The application of segmented cellular neural networks (SCNNs) for improving noisy satellite imagery performance

Author

Tolluoglu, A. Orhan

Author_Institution

Turkish War Colleges, Air War Coll., Istanbul, Turkey

fYear

2005

fDate

9-11 June 2005

Firstpage

623

Lastpage

625

Abstract

In this paper, segmented cellular neural networks (SCNN) have been applied to noisy satellite imagery to improve its performance and appearance. Because of the importance of imagery quality, SCNN has been applied to data for image processing applications that for noise filtering. Multi-level non-linear output capability of SCNN improves image quality. In training recurrent perceptron learning algorithm (RPLA) is used as a learning algorithm. They are applied to noise mounted satellite imagery successfully.

Keywords

cellular neural nets; geophysical signal processing; image processing; multilayer perceptrons; remote sensing; RPLA; SCNN; image processing applications; image quality; imagery quality; multi-level nonlinear output; noisy satellite imagery; recurrent perceptron learning algorithm; segmented cellular neural networks; Cellular neural networks; Circuit noise; Educational institutions; Equations; Filtering; Image processing; Image quality; Image segmentation; Pattern recognition; Satellites;

fLanguage

English

Publisher

ieee

Conference_Titel

Recent Advances in Space Technologies, 2005. RAST 2005. Proceedings of 2nd International Conference on

Print_ISBN

0-7803-8977-8

Type

conf

DOI

10.1109/RAST.2005.1512642

Filename

1512642