GMMB_FRAC2LHOOD Map density quantiles to PDF threshold values lhood = GMMB_FRAC2LHOOD(histS, f) histS K-element histS cell array created by gmmb_hist or gmmb_generatehist. lhood N x K array of likelihood values. f N x K array of density quantile values This function finds the likelihood threshold value corresponding to each density quantile value. For each column k in 1..K, the likelihood value is found from histS{k}, so that each column may represent a different distribution. See gmmb_hist, gmmb_generatehist, gmmb_lhood2frac, gmmb_fracthresh References: [1] Paalanen, P., Kamarainen, J.-K., Ilonen, J., Kälviäinen, H., Feature Representation and Discrimination Based on Gaussian Mixture Model Probability Densities - Practices and Algorithms, Research Report 95, Lappeenranta University of Technology, Department of Information Technology, 2005. Author(s): Pekka Paalanen <pekka.paalanen@lut.fi> Jarmo Ilonen <jarmo.ilonen@lut.fi> Joni Kamarainen <Joni.Kamarainen@lut.fi> Copyright: Bayesian Classifier with Gaussian Mixture Model Pdf functionality is Copyright (C) 2004 by Pekka Paalanen and Joni-Kristian Kamarainen. $Name: $ $Revision: 1.2 $ $Date: 2005/04/14 10:33:34 $
0001 %GMMB_FRAC2LHOOD Map density quantiles to PDF threshold values 0002 % 0003 % lhood = GMMB_FRAC2LHOOD(histS, f) 0004 % 0005 % histS K-element histS cell array created by gmmb_hist or 0006 % gmmb_generatehist. 0007 % lhood N x K array of likelihood values. 0008 % f N x K array of density quantile values 0009 % 0010 % This function finds the likelihood threshold value corresponding to 0011 % each density quantile value. 0012 % For each column k in 1..K, the likelihood value is found from histS{k}, 0013 % so that each column may represent a different distribution. 0014 % 0015 % See gmmb_hist, gmmb_generatehist, gmmb_lhood2frac, gmmb_fracthresh 0016 % 0017 % References: 0018 % [1] Paalanen, P., Kamarainen, J.-K., Ilonen, J., Kälviäinen, H., 0019 % Feature Representation and Discrimination Based on Gaussian Mixture Model 0020 % Probability Densities - Practices and Algorithms, Research Report 95, 0021 % Lappeenranta University of Technology, Department of Information 0022 % Technology, 2005. 0023 % 0024 % Author(s): 0025 % Pekka Paalanen <pekka.paalanen@lut.fi> 0026 % Jarmo Ilonen <jarmo.ilonen@lut.fi> 0027 % Joni Kamarainen <Joni.Kamarainen@lut.fi> 0028 % 0029 % Copyright: 0030 % 0031 % Bayesian Classifier with Gaussian Mixture Model Pdf 0032 % functionality is Copyright (C) 2004 by Pekka Paalanen and 0033 % Joni-Kristian Kamarainen. 0034 % 0035 % $Name: $ $Revision: 1.2 $ $Date: 2005/04/14 10:33:34 $ 0036 % 0037 0038 function lhood = gmmb_frac2lhood(histS, f); 0039 0040 if any(f(:)>1 | f(:)<0) 0041 error('Density quantile values must be in the range [0,1].'); 0042 end 0043 0044 lhood = zeros(size(f)); 0045 0046 K = size(f, 2); 0047 0048 for k = 1:K 0049 v = shiftdim(histS{k}); 0050 len_v = length(v); 0051 nf = (1-f(:,k)) .* (len_v-1) +1; 0052 i = floor(nf); 0053 0054 v(len_v+1) = v(len_v); 0055 lhood(:, k) = v(i) + ( nf-i ) .* ( v(i+1) - v(i) ); 0056 end