Remote Sensing, Landsat 8, Salt Dome, Classification, Artificial Neural Network

At the present time, remote sensing can provide the opportunity for mapping lithology, mineralogy, altered rocks, and environmental pollution, and is a useful tool for acquiring basic information, particularly on a regional scale. Significant phenomena in the field of geology are evapotranspiration, including salt domes. Evaporative structures are geological formations that are geographically expansive. One of the important morphological phenomena associated with this evapotranspiration is the structural development of salt domes. Salt domes The structures of geology are a dome of the shape formed by the movement of salt and its ascent in the diaphysmic mechanism. Salt domes and adjacent sediments are examples of a complex geological environment. Their study is due to the unique tectonic and lithological properties of salt, The existence of important resources in terms of the economic aspect and the effect of these evapotranspiration zones on the quality of resources around the salt domes is of great importance in geology, management, and human resource planning. Remote sensing technology in recent years has taken a strong role in obtaining information from these unique phenomena. So satellite imagery classification is one of the most important stages in the interpretation of satellite data, which allows users to produce various types of information, such as the production of covert maps, usage and discoveries of changes and influences.

This section consists of three steps: (1) In this study, the Landsat-8 satellite imaging imaging (OLI) image sensor on November 15, 2014 was used to carry out remote sensing studies for the classification and mapping of the global salt dome.(2) The data preprocessing stage is one of the most important steps in image processing, since all subsequent calculations are based on the image produced at this stage. The type and type of operation of this operation will vary depending on various factors such as the type of data used and the purpose of the research. In the process of preprocessing satellite imagery, it is necessary to remove any errors, such as atmospheric effects, before the identification and extraction of information.(3) PCA: The principal components analysis method is aimed at compressing the dataset in different bands of an image and in order to remove similar information. The main components of decomposition in the interpretation of digital remote sensing data are of great importance. The most important benefits of the main components of collecting and aggregating information on phenomena in different bands are less in a number of bands or components, in other words, the main components To remove excess data in satellite data, it is used extensively. The output of this method is usually a new and limited range of bands whose correlation between them is minimized, so they can be interpreted non-dependent on the original data. In general there are three stages in the classification of the neural network. The first step is an educational process using input data and educational prototypes. The second step is the validation phase that determines the success of the training and network authentication, validating and testing the network by some non-teaching samples. The last grade is the classification stage, in which a map is classified based on educational relationships during the alignment phase.

When the results of the tables are examined, several conclusions are drawn:It was observed that the class of sand-salt with 100% accuracy, clay class, 96.05%, gypsum-salt class, 99.33%, limestone class 100%, class of plants, 96/73%, sandstone class, 94/67% Salt rocks are 96.9%, gypsum soils are 93/58%. It is noteworthy that the lowest accuracy between classes is shale, which is 86.73%. Of the 185 pixels of this class, 170 pixels are correctly positioned on the shale floor, 1 pixel on the floor of the gypsum salt, 1 pixel on the clay, 3 pixels on the floor of the plants, 2 pixels on the sandstone floor, 1 pixel on the rock floor Salt and 7 pixels in gypsum soils are in error in other classes. ), It can be said that the artificial network method with the correctness of the total of 95/3501% and Kappa coefficient of 94.37% have a good performance in classifying and preparing the map of the study area.

With the launch of Landsat in 1972, remote sensing technology has opened a new horizons in the planning, research, assessment and management of natural resources. This phenomenon provides a new method for efficient and effective mapping of various terrain zones, including salt domes. Detailed information can be extracted from temporary satellite data and used as input for decision making in geographic information systems. Evaporative structures are among the geological formations that are geographically expansive in our country, including Zagros china. One of the phenomena of the morphological index associated with this evaporation structure is the structural development of salt domes. The study of salt domes due to the unique properties of salt in terms of tectonic and lithology and strong interactions between motor and thermal flows is of great importance in geology. In this study, the artificial neural skull method and Landsat 8 satellite imagery were used to classify and prepare a global salt dome map.