MPS-RR 2001-8
March 2001
In the present paper we propose a flexible continuous parametric shape model for a star-shaped planar object. The model is based on a polar Fourier expansion of the normalized radius-vector function of the object. The expected phase amplitudes are modelled by a simple regression with parameters having simple geometric interpretations. The model is a generalization of first- and second-order Gaussian shape models and is called the generalized p-order model. In particular, non-Gaussian errors are allowed. The statistical analysis is straightforward, as demonstrated on a data set concerning shape discrimination of two cell populations.
Availability: [ gzipped ps
-file ] [ pdf
-file ]
[ Help on down-loading/viewing/printing ]