Emprego de técnicas de processamento digital em imagens Landsat 8 para mapeamento geológico.

Abstract

This study aimed in the use of digital Processing techniques of remote sensing images, color composite RGB, bands subtraction and Principal Component Analysis - PCA (Crosta Technique) in scenes of OLI sensor, satellite Landsat 8, to assist in geological mapping, lithology identification and structure enhancement. The study area is located in the town of Canindé, inserted in Ceará Central Domain (neoarchans rocks to paleoproterozoic), located in the tectonic context of the northern portion of Borborema Province. The Principal Components Analysis was used in this study to determine and reduce the correlation between the bands of visible (VIS) and short waves infrared (SWIR), of the scene WRS 217/63. The eigenvalues and the eigenvectors were calculated enabling to choose the best RGB compositions, applied on acquired bands, with the aim to highlight and visualize lithological units and structures of the study area. Subtraction of the bands 4 and 2 OLI stressed soils rich in iron oxides derived from the weathering of migmatites and amphibolite lenses. The color composite RGB 543 OLI and RGB composition where the PC3 -OLI 2567 in the red channel, PC1 and PC4-0LI2456 in the green and blue channels respectively, highlighted structures NS main trend of the study area. Technical field trips had been proceeded for data acquisition and to confirm the information provided by Landsat 8 images.