This paper seeks to resolve the key problems of data processing in the research area of Reverse Engineering. The data obtained by different approaches have different characteristics. The key technologies and feature extraction are analyzed.At first, the function frame model of data processing is established to guiding the whole research project. Then the smooth data filtering of surface interpolation of scattered points are put forward by the author. After analyzing the different requirement of points reduction, the NURBS surface reduction and the Quad-Tree structure-based method of scattered points are presented. The specific algorithms of each are given. The coordination transformation and data merging method are also discussed. The approach feature extraction is presented based on the analyzing of the geometric features of points sets, Such as points lines etc. Combined the procedure and algorithm of feature extraction, the data blocking based on the feature frame is proposed. |