نوع مقاله: مقاله پژوهشی
نویسندگان
1 دانشجوی دکتری گروه آبخیزداری، دانشکده مرتع و آبخیزداری، دانشگاه علوم کشاورزی و منابع طبیعی گرگان، ایران
2 دانشیار گروه آبخیزداری، دانشکده مرتع و آبخیزداری، دانشگاه علوم کشاورزی و منابع طبیعی گرگان، ایران
3 دانشیار گروه مرتع و آبخیزداری، دانشکده منابع طبیعی و محیط زیست، دانشگاه بیرجند، ایران
4 استادیار گروه مدیریت مناطق بیابانی، دانشکده مرتع و آبخیزداری، دانشگاه علوم کشاورزی و منابع طبیعی گرگان، ایران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Land cover and land use are an important variable in natural land processes. Land use change in environmental protection programs and natural resource management plays an important role in the intensification of natural crises such as floods. The Gorganrood River basin in the Golestan province has historically experienced land use conversion. In this research was selected for land use classification using Landsat 8 OLI satellite images of the 25 June 2017. The goal of this study is to assess the accuracy of two approaches, pixel-based supervised classification and the object-oriented one base on quantity and allocation disagreement indexes. The accuracy assessment results indicated verified that for land use mapping the SVM algorithm using a 50 pixel segmentation in the object-based classification having a quantity disagreement of 2.03, an allocation disagreement of 4.58, and an overall accuracy of 92.65% and a kappa coefficient of 0.91 was more accurate than other algorithms in the object-based classification and other algorithms in the pixel-based classification. Based on this algorithm, the lowest of omission and commission error showed in forest lands and residential and industrial areas of 0.58% and 1.59% respectively. The highest of producer and user accuracy showed in forest lands and the water body of 99.44% and 99.41% respectively. The largest area of land use in the Gorganrood River basin is related to the Barren/Rangeland/Cropland class of 314110 ha. Finally, the SVM-SL50 algorithm in the object-based classification is suggested as an optimal classifier with a high accuracy for classification of land use classification maps in order to manage natural resources in Golestan province.
کلیدواژهها [English]