Irregular object measurement method based on improved adaptive slicing method

Surface shape feature is a very important index for monitoring objects. However, in the existing slicing methods, the volume measurement accuracy of point cloud is easy to be low due to the problems of indistinguishable or fuzzy boundary and static slicing. In order to solve this problem, an improved method based on dynamic adaptation is proposed on the basis of slice method, which is suitable for calculating the volume of point cloud on irregular object surface when there are multiple contour boundaries. First, after the point cloud preprocessing, the rough and fine two-step clustering method is used to separate the multiple contour boundaries in the slice, and then the adaptive mechanism is introduced to achieve adaptive slicing and calculate the point cloud volume. In addition, the bounding box algorithm is used to calculate the two-dimensional parameters (length, width and height) of the point cloud data, and the surface area is calculated by generating the grid surface of the point cloud through the greedy projection triangulation measurement algorithm. Using multiple sets of data to calculate their volume by comparing algorithms, we find that the error in this paper is reduced. In addition, the validity of the two-dimensional parameters is verified by comparing the calculated value with its theoretical value. The results show that the method is effective and accurate in segmentation of multi contour boundary, and can effectively calculate the surface shape features of point cloud, and has certain application effects.