Editorial Type:
Article
Article Type:
Research Article

Automated Invariant Alignment to Improve Canonical Variates in Image Fusion of Satellite and Weather Radar Data

Jacob S. Vestergaard DTU Informatics, Technical University of Denmark, Kongens Lyngby, Denmark

Search for other papers by Jacob S. Vestergaard in
Current site
Google Scholar
PubMed Close
and
Allan A. Nielsen DTU Space, Technical University of Denmark, Kongens Lyngby, Denmark

Search for other papers by Allan A. Nielsen in
Current site
Google Scholar
PubMed Close
Print Publication:
01 Mar 2013
DOI:
https://doi.org/10.1175/JAMC-D-12-05.1
Page(s):
701–709
Restricted access

Abstract

Canonical correlation analysis (CCA) maximizes the correlation between two sets of multivariate data. CCA is applied to multivariate satellite data and univariate radar data to produce a subspace descriptive of heavily precipitating clouds. A misalignment, inherent to the nature of the two datasets, was observed, corrupting the subspace. A method for aligning the two datasets is proposed to overcome this issue and render a useful subspace projection. The observed corruption of the subspace gives rise to the hypothesis that the optimal correspondence between a heavily precipitating cloud in the radar data and the associated cloud top registered in the satellite data is found by a scale, rotation, and translation invariant transformation together with a temporal displacement. The method starts by determining a conformal transformation of the radar data at the time of maximum precipitation for optimal correspondence with the satellite data at the same time. This optimization is repeated for an increasing temporal lag until no further improvement can be found. The method is applied to three meteorological events that caused heavy precipitation in Denmark. The three cases are analyzed with and without using the proposed method. In all cases, the use of prealignment shows significant improvements in the descriptive capabilities of the subspaces, thus supporting the posed hypothesis.

Corresponding author address: Jacob S. Vestergaard, Technical University of Denmark, Asmussens Allé, DTU Informatics, 2800 Kongens Lyngby, Denmark. E-mail: jsve@imm.dtu.dk

Abstract

Canonical correlation analysis (CCA) maximizes the correlation between two sets of multivariate data. CCA is applied to multivariate satellite data and univariate radar data to produce a subspace descriptive of heavily precipitating clouds. A misalignment, inherent to the nature of the two datasets, was observed, corrupting the subspace. A method for aligning the two datasets is proposed to overcome this issue and render a useful subspace projection. The observed corruption of the subspace gives rise to the hypothesis that the optimal correspondence between a heavily precipitating cloud in the radar data and the associated cloud top registered in the satellite data is found by a scale, rotation, and translation invariant transformation together with a temporal displacement. The method starts by determining a conformal transformation of the radar data at the time of maximum precipitation for optimal correspondence with the satellite data at the same time. This optimization is repeated for an increasing temporal lag until no further improvement can be found. The method is applied to three meteorological events that caused heavy precipitation in Denmark. The three cases are analyzed with and without using the proposed method. In all cases, the use of prealignment shows significant improvements in the descriptive capabilities of the subspaces, thus supporting the posed hypothesis.

Corresponding author address: Jacob S. Vestergaard, Technical University of Denmark, Asmussens Allé, DTU Informatics, 2800 Kongens Lyngby, Denmark. E-mail: jsve@imm.dtu.dk
Save
Cover Journal of Applied Meteorology and Climatology
Journal of Applied Meteorology and Climatology

  • Anderson, T. W., 1984: An Introduction to Multivariate Statistical Analysis. 2nd ed. Wiley-Interscience, 686 pp.

  • Bell, A., and T. Sejnowski, 1995: An information-maximization approach to blind separation and blind deconvolution. Neural Comput., 7, 1129–1159.

    • Search Google Scholar
    • Export Citation

  • Breiman, L., and J. H. Friedman, 1985: Estimating optimal transformations for multiple regression and correlation. J. Amer. Stat. Assoc., 80, 580–598.

    • Search Google Scholar
    • Export Citation

  • D’Errico, J., cited 2012: Bound constrained optimization using fminsearchbnd. MathWorks. [Available online at http://www.mathworks.com/matlabcentral/fileexchange/8277-fminsearchbnd.]

  • Eldén, L., 2007: Matrix Methods in Data Mining and Pattern Recognition. Fundamentals of Algorithms, Society for Industrial and Applied Mathematics, 224 pp.

  • Green, A. A., M. Berman, P. Switzer, and M. Craig, 1988: A transformation for ordering multispectral data in terms of image quality with implications for noise removal. IEEE Trans. Geosci. Remote Sens., 26, 65–74.

    • Search Google Scholar
    • Export Citation

  • Hotelling, H., 1936: Relations between two sets of variates. Biometrika, 28, 321–377.

  • Jolliffe, I. T., 2002: Principal Component Analysis. 2nd ed. Springer, 487 pp.

  • Mecikalski, J. R., W. M. MacKenzie, M. Koenig, and S. Muller, 2010: Cloud-top properties of growing cumulus prior to convective initiation as measured by Meteosat Second Generation. Part I: Infrared fields. J. Appl. Meteor. Climatol., 49, 521–534.

    • Search Google Scholar
    • Export Citation

  • Müller, J., 2010: MSG level 1.5 image data format description. EUMETSAT Doc. EUM/MSG/ICD/105, 127 pp. [Available online at http://www.eumetsat.int/groups/ops/documents/document/pdf_ten_05105_msg_img_data.pdf.]

  • Nielsen, A. A., 2011: Kernel maximum autocorrelation factor and minimum noise fraction transformations. IEEE Trans. Image Process., 20, 612–624.

    • Search Google Scholar
    • Export Citation

  • Nielsen, N. W., 2008: Et regnskyl af dimensioner: Skybruddet ved Gråsten i Sønderjylland den 20. august 2007 (A severe downpour: The cloudburst at Gråsten in Southern Jutland on August 20, 2007). Vejret,114 (1).

  • Schmetz, J., P. Pili, S. Tjemkes, D. Just, J. Kerkmann, S. Rota, and A. Ratier, 2002: An introduction to Meteosat Second Generation (MSG). Bull. Amer. Meteor. Soc., 83, 977–992.

    • Search Google Scholar
    • Export Citation

  • Schölkopf, B., A. Smola, and K.-R. Müller, 1998: Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput., 10, 1299–1319.

    • Search Google Scholar
    • Export Citation

  • Schowengerdt, R., 2007: Remote Sensing: Models and Methods for Image Processing. 3rd ed. Academic Press, 515 pp.

  • Vestergaard, J. S., 2011: Improved nowcasting of heavy precipitation using satellite and weather radar data. M.Sc. thesis, DTU Informatics, Technical University of Denmark, 177 pp.

  • Wackernagel, H., 1995: Multivariate Geostatistics: An Introduction with Applications. Springer, 256 pp.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 156 59 8
PDF Downloads 72 33 0