Multiresolution compression and feature line reconstruction for Reverse Engineering

In this paper we analyze applications of Douglas-Peucker algorithm to compression and feature line extraction of 3D models from digital scans introducing a multiresolution model that enables a fast description of the object at different levels of details through the use of a unique rearrangement, stored in a file, of the input data set. The multiresolution algorithm, which reaches high data compression rates while producing fair approximation, is used to simplify each point of the input data set according to its scale, its basic shape type and its geometric parameters. Finally, object feature lines are reconstructed using shape information previously found. The proposed algorithm is able to segment the object into meaningful patches, which represent form features of the object, resulting in an important tool for reverse engineering and rapid prototyping.