Title :
Classification of sea ice types from ScanSAR RADARSAT images using pulse-coupled neural networks
Author :
Karvonen, Juha ; Similä, Markku
Author_Institution :
Finnish Inst. of Marine Res., Helsinki, Finland
Abstract :
Pulse-coupled neural networks (PCNN) are used for classifying sea ice in the Baltic Sea based on RADARSAT SAR images. The PCNN-classification is improved by computing additional features from the image and defining some rules to split and merge the preliminary classification produced by PCNN
Keywords :
geophysical signal processing; geophysics computing; image classification; neural nets; oceanographic techniques; radar imaging; remote sensing by radar; sea ice; spaceborne radar; synthetic aperture radar; Baltic Sea; RADARSAT; SAR; ScanSAR; image classification; image processing; measurement technique; neural net; ocean; pulse coupled neural network; radar image; radar imaging; radar remote sensing; sea ice type; Image coding; Image edge detection; Image resolution; Image segmentation; Joining processes; Layout; Neural networks; Radar scattering; Sea ice; Testing;
Conference_Titel :
Geoscience and Remote Sensing Symposium Proceedings, 1998. IGARSS '98. 1998 IEEE International
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-4403-0
DOI :
10.1109/IGARSS.1998.702260
