During the pre-processing step, the radiometry of satellite images was corrected; then, the Dark Object Subtraction (DOS) technique was applied to remove atmospheric noises and the shadow in the image (Atari-Hajipirloo et al., 2016)
Based on the goals, the proposed digital image processing has two parts: structural lineament extraction and lithological mapping. Image processing “enhancement” utilizes these to make the geological features clear. A color composite (false color RGB) was examined to differentiate surface geological units. The RGB composite 567 (equal to RGB 457 in Landsat 7 ETM+) is the best and most efficient composite for displaying the lithological units and lineament. Identical to the Landsat 7 ETM+, Band 5 (near-infrared: 0.851–0.879 m) is a helpful tool for presenting water or land surfaces because most of the radiation in this wavelength range is absorbed by water. Because of its sensitivity to moisture, band 6 (initial shortwave infrared: 1.566–1.651 m) is an excellent choice for monitoring plants. Band 7 (second shortwave infrared: 2.107–2.294 m) helps interpret the soil and geology. A series of interpretive components, such as color or tone, texture, pattern, and association, were used to determine the geological units and lineament. The geology map of Sarpol Zahab 1:100.000 and the topography map with a contour interval of 12.5 m were employed as references in the interpretation procedure (Atari-Hajipirloo et al., 2016). The steps of landsat8 OLI processing are described in (Figure 2.9).
Principal Component Analysis (PCA)
The principal component analysis (PCA) technique is a statistical approach widely used in geological investigations. The method transforms the information found in the original tapes can be assembled into new tapes called principal components (Adiri et al., 2017; Gabr et al., 2010). Consequently, PCA enhances and distinguishes specific spectral signatures from the background and discriminates between lithological units and geological structures (Gabr et al., 2010). The study focuses on PCA1 because it is sharp and has a more significant percentage of variance on the PCA1 axis than on the PCA2 axis, which is higher than on the PCA3 axis (Himyari et al., 2002).
The PC1 was utilized by automatically extracting lineaments, which were then applied to the PC1's directional filters in several orientations (NWSE, N-S, NE-SW, and E-W). The lineament identification was made visually based on geomorphological characteristics such as valley and river lineament, ridge rapid river shifting, ridge line, scarp face, and straight drainage segments. Rivers and other drainage features are the most visible evidence of underlying fracture zones (Gannouni and Gabtni, 2015; Lloyd, 1999). Straight river segments are so commonly regarded as geological lineaments. The vegetation indicator is another important method for determining the geological lineament. The pattern of vegetated and less vegetated zones, particularly during the dry season, can also be used to identify geological lineaments (Lloyd, 1999).
