Beach hydromorphological classification through image classification techniques applied to remotely sensed data

Evaluation of beach hydromorphological behavior and its classification is extremely complex. Several aerial photographs, using visual interpretation on a GIS environment, were previously used on the identification of coastal hydroforms and hydromorphologies, and to classify beach morphological stage in a selected area of the NW Portuguese coast. The goal of this study is to improve and develop new methodologies to identify coastal features and coastal patterns. In order to achieve that, pixel-based classification and object-oriented classification algorithms were employed, with the aim to identify and analyze morphological features and hydrodynamic patterns and to compare these results with the visual interpretation already performed. The dataset is composed by two aerial photographs (1996 and 2001) and one IKONOS-2 image (2004). The supervised classification algorithms presented good results both for aerial photographs and for IKONOS-2 image, demonstrated by its overall accuracy and Kappa coefficient values. For the two aerial photographs the best results were found for the maximum likelihood classifier and for the IKONOS-2 image the best result was archived with the parallelepiped classifier. The object-oriented classification performance for the aerial photographs was very good, identifying the classes of interest. The results obtained with the IKONOS-2 image were worst.