Research On Data Process Of Point-Cloud And Virtual Human Body Modeling Based On Human Body

In recent years, the technology of measure system is rapidly improved, and scattered point-cloud data which contain more details of the complex shape object, can be obtained conveniently. The processing technology of 3D point-cloud data is one of key technologies in the research and application of modeling techniques. Under the most circumstance, the scattered data will be transformed into the models of triangular mesh. Then the processing and model reconstruction is conducted on the basis of the triangle mesh. As the scattered data transform into a triangular mesh, the amount of information doubles, and the higher quality of the subsequent processing and the modeling of the mesh is required. It is helpful to improve the ability of the processing and the reconstruction. So, the research of the data processing and model reconstruction from the scattered 3D point-cloud should be valuable and be carried out urgently.Based on the mannequin data from a non-contact 3D body scanner, this paper discussed how to store the point-cloud data, how to do the accurate surface fitting and the reconstruction of the human body model. Specific researches include the improved storage structure of the point-cloud data based on the 2D chain-table, the algorithm of compression and compaction of the data based on mathematical expectation, a new method of the ICP algorithm of data registration based on moving least-squares methods and an improved algorithm of the point set in the plane based on Delaunay triangulation. Finally, they are applied in MATLAB platform and the examples with photos are also given.