Research And Implementation Of The Volume Calculation Of Cattle Based On Point Cloud

Beef cattle industry in national economic construction is of great significance, it is the industry that closely related to the people’s livelihood. As an important criterion for the evaluation of beef cattle production, meat production is urgent needed to achieve automatic determination. Volume parameter as the basic geometric properties of cattle, it is an important prerequisite for the study of the correlation between them. To solve this problem, this paper mainly used the computer graphics technology study the method of volume calculation based on point cloud, and adopted a method of tetrahedral generation to realize the automatic division of cattle point cloud. Finally this paper optimized the tetrahedral mesh, and achieved the calculation of cattle volume. The main study of this paper includes the following aspects:(1) Initial tetrahedralization: By comparing and analyzing the advantages and disadvantages of several common generation algorithm of tetrahedral mesh, the Delaunay triangulation method was deeply studied. Based on the computational geometry theory of triangulation, this paper studied the concepts and criterions of the convex hull, Voronoi graph and Delaunay triangulation and researched their relationships. On the basis of the incremental algorithm and the local transformation method, the incremental transformation algorithm was proposed to generate the initial tetrahedral mesh. The stochastic perturbation method was introduced to improve the quality of the tetrahedral element based on it.(2) Boundary recovery: Due to the mesh was convex hull after generating the initial tetrahedral mesh, it is necessary to realize the surface boundary recovery of cattle under the condition of adding boundary constraints. Boundary recovery mainly included edge recovery and surface recovery. In this paper, the method of adding auxiliary points on the missing edge was used to achieve edge recovery; the surface recovery was to reconstruct and mark the generated mesh through the cavity, and then eliminate the in vitro tetrahedral. The constrained tetrahedral mesh can be obtained and the accuracy of volume calculation can be improved through the boundary recovery.(3) Mesh optimization: According to the method of measuring the quality of tetrahedral element, the size and quality of tetrahedral mesh can be controled. This paper removed the sliver tetrahedrons by the decomposition method, and generated the homogenized tetrahedral mesh through the local transformation of adjacent elements and the mesh smoothing technique. According to the optimization of tetrahedral mesh, the volume of each tetrahedral of cattle was calculated by the determinant method, the sum of the tetrahedral volume was the volume of the whole cattle.The algorithm was designed and implemented by using the simplified data structure. The experimental results show that the average accuracy of cattle volume in this algorithm is 91.4%, the error range is less compared with the accuracy of the geometric object with regular shape. It can be seen that the algorithm in this paper has better adaptability to the cattle body with complex surface characteristics, and has higher accuracy of the calculation of the volume.