Classification of textural distribution of seabed using hydroacoustics data: Case study of Ponta Negra beach, Natal/RN-Brazil

摘要

This work was based on the analysis of geophysical data acquired in hydroacoustic surveys of bathymetry and sonography using the inteferometric sonar EdgeTech 4600. The study area is a sector of the inner northeast Brazilian shelf adjacent to Natal (RN). The objective of the work was to identify and draw the different textural patterns which characterize the sea bottom of the study area and make the correlation between the seabed morphology and the textural patterns. In addition, two approaches for textural classification of sonograms were compared and evaluated, which were: GLCM automatic classification available in SonarWiz software; and the semi-automatic Maximum Likelihood, available in ArcGIS software. The comparison tested the capacity for identifying and drawing the textural patterns distribution. The automated classification identified 4 patterns while on the semi-automated 6 patterns were identified. It was made the correlation between the textural patterns found in each classification, besides the correlation between textural patterns and the levels of intensity of reflectance presents on the sonogram.