A simplified pulse-coupled neural network based sea-ice classifier with graphical interactive training

Author

Karvonen, Juha

Author_Institution

Finnish Inst. of Marine Res., Finland

Volume

2

fYear

2000

fDate

2000

Firstpage

681

Abstract

The author has applied the pulse-coupled neural network (PCNN) for sea ice segmentation and classification. In this paper a simplified version of the PCNN, which is faster to execute and requires less memory, is used. Additionally, the author is using an on line training algorithm. During the interactive training phase, the users can adjust some parameters of the PCNN using a set of training images

Keywords

geophysical signal processing; geophysics computing; image classification; image segmentation; learning (artificial intelligence); neural nets; oceanographic techniques; remote sensing; sea ice; PCNN; geophysical measurement technique; graphical interactive training; image classification; image segmentation; neural net; ocean; on line training algorithm; pulse-coupled neural network; remote sensing; sea ice; sea-ice classifier; training image; Classification algorithms; Equations; Image resolution; Image segmentation; Instruments; Iterative algorithms; Navigation; Neural networks; Sea ice; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium, 2000. Proceedings. IGARSS 2000. IEEE 2000 International

Conference_Location

Honolulu, HI

Print_ISBN

0-7803-6359-0

Type

conf

DOI

10.1109/IGARSS.2000.861670

Filename

861670