Presentation 1998/4/17
Applying the Prediction Method based upon the Scale Expansion to the Feature Extraction of Fractal Geometry
Shozo TOKINAGA, Yoshinori KISHIKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) It is known that the natural phenomina and the data representing social activities have the self-similarity, and are explained by using the fractal theroy. But usually, real observed data include a kind of white noise and specific pattern which are superposed on the ideal fractal geometry. Evenmore, the parameters of the fractal geometry such as the fractal dimension vary along the time. Therefore, it is necessary for the design of database management system of the fractal geometry to develop a efficient method to detect such kind of features and to divide to several segments of fractal geometry where the parameters are constant. This report deals with the characterization of fractal geometry(noise reduction, extarction of feature patterns and segmentation)by using the prediction of fractal geometry based upon the scale expansion given by the authors. The scale expansion corresponds to a kind of nonlinear fltering to detect the feature patterns from the fractal geometry. As an application, the the feature extraction for stock trend and the communication traffic are shown for the cases of fractral time series. And we also show the extraction of the feature patterns of cloud.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Fractal / Prediction / Feature extraction / Wavelet transform
Paper #
Date of Issue

Conference Information
Committee DSP
Conference Date 1998/4/17(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Digital Signal Processing (DSP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Applying the Prediction Method based upon the Scale Expansion to the Feature Extraction of Fractal Geometry
Sub Title (in English)
Keyword(1) Fractal
Keyword(2) Prediction
Keyword(3) Feature extraction
Keyword(4) Wavelet transform
1st Author's Name Shozo TOKINAGA
1st Author's Affiliation Faculty of Economics, Kyushu University, Fukuoka, 812 Jaoan()
2nd Author's Name Yoshinori KISHIKAWA
2nd Author's Affiliation Faculty of Economics, Kyushu University, Fukuoka, 812 Jaoan
Date 1998/4/17
Paper #
Volume (vol) vol.98
Number (no) 3
Page pp.pp.-
#Pages 8
Date of Issue