DocumentCode
3480958
Title
Segmentation of remote-sensing images by artificial neural networks
Author
Olmez, Tamer ; Dokur, Zumray
Author_Institution
Istanbul Tech. Univ., Turkey
fYear
2004
fDate
28-30 April 2004
Firstpage
84
Lastpage
86
Abstract
A novel unsupervised incremental neural network (Turkish acronym - DAYS) is proposed for the segmentation of remote-sensing images. Feature vectors are formed by the intensity of one pixel of each channel. The training set of a DAYS network is formed using all pixels of the image. The remote-sensing image is segmented according to the decision of the network. In the study, the segmentation results of DAYS and Kohonen networks are compared.
Keywords
geophysical signal processing; image segmentation; neural nets; remote sensing; self-organising feature maps; unsupervised learning; Kohonen networks; artificial neural networks; feature vectors; remote-sensing image segmentation; training set; unsupervised incremental neural network; unsupervised neural network; Artificial neural networks; Image segmentation; Neural networks; Pixel; Remote sensing;
fLanguage
English
Publisher
ieee
Conference_Titel
Signal Processing and Communications Applications Conference, 2004. Proceedings of the IEEE 12th
Print_ISBN
0-7803-8318-4
Type
conf
DOI
10.1109/SIU.2004.1338263
Filename
1338263
Link To Document