Evaluation of Beach Hydromorphological Behaviour and Classification Using Image Classification Techniques

• Tipo: Artigo em Revista Científica Internacional
• Data: 2009

Autores

Teodoro, AC (Autor) FCUP; Pais Barbosa, J (Autor) Outra; Veloso Gomes, F (Autor) FEUP; Taveira Pinto, F (Autor) Outra

Revista

Título: Journal of Coastal Research

Nº SI 39

Páginas: 1607-1611

ISSN: 0749-0208

Editora: Coastal Education & Research Foundation, Inc.

Indexação

ISI Web of Knowledge - 24 Citações

Scopus - 26 Citações

Classificação Científica

FOS: Ciências exactas e naturais > Ciências da terra e ciências do ambiente

Outras Informações

ID Authenticus: P-003-QYG

Abstract (EN): The development of new techniques on coastal studies, such as Geographic Information Technologies (GIT), using aerial photography and high resolution satellite images is an important issue on Coastal Engineering and Geo-Spatial Sciences research. Several aerial photographs (between 1958 and 2003), using visual interpretation on a Geographic Information System (GIS) environment, were used on the identification of the coastal features in a high energetic selected area of the NW Portuguese coast, between Esmoriz and Mira beach. The main goal of this work, based on the image classification techniques, is to identify and to analyse morphological features and hydrodynamic patterns and to compare these results with the visual interpretation. To achieve these objectives different techniques of image analysis were tested. Different supervised classification algorithms such as parallelepiped, minimum distance and maximum likelihood, were applied in order to identify morphological features and hydrodynamic patterns. Unsupervised classification algorithms (clustering) were also employed in order to determine the natural groupings or structures in the aerial photographs. The supervised classification algorithm presents good performance, demonstrated by the results of the confusion matrix, Kappa coefficient and overall accuracy. The parallelepiped classifier presents an overall accuracy of 95.65% and a Kappa statistics of 0.95661. The results for maximum likelihood were similar, 95.85% and 0.95840, respectively. The minimum distance classifier presents lower performance. The unsupervised classification algorithms (K-means and ISODATA) allows for indentify several classes such as breaking zone, beach face and beach. The results of the supervised and unsupervised classification algorithms were compared with the visual identification. The obtained results were in agreement with the visual interpretation.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 5