Remote sensing and field assessment of tsunami effects on coastal pond aquaculture in northern Sumatra

An attempt was made to conduct spatial assessment of the pattern and extent of damage to coastal aquaculture ponds along the east coast of Aceh province in Sumatra, Indonesia, resulting from the tsunami event of 26 December 2004. High-resolution satellite imagery, i.e., SPOT-5 multispectral scenes covering the 700 km stretch of the coast, acquired before and after the tsunami, were digitally enhanced and visually interpreted to delineate pockets of aquaculture ponds that were discerned to be damaged and relatively intact. Field checks were conducted at 87 sites in the four eastern coastal districts. The results indicate that SPOT-5 multispectral imagery was minimally sufficient to detect areas of damaged and relatively intact aquaculture ponds, but the 10-m spatial resolution poses limitations to evaluating the extent of pond damage. Nevertheless, the 60 km swath of the imagery makes it reasonably affordable for large-area assessment to identify pockets of severe damage for targeting more detailed assessments. The image maps produced from a mosaic of the SPOT-5 scenes can also serve as base maps for spatial planning in the challenging task of reconstruction and rehabilitation of the disrupted livelihoods of the coastal communities.

Application of remote sensing to marine fisheries: alternative approaches

A discussion is presented on the applications of remote sensing to fisheries. The measurement of temperature, wind stress, and ocean colour using remote sensing techniques is considered.

Semantic Discriminant mapping for classification and browsing of remote sensing textures and objects

We present a new approach based on Discriminant Analysis to map a high dimensional image feature space onto a subspace which has the following advantages: 1. each dimension corresponds to a semantic likelihood, 2. an efficient and simple multiclass classifier is proposed and 3. it is low dimensional. This mapping is learnt from a given set of labeled images with a class groundtruth. In the new space a classifier is naturally derived which performs as well as a linear SVM. We will show that projecting images in this new space provides a database browsing tool which is meaningful to the user. Results are presented on a remote sensing database with eight classes, made available online. The output semantic space is a low dimensional feature space which opens perspectives for other recognition tasks. © 2005 IEEE.