Surface Reconstruction Based On Point Cloud Data Of Objects By Ultrasonic Detection

With the development of industrial and medical needs, requirements that ultrasound probe can be solved "three-confirm" problem. The detecting objects accurately determine its size, location and nature. The solution to this problem must be by surface reconstruction of point cloud data to complete and the point cloud data is from ultrasonic detection of space objects.This paper analyzes the characteristics of point cloud data with ultrasonic probe, there is an designed algorithm of the grid provided. It works for space partition of the point cloud data; it can help searthing neiborhood points with much less time.In general, the point cloud data contains multiple objects, the data are mixed together, this paper designs different objects separation algorithm for point cloud data. This can greatly reduce the search volume when make based-graphics with the point cloud data. And is also a foundation to determine the size, location and nature of the detection object.Basic graphical of point cloud data is crucial to the reconstruction process of the surface. This paper designs a new algorithm of ball field graphical on based side. Basic primitives by this algorithm are that point, segment, triangle. And, it has a few characteristics, for example, wide applicability, rapid-speed, small memory consumption, and so on.OpenGL is a professional three-dimensional graphics language. In this paper, it has completed the final step in surface reconstruction. Use OpenGL assembling with the basic primitives into three-dimensional graphics. In the program, the added space-roaming function, the viewer can fully observe the graphics in the perigon's orientation.The data and graphics from running program meets the design requirements, there is a high degree of consistent point cloud data between the output graphics and input. Those prove that design of the algorithm and the program is correct.