Effect Of Spectral Indices Over Land Use/cover Classification: Istanbul, Beylikduzu District, Land Use Change Summary

Tez (Yüksek Lisans) -- İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, 2015Thesis (M.Sc.) -- İstanbul Technical University, Instıtute of Science and Technology, 2015Günümüzde uzaktan algılama ile yeryüzünün büyük bir kısmı eş zamanlı ve farklı spektral bölgelerde gözlenmektedir ve elde edilen veriler birçok uygulama alanında kullanılmaktadır. En yaygın kullanılan uygulama alanlarından biri de kentsel gelişimin izlenmesidir. Çok zamanlı uydu görüntü verileriyle zaman içerisinde kentsel değişimin izlenebilmesi, kentsel gelişme sonucu arazi örtüsü/kullanımındaki değişimin belirlenmesi mümkün olabilmektedir. Beylikdüzü, 1990’lı yılların sonundan itibaren hızlı bir kentsel gelişim göstermiş, kentin önemli merkezlerinden biri haline gelmiştir. Bu çalışmada, Beylikdüzü ilçesi yerleşim alanlarının ve çevresindeki arazi kullanımlarının uzaktan algılama yöntemi kullanılarak yıllara göre değişimi incelenmiştir. Kullanılan veriler, 12 Haziran 1984 ve 23 Haziran 2011 tarihli LANDSAT 5 TM uydu görüntüleridir. Kullanılan yöntem değişim saptama analizi tekniklerinden görüntü sınıflandırma ile spektral indekslerin birlikte kullanımını içermektedir. Çalışmanın giriş bölümünde uzaktan algılama tanımı ile birlikte çalışma konusu hakkında genel bilgiler verilmiştir. İkinci bölümde ise uzaktan algılamanın temel kavramları olan elektromanyetik enerji, elektromanyetik spektrum, atmosfer ile elektromanyetik enerji arasındaki etkileşim ve son olarak da başlıca yeryüzü cisimlerinin spektral özelliklerine yer verilmiştir. Üçüncü bölümde, kentleşme hakkında genel bilgiler verilmiş, kentsel gelişimin belirlenmesinde uzaktan algılamanın sağladığı kolaylıklar anlatılarak literatürdeki kentsel gelişim ile ilgili çalışmalardan örnekler gösterilmiştir. Dördüncü bölümde ise dijital uydu görüntüsü kavramı açıklanarak, uzaktan algılamada kullanılan dijital görüntü işleme teknikleri (önişleme, görüntü zenginleştirme, sınıflandırma ve değişim saptama yöntemleri) anlatılmıştır. Uygulama bölümünde öncelikle çalışmanın uygulama alanı ve kullanılan veriler hakkında bilgiler verilmiştir. Ardından farklı tarihli LANDSAT-TM uydu görüntülerine kontrollü sınıflandırma işlemi uygulanmıştır ve yapılan kontrollü sınıflandırmaların doğruluğu analiz edilmiştir. Daha sonra 1984 ve 2011 yıllarına ait LANDSAT-TM uydu görüntülerine yerleşim alanlarının belirlenmesine katkı sağlayan 7 farklı spektral indeks uygulanmıştır. Spektral indeks görüntüleri ve orijinal görüntüler arasında korelasyon analizi yapılarak en uygun indeks görüntüleri (NDVI ve NDBI) orijinal görüntülere dahil edilerek tekrar kontrollü sınıflandırma işlemi yapılmış ve kontrollü sınıflandırmalara ait doğruluk analizleri tekrar irdelenmiştir. Son olarak değişim analizi yapılarak göz önüne alınan tarihler arasında Beylikdüzü ilçesindeki kentsel gelişim nicel ve nitel olarak belirlenmiştir. Sonuç kısmında, uygulama bölümündeki sonuçlar karşılaştırılarak, uydu görüntülerinin kentsel gelişimde kullanılabilirliği değerlendirilmiştir.Today, with remote sensing, a large part of the Earth surface is being monitored in different spectral regions synchronously. Data, obtained by using remote sensing satellites, is used in several applications. One of the most commonly used applications is the monitoring of the urban development. With using multi-date satellite images, monitoring urban development and determination land cover/land use changes after urban expansion are possible. The district of Beylikdüzü shows a rapid urban development from the end of the 1990s and has become one of the most important district of the city. In this study, the changes in the residential areas and surrounding land cover of the district of Beylikdüzü in 27 years are analyzed using remote sensing methods. In the analysis, multi-temporal data collected from LANDSAT 5-TM satellite (June 12, 1984 and June 23, 2011) are used. As an image processing method, image classification, which is the one of the technique used in change detection analysis, together with several spectral indices is performed. In the introduction part of the study, definition of remote sensing and a brief summary of the topic are given. In the second chapter, electromagnetic energy, electromagnetic spectrum, energy interaction with the atmosphere and lastly main earth surface spectral features are explained as the fundamentals of remote sensing. In the third chapter, general information about urbanization is given and the feasibility of remote sensing techniques in urban studies is explained providing some examples cited in the literature. In the fourth chapter, digital image is defined and digital image processing methods used (preprocessing, image enhancement, classification and change detection techniques) in remote sensing are explained. Preprocessing operations could include radiometric processing to adjust digital values for effects of a hazy atmosphere and/or geometric processing to bring an image into registration with a map or another image. Preprocessing forms a preparatory phase that, improves image quality as the basis for later analyses that will extract information from the image. It should be emphasized that, although certain preprocessing procedures are frequently used, there can be no definitive list of standard preprocessing steps, because each project requires individual attention and some preprocessing decisions may be a matter of personal preference. Furthermore, the quality of image data varies greatly, so some data may not require the preprocessing that would be necessary in other instances. Image enhancement techniques improve the quality of an image as perceived by a human. These techniques are most useful because many satellite images when examined on a color display give inadequate information for image interpretation. There is no