Automatic classification and detection of landforms is one of the basic needs of environmental sciences, through which work units are extracted with high speed and accuracy. The aim of this study was to automatically detect landslides in Sefid Dasht basin using topographic position index in SAGA GIS software. The results of the present study showed that out of 10 landforms identified in the study area, the plains and open slopes with an area of 107.7 and 87.4 square kilometers, respectively, have the highest area among other landforms. Also, upland Drainages landforms and local ridges with 3.3 and 0.6 square kilometers, respectively, have the lowest area among other landforms. Landform mapping can be used in a variety of sciences such as environmental planning, natural resource conservation, and natural disaster forecasting.