Data processing technology of 3D points cloud

This paper presents the methods for data processing technology of 3D points cloud with large unorganized and noisy point sets. Firstly, the noise points will be selected and deleted by the boundary points detection algorithm (BPDA). Secondly, we parameterize the interior points to the sphere model, the octahedron model and the geometry image model. Lastly, we give the error measure of the parameterization. The experiments show that the noisy point sets will be selected well and be deleted. The geometry deformation of our parameterization algorithm is less through the method of the error measure of geometry deformation.