Share Email Print
cover

Proceedings Paper

An efficient 2D-WTMM and PNN approach to remove spurious radar echoes
Author(s): Mohamed Khider; Boualem Haddad
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The proposed method aims to reduce the spurious echoes in weather radar images collected at Melbourne radar site, using parameters from 2D-WTMM method based on the continuous wavelet transform, and including the PNN probabilistic neural network for the classification of pixels into two types of echoes : precipitation or parasite. Indeed, we propose the introduction of parameters related to wavelet transform skeletons, these parameters are proportional to the image texture roughness, anisotropy and the distance of separation between non-zero radar echoes cells and give good separation between rain and non-rain echoes. Radar image is first segmented with Voronoi’s cells according to the spatial distribution of Holder exponents. By comparing with a direct method of classification which takes into account only one parameter at a time by using a threshold, it was found that the combination of these three parameters with PNN approach improves the final results in terms of preserving precipitation echoes and elimination of weather radar clutter. Initial results show approximately the removal of 98% of clutter and preservation of 97% of precipitation echoes.

Paper Details

Date Published: 14 March 2013
PDF: 6 pages
Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 876846 (14 March 2013); doi: 10.1117/12.2011192
Show Author Affiliations
Mohamed Khider, Univ. des Sciences et de la Technologie Houari Boumediene (Algeria)
Boualem Haddad, Univ. des Sciences et de la Technologie Houari Boumediene (Algeria)


Published in SPIE Proceedings Vol. 8768:
International Conference on Graphic and Image Processing (ICGIP 2012)
Zeng Zhu, Editor(s)

© SPIE. Terms of Use
Back to Top