Hi Luis,<br><br>Thanks very much for the answer and the reference! However, after the manual of itk::MultivariateLegendrePolynomial seems to suggest that this function works only for 2D and 3D data. Although it's possible to add one dummy dimension (eg. all zero), I'm wondering whether there's a function more suitable for 1D signal.
<br><br>Best,<br>- Karen<br><br><div><span class="gmail_quote">On 6/6/07, <b class="gmail_sendername">Luis Ibanez</b> <<a href="mailto:luis.ibanez@kitware.com">luis.ibanez@kitware.com</a>> wrote:</span><blockquote class="gmail_quote" style="margin-top: 0; margin-right: 0; margin-bottom: 0; margin-left: 0; margin-left: 0.80ex; border-left-color: #cccccc; border-left-width: 1px; border-left-style: solid; padding-left: 1ex">
<br>Hi Karen,<br><br><br>You can use the Multivariate Legendre Polynomials,<br>and combine them with the Linear Kalman estimator,<br>or with the Levenberg-Marquard optimizer.<br><br><a href="http://www.itk.org/Insight/Doxygen/html/classitk_1_1MultivariateLegendrePolynomial.html">
http://www.itk.org/Insight/Doxygen/html/classitk_1_1MultivariateLegendrePolynomial.html</a><br><a href="http://www.itk.org/Insight/Doxygen/html/classitk_1_1KalmanLinearEstimator.html">http://www.itk.org/Insight/Doxygen/html/classitk_1_1KalmanLinearEstimator.html
</a><br><a href="http://www.itk.org/Insight/Doxygen/html/classitk_1_1LevenbergMarquardtOptimizer.html">http://www.itk.org/Insight/Doxygen/html/classitk_1_1LevenbergMarquardtOptimizer.html</a><br><br><br>In both cases you will be using the sum of squared
<br>differences between your data and the polynomial,<br>as the metric to minimize.<br><br><br>Please look at the recent trail by Mathieu Malaterre<br>on this topic in the users list.<br><br><br><br> Regards,<br><br><br>
Luis<br><br><br>--------------------<br>Karen Guan wrote:<br>> Dear all,<br>><br>> I'm working on fcMRI processing, and need polynomial fitting (3rd order,<br>> ax^3 + bx^2 + cx +d) for each time course (with about 600 time points)
<br>> to remove B0 fluctuation or shifting. The entire data set<br>> has 128 * 128 * 7 samples ( i.e. time courses).<br>><br>> The questions are:<br>> 1. Is there such an algorithm in ITK (including vnl/vcl)?
<br>> 2. If so, for fast processing of 1-D signal with 600 samples, what are<br>> be best choices?<br>><br>> I appreciate the help!<br>><br>> - X.<br>><br>><br>><br>> ------------------------------------------------------------------------
<br>><br>> _______________________________________________<br>> Insight-users mailing list<br>> <a href="mailto:Insight-users@itk.org">Insight-users@itk.org</a><br>> <a href="http://www.itk.org/mailman/listinfo/insight-users">
http://www.itk.org/mailman/listinfo/insight-users</a><br></blockquote></div><br>