Virtual edge method for point cloud slicing in 3D printing is a very common algorithm due to the advantage of low computational complexity. However, this approach is not robust to density variations. To overcome this shortcoming, in this paper, a concept of column neighborhood is introduced to estimate whether the virtual edge model should be used in certain situations. Concretely, based on the technic of generating a set of multiple contours, we present an improved algorithm for high resolution point cloud slicing. To deal with multiple contours, a classify method based on convex hull is proposed, which reduces considerable calculations and separates each module from the others. At the last step, a method based on minimum angle is proposed to generate a single contour for each class. This algorithm is verified to be efficient by testing on kinds of typical benchmarks.
展开▼
