Once pre-processing phase has finished, the hue and saturation values for each selected pixel are use to
infer the model, that is, ( ,..., ) 1 n x x x r r r = , where n is the number of samples and a sample is ( , ) i i i x = h s r . As
a result of a testing and comparing phase with several statistical models such as mixture of gaussians or
discrete histograms, the best results have been obtained using a Gaussian model. The values for the
parameters of the Gaussian model (mean, x , and covariance matrix,Σ ) are computed from the sample set
using standard maximum likelihood methods [9]. Once they are found, the probability that a new pixel,
x = (h,s) r , is skin can be calculated as
Once pre-processing phase has finished, the hue and saturation values for each selected pixel are use toinfer the model, that is, ( ,..., ) 1 n x x x r r r = , where n is the number of samples and a sample is ( , ) i i i x = h s r . Asa result of a testing and comparing phase with several statistical models such as mixture of gaussians ordiscrete histograms, the best results have been obtained using a Gaussian model. The values for theparameters of the Gaussian model (mean, x , and covariance matrix,Σ ) are computed from the sample setusing standard maximum likelihood methods [9]. Once they are found, the probability that a new pixel,x = (h,s) r , is skin can be calculated as
การแปล กรุณารอสักครู่..
