computes the parameters of orthogonal signal correction (OSC) according to the direct method
[ploads,weights,tscores,evpcent]=osc(x_div,ndim);
a matrix (n x q) or a Div structure
the number of dimensions
the loadings or basis of the dropped space
ploads.d is a matrix of dimensions (q x ndim)
the weights
weights.d is a matrix of dimensions (q x ndim)
the scores of the observations for the ndim dimensions
tscores.d is a matrix of dimensions (n x ndim)
the eigenvalues, in percent
evpcent.d is a vector of dimensions (ndim x 1)