Share Email Print
cover

Proceedings Paper

Gross error detection and correction based on wavelet transform and support vector machine
Author(s): Tingye Tao; Fei Gao; Zhaofu Wu
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In order to obtain high accuracy results, the gross errors in observations must be correctly detected and repaired. In this paper, the theory and methods of singularity detection based on wavelet transform, support vector machine regression model are introduced. The wavelet multi-resolution analysis (MRA) was carried out and the location of the gross errors can be detected by ascertaining the points of modulus maximal value of the wavelet coefficients since the gross error can be regarded as the singular point of the observation time series. Then the time series regression model based on support vector machine (SVM) was established to repair the gross errors. Practical test results indicate that the gross errors can be validly detected by wavelet method as well as be correctly repaired by the method based on support vector machine.

Paper Details

Date Published: 14 October 2009
PDF: 8 pages
Proc. SPIE 7492, International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining, 74921E (14 October 2009); doi: 10.1117/12.838574
Show Author Affiliations
Tingye Tao, Wuhan Univ. (China)
Hefei Univ. of Technology (China)
Fei Gao, Hefei Univ. of Technology (China)
Zhaofu Wu, Hefei Univ. of Technology (China)


Published in SPIE Proceedings Vol. 7492:
International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining
Yaolin Liu; Xinming Tang, Editor(s)

© SPIE. Terms of Use
Back to Top