object oriented

Most of critical issues such as increase in pollution levels, sudden climatic changes and the rise of temperature in the urban area, leading to the formation of Urban Heat Islands (UHI), have been resulted from urbanization. As population density increases, most terrestrial areas become cities, and cities grow very fast. The reason to do the current study is to compare Single-Channel, SEBAL and Split-Window methods and then choose the best method for estimating land surface temperature. The objectives are as follows: Three independent studies were conducted using a series of Landsat data: (i) to land-use/land-cover (LU/LC) classification by object-oriented method and change detection; (ii) to understand the connection between particular LU/LC class and Land Surface Temperature(LST); and (iii) LST recovery using Single-Channel, SEBAL and Split-Window, as well as comparing these methods together. The results of land-use classification and change detection indicated that urban areas have increased, while agriculture has declined. The results of validation of the three temperature recovery methods demonstrated that due to using two thermal bands simultaneously, the Split-Window method functions better and in these three algorithms, water bodies and wet soils exhibit minimum surface temperatures. Due to less vegetation, areas such as deserts, saline soils and residential area display a higher surface temperature. Vegetation has always been an obstacle for heat input and inversely related to surface heat. In addition, due to fuel pollution of machinery and factory, urban areas experience high temperatures. The only gap of this study was the utilizing 5-cm surface temperature data, which was only available at airports and was not available.

Soil erosion and sediment yield in the downstream areas, water transfer canals, and dams are the most serious problems in the world today. Soil erosion threatens soil resources, causes severe damage to infrastructures, and imposes high costs on agriculture, watershed management, and natural resources. Reducing these hazards and damages due to soil erosion and sediment yield requires the use of quantitative data to identify critical areas that require immediate protection. Due to the high cost and time consuming of conventional methods, the use of new remote sensing technologies and satellite imagery is essential. This study used the MPSIAC model, one of the most well-known models for estimating soil erosion and sediment yield in Iran, geographical information system (GIS), and satellite image processing with object-oriented and pixel-based methods. For this purpose, basic data were prepared using base maps, Sentinel-2 satellite imagery, meteorological and hydrometric data, and fieldwork. After establishing a database, the score for each of the nine factors of the MPSIAC model was determined using the three common, object-oriented, and pixel-based processing methods. The extent of soil erosion and sediment yield of the watershed was determined within each hydrologic unit. Based on the results, the soil erosion and sediment yield intensities of the Lighvan watershed were classified as medium class (III). However, the comparison of the specific soil erosion and sediment yield values obtained from the three methods showed that the use of object-oriented methods in determining the values for land cover, land use, and current soil erosion state increased the accuracy of the predictions (with the estimated error of 12.18% and 13.15% for sediment yield and erosion, respectively) compared to common (with the estimated error of 15.73% and 16.71%) and pixel-based (with the estimated error of 18.78% and 19.45%) methods.