Wavelet fuzzy classification for detecting and tracking region outliers in meteorological data

In this paper, a wavelet fuzzy classification approach is proposed to detect and track region outliers in meteorological data. First wavelet transform is applied to meteorological data to bring up distinct patterns that might be hidden within the original data. Then a powerful image processing technique, edge detection with competitive fuzzy classifier, is extended to identify the boundary of region outlier. After that, to determine the center of the region outlier, the fuzzy-weighted average of the longitudes and latitudes of the boundary locations is computed. By linking the centers of the outlier regions within consecutive frames, the movement of a region outlier can be captured and traced. Experimental evaluation was conducted on a real-world meteorological data to examine the effectiveness of the proposed approach. This work will help discover interesting and implicit information for large volume of meteorological data.

[1]  Shashi Shekhar,et al.  A Unified Approach to Detecting Spatial Outliers , 2003, GeoInformatica.

[2]  F. Russo,et al.  A user-friendly research tool for image processing with fuzzy rules , 1992, [1992 Proceedings] IEEE International Conference on Fuzzy Systems.

[3]  Chang-Tien Lu,et al.  Detecting region outliers in meteorological data , 2003, GIS '03.

[4]  Ambuj K. Singh,et al.  SWAT: hierarchical stream summarization in large networks , 2003, Proceedings 19th International Conference on Data Engineering (Cat. No.03CH37405).

[5]  Carl G. Looney,et al.  Pattern recognition using neural networks: theory and algorithms for engineers and scientists , 1997 .

[6]  F. Russo.  A new class of fuzzy operators for image processing: design and implementation , 1993, [Proceedings 1993] Second IEEE International Conference on Fuzzy Systems.

[7]  Carl G. Looney,et al.  A Fuzzy Classifier Network with Ellipsoidal Epanechnikovs , 2001 .

[8]  Douglas M. Hawkins.  Identification of Outliers , 1980, Monographs on Applied Probability and Statistics.

[9]  Carl G. Looney,et al.  Nonlinear Rule-based Convolution for Refocusing , 2000, Real Time Imaging.

[10]  Lily R. Liang,et al.  Competitive fuzzy edge detection , 2003, Appl. Soft Comput..

[11]  Vic Barnett,et al.  Outliers in Statistical Data , 1980 .

[12]  Vijayalakshmi Atluri,et al.  Neighborhood based detection of anomalies in high dimensional spatio-temporal sensor datasets , 2004, SAC '04.

[13]  Raymond T. Ng,et al.  A unified approach for mining outliers , 1997, CASCON.

[14]  Hans-Peter Kriegel,et al.  OPTICS-OF: Identifying Local Outliers , 1999, PKDD.

[15]  Philip S. Yu,et al.  A Framework for Clustering Evolving Data Streams , 2003, VLDB.

[16]  Theodore Johnson,et al.  Fast Computation of 2-Dimensional Depth Contours , 1998, KDD.

[17]  Carl G. Looney,et al.  Radial basis functional link nets and fuzzy reasoning , 2002, Neurocomputing.

[18]  Giovanni Ramponi,et al.  Fuzzy operator for sharpening of noisy images , 1992 .

[19]  Jiawei Han,et al.  Spatial Data Mining: Progress and Challenges , 1996, Workshop on Research Issues on Data Mining and Knowledge Discovery.

[20]  Shenghuo Zhu,et al.  A survey on wavelet applications in data mining , 2002, SKDD.

[21]  Nick Efford,et al.  Digital Image Processing: A Practical Introduction Using Java , 2000 .

[22]  Aidong Zhang,et al.  WaveCluster: A Multi-Resolution Clustering Approach for Very Large Spatial Databases , 1998, VLDB.

[23]  Daniel Zeng,et al.  A comparative study of spatio-temporal hotspot analysis techniques in security informatics , 2004, Proceedings. The 7th International IEEE Conference on Intelligent Transportation Systems (IEEE Cat. No.04TH8749).

[24]  Xuefeng Ya.  Research issues in spatio-temporal data mining , 2003 .

[25]  C. Torrence,et al.  A Practical Guide to Wavelet Analysis. , 1998 .

[26]  Carl G. Looney,et al.  Pattern recognition using neural networks , 1997 .

[27]  Tao Cheng,et al.  A multi-scale approach to detect spatio-temporal outliers , 2004 .