Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/17057
Title: Tropical cyclone forecast using angle features and time warping
Authors: Liu, J
Feng, BO
Wang, M
Luo, W
Keywords: Atmospheric movements
Feature extraction
Geophysics computing
Gradient methods
Weather forecasting
Issue Date: 2006
Publisher: IEEE
Source: International Joint Conference on Neural Networks, 2006 : IJCNN '06, July 16-21, 2006, Vancouver, BC, p. 4330-4337 How to cite?
Abstract: The most popular approach to comparing two given tropical cyclones (TCs) is to measure the distance between various contour points of the TC extracted from a satellite image. However, this measure has a very high computational cost as it involves many point-to-point calculations. Moreover, this measure does not reflect the most distinctive features of a tropical cyclone, their spiral shape. In this paper, we propose the use of angle features and time warping for TC forecast. The gradient vector flow (GVF) snake model is applied to extract the contour points of a dominant tropical cyclone from the satellite image. Dvorak templates are used as references to predict the intensity of the tropical cyclone. Given two sets of contour points, one for each tropical cyclone, we retrieve the similarity of two shapes using angle features found among the successive contour points. We adopt a time warping approach to produce a fast and accurate result. Experimental results have shown that our approach is better than other conventional comparison approaches such as Hausdorff distance measure.
URI: http://hdl.handle.net/10397/17057
ISBN: 0-7803-9490-9
DOI: 10.1109/IJCNN.2006.247009
Appears in Collections:Conference Paper

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

31
Last Week
2
Last month
Checked on Aug 14, 2017

Google ScholarTM

Check

Altmetric



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.