RECONSTRUCTION OF SURFACE FOR IMAGE PROCESSING

In this paper we are going to use a physically motiva ted method for surface reconstruction that can recover smooth surfaces from noisy and sparse data sets. No orientation information is required. By a new technique based on regularized - membrane potentials the input sample points are aggregated, leading to i mproved noise tolerability and outlier removal, without sacri ficing much with respect to detail (feature) recovery. In this method, sample points are first aggregated on a volumetric grid. A labeling algorithm that relies on intrinsic properties of the smooth scalar field which emerging after the aggregation, is us ed to classify grid points as exterior or interior to the surface. We also introduce a mesh - smoothing paradigm based on a mass - spring system, enhanced with a bending - energy minimizing term to ensure that the final triangulated surface is smoother than piece wise linear. The method compares favorably with respect to previous approaches i n terms of speed and flexibility