| Revista: | Boletim de ciencias geodesicas |
| Base de datos: | PERIÓDICA |
| Número de sistema: | 000459898 |
| ISSN: | 1413-4853 |
| Autores: | Ferreira, Italo Oliveira1 Santos, Afonso de Paula dos1 Oliveira, Julio Cesar de1 Medeiros, Nilcilene das Gracas1 Emiliano, Paulo Cesar2 |
| Instituciones: | 1Universidade Federal de Vicosa, Departamento de Engenharia Civil, Vicosa, Minas Gerais. Brasil 2Universidade Federal de Vicosa, Departamento de Estatística, Vicosa, Minas Gerais. Brasil |
| Año: | 2019 |
| Volumen: | 25 |
| Número: | 3 |
| País: | Brasil |
| Idioma: | Inglés |
| Tipo de documento: | Artículo |
| Enfoque: | Analítico, descriptivo |
| Resumen en inglés | Currently, during the operation in shallow waters, scanning systems, such as multibeam systems, are capable of collecting thousands of points in a short time, promoting a greater coverage of the submerged bottom, with consequent increase in the detection capacity of objects. Although there has been an improvement in the accuracy of the depths collected, traditional processing, that is, manual, is still required. However, mainly due to the increased mass of data collected, manual processing has become extremely time-consuming and subjective, especially in the detection and elimination of spikes. Several algorithms are used to perform this task, but most of them are based on statistical assumptions hardly met and/or verified, such as spatial independence and normality. In this sense, the goal of this study is to present the SODA (Spatial Outlier Detection Algorithm) methodology, a new method for detection of spikes designed to treat bathymetric data collected through swath bathymetry systems. From computational simulation, promising results were obtained. SODA, in some cases, was capable to identify even 90% of spikes inserted on simulation, showing that the methodology is efficient and substantial to the bathymetric data treatment |
| Disciplinas: | Geociencias |
| Palabras clave: | Oceanografía, Spikes, Valores atípicos, Ecosondas multihaz, Tratamiento de datos multihaz |
| Keyword: | Oceanography, Spikes, Outliers, Multibeam echo sounders, Multibeam Data Processing |
| Texto completo: | Texto completo (Ver HTML) Texto completo (Ver PDF) |
