Dark formation detection using neural networks

Topouzelis, K; Karathanassi, V; Pavlakis, P; Rokos, D

dc.contributor.author	Topouzelis, K	en
dc.contributor.author	Karathanassi, V	en
dc.contributor.author	Pavlakis, P	en
dc.contributor.author	Rokos, D	en
dc.date.accessioned	2014-03-01T01:28:05Z
dc.date.available	2014-03-01T01:28:05Z
dc.date.issued	2008	en
dc.identifier.issn	0143-1161	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/18708
dc.subject	Neural Network	en
dc.subject.classification	Remote Sensing	en
dc.subject.classification	Imaging Science & Photographic Technology	en
dc.subject.other	Accidents	en
dc.subject.other	Artificial intelligence	en
dc.subject.other	Backscattering	en
dc.subject.other	Computer architecture	en
dc.subject.other	Deformation	en
dc.subject.other	Detectors	en
dc.subject.other	Feedforward neural networks	en
dc.subject.other	Hazardous materials spills	en
dc.subject.other	Image classification	en
dc.subject.other	Image enhancement	en
dc.subject.other	Imaging systems	en
dc.subject.other	Imaging techniques	en
dc.subject.other	Lightning	en
dc.subject.other	Marine pollution	en
dc.subject.other	Marine radar	en
dc.subject.other	Multilayer neural networks	en
dc.subject.other	Network architecture	en
dc.subject.other	Nonlinear optics	en
dc.subject.other	Oil spills	en
dc.subject.other	Pollution	en
dc.subject.other	Radar	en
dc.subject.other	Radial basis function networks	en
dc.subject.other	Synthetic aperture radar	en
dc.subject.other	Synthetic apertures	en
dc.subject.other	Target drones	en
dc.subject.other	Topology	en
dc.subject.other	Tracking radar	en
dc.subject.other	Vegetation	en
dc.subject.other	Artificial neural network (ANNs)	en
dc.subject.other	Feed forward (FF)	en
dc.subject.other	High resolution imagery	en
dc.subject.other	High resolution satellite images	en
dc.subject.other	High resolutions	en
dc.subject.other	Local weather conditions	en
dc.subject.other	Marine environments	en
dc.subject.other	Multi layer perceptron (MLP)	en
dc.subject.other	Multilayer perceptron (MLP) networks	en
dc.subject.other	Nonlinear behaviours	en
dc.subject.other	Oil spills	en
dc.subject.other	Pixel values	en
dc.subject.other	Radar backscatter	en
dc.subject.other	Radial-basis function (RBF)	en
dc.subject.other	SAR Images	en
dc.subject.other	Sea waves	en
dc.subject.other	Synthetic aperture radar (SAR) images	en
dc.subject.other	Neural networks	en
dc.subject.other	architecture	en
dc.subject.other	artificial neural network	en
dc.subject.other	detection method	en
dc.subject.other	image analysis	en
dc.subject.other	nonlinearity	en
dc.subject.other	pixel	en
dc.subject.other	remote sensing	en
dc.subject.other	satellite imagery	en
dc.subject.other	synthetic aperture radar	en
dc.subject.other	threshold	en
dc.subject.other	topology	en
dc.title	Dark formation detection using neural networks	en
heal.type	journalArticle	en
heal.identifier.primary	10.1080/01431160801891770	en
heal.identifier.secondary	http://dx.doi.org/10.1080/01431160801891770	en
heal.language	English	en
heal.publicationDate	2008	en
heal.abstract	Synthetic Aperture Radar (SAR) images are extensively used for dark formation detection in marine environment, as they are not affected by local weather conditions and cloudiness. Dark formations can be caused by man-made actions (e.g. oil spills) or natural ocean phenomena (e.g. natural slicks and wind front areas). Radar backscatter values for oil spills are very similar to backscatter values for very calm sea areas and other ocean phenomena because they dampen the capillary and short gravity sea waves. Thus, traditionally, dark formation detection is the first stage of the oil-spill detection procedure and in most studies is performed manually or using a fixed size window in which a threshold value is adopted. In high-resolution imagery, dark formation detection may fail due to the nonlinear behaviour of the pixel values contained in the dark formation and in the area around it. In this paper, we examine the ability of two feed-forward neural network families, i.e. Multilayer Perceptron (MLP) and the Radial Basis Function (RBF) networks, to detect dark formations in high-resolution SAR images. The general objective of this paper is to test the potential of artificial neural networks for dark formation detection using SAR high-resolution satellite images. Both the type and the architecture of the network are subjects of research. The inputs into the networks are the original SAR images. Each network is called to classify an area of the image as dark area or sea. The group of MLP networks can be recognized as the most suitable group for dark formation detection, as it presents reliable stable results for all the examined accuracies. Nevertheless, in terms of single topology, there is no an MLP topology that performs significantly better than the others.	en
heal.publisher	TAYLOR & FRANCIS LTD	en
heal.journalName	International Journal of Remote Sensing	en
dc.identifier.doi	10.1080/01431160801891770	en
dc.identifier.isi	ISI:000257997000006	en
dc.identifier.volume	29	en
dc.identifier.issue	16	en
dc.identifier.spage	4705	en
dc.identifier.epage	4720	en