Space Sparse Point Cloud Surface Reconstruction Based On Complex Constraints

In recent years, point cloud space reconstruction technology has been greatly developed. It is largely used in many areas such as CAD modeling, geological modeling and so on. With the research in each area going deeper and deeper, many algorithms have been brought up regarding its own area. In geologic surface reconstruction, the main problem is the existence of a variety of geologic constraints and the thin the raw data.In this paper, the problem of sparse point cloud surface reconstruction has been studied with constrained spatial point cloud reconstruction technology. The geologic surface reconstruction method for a variety of different constraints is proposed. The main contents of this article are in the following:1. Proposed a algorithm solves the problem on how to correctly connect curved surface and reconstruct it as a quadrilateral nets and triangle nets curved surface even with reverse fault, vertical fault and normal fault. According to the loose and overlapping properties of original geological data, Delanuany triangle net technology with constraints is used in order to separate the overlapping area between single vale and multi value. After establishing the two-dimensional triangular net topology which is used for separating each area, according to the connectivity properties of surface, generated corrected surface. Through a practical data collecting and reconstruction of original geological data with the algorithm, it is proved that the algorithm is correct and practical.2. With the characteristic of reconstructed topology, the partial subdivision method is proposed in this article based on the subdivision surface method. According to the few of original geological data leading undesirable interpolation and the uneven of reconstruction, subdivision surface method is used. By using discrete Gaussian curvature as scale, this method preserves the original data and modifies the reconstructed topology with smoother surface and less operation quantity while remains same uniform property of nets. Through a real data measurement, the algorithm has been confirmed as practical and correct.3.Complete the algorithm module. According to the characteristics of the actual geological and the two proposed algorithm, a test module has been programmed. The module has tested a variety of different geological work area data and got good results.The work of the design, realization and measurement of this algorithm proves that this algorithm can correctly reconstruct surface using geological data with multi and complex constraints. This algorithm provides a new way of thinking for point cloud surface reconstruction technology.