. The use of TLS for health monitoring of structures is a nondestructive technique, which
allows to obtain the geometry of the object (shape, size and dimensions) even if it is
inaccessible.
. The real possibility of acquiring dense geometric information by TLS allows us to analyse
range dataset through different statistical and modelling approaches not considered until now.
. On the one hand, a parametric surface 3D modelling based on RBF function has allowed us to
carry out an exhaustive structural monitoring applied to a large and complex structure,
especially for fitting surfaces to non-uniformly sampled point-clouds and partial meshes that
contain large irregular holes.
. On the other hand, a novelty statistical approach based on re-weighted extended Procrustes
analysis has been developed to georeference range dataset and to control its accuracy.
Particularly, a modified Danish robust estimator has performed efficiently when a large
number of observations are available, providing more precision and reliability than other
classical methods.
. Therefore, a novelty flowline for health monitoring of complex and large structures which
incorporates a new way of modelling real-world (scanned) objects has been developed.