conscious effort to improve the fidelity of the image with regard to some ideal form of the image. There exists a wide variety of techniques for improving image quality. The contrast stretch, principal component analysis, spatial filtering, arithmetic band operations and spectral indices are the most commonly used techniques. In this study, seven different spectral indices are examined. One of them is the Normalized Difference Vegetation Index (NDVI). This index is a numerical indicator that uses the visible and near-infrared bands of the electromagnetic spectrum and is adopted to analyze remote sensing measurements and assess whether the target being observed contains live green vegetation or not. Generally, healthy vegetation will absorb most of the visible light that falls on it, and reflects a large portion of the near-infrared light. Unhealthy or sparse vegetation reflects more visible light and less near-infrared light. Bare soils on the other hand reflect moderately in both the red and infrared portion of the electromagnetic spectrum. To ensure the authenticity of the urban boundary, another spectral index, the Normalized Difference Built-up Index (NDBI) was used to extract the urban boundary. NDBI takes advantage of the unique spectral response of built-up areas and the other land covers. Another spectral index used in urban analyses is the Normalized Difference Impervious Surface Index (NDISI). The visible bands and middle infrared bands together with the near infrared band, can form a weak-reflectance group of impervious surface (related to urbanization feature classes), and the ratio of thermal band to this group should be effective to enhance impervious surface and suppress soil, sand and water noise. This index is designed to maximize the radiation that is emitted from impervious surface in the form of heat by using thermal wavelengths, minimize the low reflectance of near infrared, middle infrared and visible bands by impervious surface and take advantage of the high reflectance of middle infrared by sand and soil. Another one, the Soil and Vegetation Index (SVI) is a spectral index based on spectral response in the shortwave infrared domain and has demonstrated good performances in discriminating impervious from non-impervious surfaces. Band Ratio for Built-up Area (BRBA) and Soil index were also developed to extract built up areas and bare soil respectively. Lastly, in order to effectively represent major biophysical compositions in an urban environment, the BCI (Biophysical Composition Index) was designed to follow the mechanism of V–I–S (Vegetation-Impervious surface -Soil) triangle model. With the BCI, impervious surfaces are expected to have positive and relatively high values; vegetation is expected to be differentiated from other land covers through its negative and low values; and bare soil is expected to have a value of near zero, and can be separable from impervious surfaces. To reach this objective, a reexamination of Tasseled Cap (TC) transformation is conducted and evaluated whether the BCI can be derived using TC components. Image classification refers to the computer-assisted interpretation of remotely sensed images. There are two general approaches to image classification: supervised and unsupervised. They differ in how the classification is performed. In the case of supervised classification, the software system delineates specific land cover types based on statistical characterization data drawn from known examples in the image (known as training sites). With unsupervised classification, however, clustering software is used to uncover the commonly occurring land cover types, with the analyst providing interpretations of those cover types at a later stage. A vital step in the classification process, whether supervised or unsupervised, is the assessment of the accuracy of the final images produced. This involves identifying a set of sample locations that are visited in the field. The land cover found in the field is then compared to that which was mapped in the image for the same location. Statistical assessments of accuracy may then be derived for the entire study area, as well as for individual classes. Change detection is a technology ascertaining the changes of specific features within a certain time interval. It provides the spatial distribution of features and qualitative and quantitative information of features changes. The quantitative analysis and identifying the characteristics and processes of surface changes are carried through from the multi-temporal of remote sensing data. There are many types of change detection methods of multi-spectral image data. Some of these are band rationing, regression analyses, post classification, spectral indices, principal component analyses etc. In the application chapter, firstly study area and satellite data used are defined. After, supervised classification is applied to two LANDSAT-TM images taken in the years 1984 and 2011. Then, the accuracy assessments of the classifications are done. As a next step, seven different spectral indices that are used in urban analysis are performed to the LANDSAT-TM images. As a next step, the correlation analysis is done between the original images and spectral indices. According to the evaluation, the NDVI and NDBI are taken into consideration as the best spectral indices. Afterwards, these two indices are added as a new band to the original LANDSAT images and reclassified. Regarding to these new classifications, the accuracy assessment are redone and compared with the previous ones. Finally, the change detection analysis is performed and the urban development of Beylikdüzü in 27 years is evaluated both qualitatively and quantitatively. In the final chapter, the results obtained in the application stage are discussed and the usage and the efficiency of the satellite images for urban analysis are outlined.Yüksek LisansM.Sc