Title :
Contour analysis using time-varying autoregressive model
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., George Washington Univ., Washington, DC, USA
Abstract :
Contour modeling by a time-varying autoregressive (TVAR) model is considered. A least squares estimator of the TVAR model parameters is presented, and the maximum likelihood approach for determining the model order is also presented. In the experiment, curvature extrema points of synthesized contours are detected from the time frequency distribution estimated with TVAR model. In the classification experiment with contours of various planar shapes, about 97% of samples are correctly classified.
Keywords :
autoregressive processes; image classification; least squares approximations; maximum likelihood estimation; time-frequency analysis; TVAR model parameters; classification experiment; contour analysis; curvature extrema points; least squares estimator; maximum likelihood estimation; model order; planar shapes; synthesized contours; time frequency distribution; time-varying AR model; time-varying autoregressive model; Covariance matrix; Frequency estimation; Gaussian processes; Maximum likelihood detection; Maximum likelihood estimation; Parameter estimation; Polynomials; Shape; Testing; Time frequency analysis;
Conference_Titel :
Image Processing, 2000. Proceedings. 2000 International Conference on
Conference_Location :
Vancouver, BC, Canada
Print_ISBN :
0-7803-6297-7
DOI :
10.1109/ICIP.2000.899857
