| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 843.15 KB | Adobe PDF |
Authors
Abstract(s)
O conhecimento do fundo marinho é um fator crucial para a segurança da navegação, assim como para as diversas atividades marítimas. O progresso da hidrografia tem mostrado novos horizontes trazendo diversos sistemas complexos que prometem resolver muitos dos desafios nesta área.
Respetivamente à aquisição de dados batimétricos através de sondadores de feixe-simples, este trabalho surge da necessidade de colmatar os erros dependentes do fator humano, reduzir o tempo de processamento e melhorar a qualidade do mesmo. Para tal, são necessárias ferramentas computacionais que auxiliem a fase de processamento de dados e a elaboração de produtos finais.
Deste modo, foi criada uma ferramenta, com auxílio a vários algoritmos e métodos desenvolvidos. Um dos utilizados é o Self Organizing Map (SOM) ou mapas auto-organizados. O estudo apresenta, sinteticamente, conceitos básicos de Levantamentos Hidrográficos para uma melhor compreensão dos dados analisados. De seguida, apresenta-se o modo de funcionamento do algoritmo SOM, abordando aspetos necessários para a realização de agrupamento de dados (clustering) como parte da solução da ferramenta desenvolvida. A partir deste, apresenta-se a metodologia desenvolvida para a realização da seleção de dados (sondas) necessária à criação de produtos para representação do fundo marinho. A ferramenta desenvolvida, Processamento de Dados Garmin com Self-Organizing Map (PDGSOM), é confrontada com outros sistemas de processamento de dados batimétricos, nomeadamente o software CARIS. Deste modo, comparando modelos batimétricos realizou-se uma análise de parâmetros de funcionamento e eficácia da ferramenta desenvolvida.
Em síntese, a presente dissertação tem como objetivo a construção de uma ferramenta com uma interface de fácil utilização, fornecendo uma informação segura na utilização dos dados e produtos obtidos, tendo como motivo de estudo a “segurança na navegação”.
The need of knowledge of the seabed in its whole, has been brought many challenges to maritime navigation. The safety of navigation is a crucial factor, and even with all the progress and the new horizons that hydrography science has shown with its numerous complex systems, there are still many obstacles to overcome. Regarding the acquisition of bathymetric data with single beam echo sounder, this work arises from the need to bridge the errors reflected on the human factor, improve the processing of data acquisition and time reduction. For all the reasons stated above, computational tools are of a main importance on the data processing phase and the elaboration of final products. For this study, I developed a tool, using Decision Support System, based on the algorithm of self-organizing map Kohonen. Better known for its clustering, visualization and classification capabilities with comprehensive solutions for selection of bathymetric data. The study begins by presenting, the basic concepts for a better understanding of the studied area, with respect to the Hydrographic Surveys. Next, the operation of the Self Organizing Map algorithm is presented, addressing aspects necessary to perform data grouping (clustering) as part of the solution of the developed tool. From this, I present the methodology used to develop the realization of data selection necessary to create products for representing the seabed. The developed tool, Garmin Data Processing with Self-Organizing Map (PDGSOM - Portuguese version) is compared with other bathymetric data processing systems, CARIS software. This allowed the comparison of digital elevation models, analysis of operation parameters and effectiveness of the developed tool. Succinctly, this dissertation has as main goal the construction of a tool with an easy-use interface, providing safe information in the use of the data and obtained products, based on the principles of "safety of navigation ".
The need of knowledge of the seabed in its whole, has been brought many challenges to maritime navigation. The safety of navigation is a crucial factor, and even with all the progress and the new horizons that hydrography science has shown with its numerous complex systems, there are still many obstacles to overcome. Regarding the acquisition of bathymetric data with single beam echo sounder, this work arises from the need to bridge the errors reflected on the human factor, improve the processing of data acquisition and time reduction. For all the reasons stated above, computational tools are of a main importance on the data processing phase and the elaboration of final products. For this study, I developed a tool, using Decision Support System, based on the algorithm of self-organizing map Kohonen. Better known for its clustering, visualization and classification capabilities with comprehensive solutions for selection of bathymetric data. The study begins by presenting, the basic concepts for a better understanding of the studied area, with respect to the Hydrographic Surveys. Next, the operation of the Self Organizing Map algorithm is presented, addressing aspects necessary to perform data grouping (clustering) as part of the solution of the developed tool. From this, I present the methodology used to develop the realization of data selection necessary to create products for representing the seabed. The developed tool, Garmin Data Processing with Self-Organizing Map (PDGSOM - Portuguese version) is compared with other bathymetric data processing systems, CARIS software. This allowed the comparison of digital elevation models, analysis of operation parameters and effectiveness of the developed tool. Succinctly, this dissertation has as main goal the construction of a tool with an easy-use interface, providing safe information in the use of the data and obtained products, based on the principles of "safety of navigation ".
Description
Keywords
Levantamentos Hidrográficos Expeditos Sondador de Feixe Simples Modelo batimétrico Mapas Auto-organizáveis Agrupamento de dados Dados Garmin Seleção de Sondas Métodos de Seleção Batimétrica Produtos Finais Ferramenta de Processamento de Dados Hidrográficos Google Earth Pro Minor Hydrographic Surveys Single-Beam Echo Sounder Self-Organizing Map SOM Clustering Bathymetric data processing Data Garmin Sounding Selection and Bathymetric Automated sounding selection Methodology Selection Hydrographic Data Processing Tool Modeling Products Digital Elevation Model Final Products Visualization with Google Earth Pro