increased especially in the last 30 years. Therefore, urban areas having a dynamic structure should be regularly monitored and the urban region should be determined. There is a need of accurate current data that enables to monitor the temporally changing urban development and land use at regular intervals. In order to determine the urban sizes the 2010 DMSP - OLS (US Air Force Meteorological Satellite Program – operational line scan system) night-time images are used. This study is adapted to all provinces of Turkey. Night-time images in the literature are mainly used for examining the global economic and demographic differences between countries. In order to determine the urban growth from night-time images two different image analysis including object-based and cell-based methods are used in this study. Object-oriented image analysis method includes image segmentation studies and cell-based classification method contains natural breaks and otsu method.
TopBackground
Urbanization is quickly increasing in the world and according to data of the United Nations Population Fund it is estimated that the world's population will be 12 billion by 2050, 80% of this population will be living in cities. In our country where an intensely urban migration from villages has been observed, there has been a clear increase in the urban areas. The effects of this increase occurring in urban areas on the social, economic and cultural characteristics of a city and population are inevitable.
Observation and evaluation of these changes occurring in urban areas are crucial for making balanced and sustainable urban development. Today, through the use of remote sensing and information technologies, the assessment process of data is accelerating and the productivity is increasing. In recent years, night-time images have been often used in the studies that determine the urban areas. The fact that these images can be accessed quickly and their assessment process is fast increases the interest in the night images.
The method was used to detect a threshold value for Chicago, Miami and Sacramento states in the USA. Maps of urban areas were created from night-time images in accord with this threshold value (Imhoff et al., 1997).
It has been showed that a single threshold value in night-time images has been more appropriate to determine the urban areas for different metropolitan regions. Therefore, Henderson et al. have indicated that the threshold value of the data in night-time images should be determined separately for each year for which a temporal analysis needs to be carried out for urban growth detection (Henderson et al., 2003).
As there is very little fluctuation in light intensity in Italy's Sicilian city at night, Elvidg et al. have implemented a quadratic regression model to inter-calibrate among years by using the data of the city as the reference data. The purpose of using a single city for calibration in night-time images is to correct the changes according to the years in the satellite and detector sensors (Elvidg et al., 2009).
By examining the global economic and demographic differences of night-time images, Levin and Duke conducted a study that could be used as an indication of the demographic and socio-economic characteristics of the residential area at a local scale. In the study, the population data of the settlements Israel and West Bank (Arab and Jewish) were compared. It has been stated that the economic and geopolitical differences between Israel and the West Bank lead to pattern differences in night-time images (Levin & Duke, 2012).
Key Terms in this Chapter
Object-Based Classification: Object-based image analysis, a technique used to analyze digital imagery, was developed relatively recently compared to traditional pixel-based image analysis. While pixel-based image analysis is based on the information in each pixel, object-based image analysis is based on information from a set of similar pixels called objects or image objects.
Moran's I: In statistics, Moran's I is a measure of spatial autocorrelation developed by Patrick Alfred Pierce Moran. Spatial autocorrelation is characterized by a correlation in a signal among nearby locations in space. Spatial autocorrelation is more complex than one-dimensional autocorrelation because spatial correlation is multi-dimensional and multi-directional.
Correlation: In statistics, dependence is any statistical relationship between two random variables or two sets of data. Correlation refers to any of a broad class of statistical relationships involving dependence, though in common usage it most often refers to the extent to which two variables have a linear relationship with each other.
Histogram: A graph showing the distribution of values in a set of data. Individual values are displayed along a horizontal axis, and the frequency of their occurrence is displayed along a vertical axis.
Pixel-Based Classification: The traditional pixel-based classification process automatically categorizes all the pixels of an image into the thematic classes utilizing only the spectral information (relative reflectance) of each pixel in the image. The spectral information used by each pixel is also termed as a spectral signature, which is estimated by the relative reflectance in different wavelength bands.
Night-Time Images: Time series data of DMSP/OLS stable night-time light imagery spanning years 1992–2012 were derived from NOAA's National Geophysical Data Center. These image products provide gridded cell based annual cloud-free composited stable night-time lights with a digital number (DN) ranged from 0 to 63.
Urban Area: An urban area is a location characterized by high human population density and many built environment features in comparison to the areas surrounding it.
Remote Sensing: Remote sensing is the acquisition of information about an object or phenomenon without making physical contact with the object and thus in contrast to on site observation.