In this paper, a method for classifying 3D unorganized points into interior and boundary points using a deep neural network is proposed. The classification of 3D unorganized points into boundary and interior points is an important problem in the nonuniform rational B-spline (NURBS) surface reconstruction process. A part of an existing neural network PointNet, which processes 3D point segmentation, is used as the base network model. An index value corresponding to each point is proposed for use as an additional property to improve the classification performance of the network. The classified points are then provided as inputs to the NURBS surface reconstruction process, and it has been demonstrated that the reconstruction is performed efficiently. Experiments using diverse examples indicate that the proposed method achieves better performance than other existing methods.

In this paper, we propose a method for registering unorganized point clouds without using targets or markers. Motivated by the 4-points congruent sets (4PCS) algorithm, which is a nontarget-based registration method commonly used in the related fields, we develop a feature-based 4PCS algorithm (F-4PCS). The method combines the basic approach of the 4PCS algorithm with geometric feature information to produce consistent global registration results efficiently. We use the features from the point feature histogram descriptor and the ones that capture the surface curvature. The experimental results show that the proposed method successfully registers point clouds of both the outdoor and indoor scenes and demonstrates better performance than the existing 4PCS-based registration methods.

No abstract

This article introduces an artificial neural network (ANN) model to determine cycle-times for forming curved hull plates when the target shape is known. The proposed model aids shipbuilding companies in predicting the cycle-times required for ship fabrication. The input parameters are geometric information extracted from the target shape (curvedness, Gaussian curvature, width, and height of the hull plate), and the output parameter is the heating duration per unit area. The structure of the proposed model, which predicts cycle-times for line heating after the cold forming case, consists of two hidden layers. The proposed model is convenient to use and flexible because it only requires retraining when the dataset is changed. The performance of the proposed model was analyzed by five-fold cross-validation and compared with that of a mathematical model obtained from the linear regression analysis method and predefined formulas. The results show that the ANN model is reliable and accurate for the cycle-time prediction of curved hull plates in shipbuilding applications. Introduction Shipbuilding companies generally estimate the production cost of a ship based on their previous ships for various purposes before the production planning department begins to optimize the fabrication process. They use the estimated value to refine the overall fabrication process or improve it by reducing unnecessary tasks and maximize the overall production efficiency.