Segmentation of point clouds using smoothness constraints

摘要: For automatic processing of point clouds their segmentation is one the most important processes. The methods based on curvature and other higher level derivatives often lead to over segmentation, which later needs a lot manual editing. We present method for using smoothness constraint, finds smoothly connected areas in clouds. It uses only local surface normals connectivity can be enforced either k-nearest or fixed distance neighbours. presented requires small number intuitive parameters, provide tradeoff between underand over-segmentation. application algorithm industrial shows its effectiveness compared approaches.