An Improved Method for NURBS Free-Form Surface Based on Discrete Stationary Wavelet Transform

• Main Authors: Deming Kong, Xiaoqiang Tian, Dehan Kong, Xiaodan Zhang, Li Yuan
• Format: Article
• Language: English
• Published: IEEE 2020-01-01
• Series: IEEE Access
• Subjects: 3D measurement; reverse engineering; discrete stationary wavelet transform; NURBS free-form surface
• Online Access: https://ieeexplore.ieee.org/document/9060928/

In order to improve fitting accuracy of free-form surface, an improved method for the currently recognized NURBS free-form surface based on discrete stationary wavelet transform is proposed. Firstly, the elevation image of the free-form surface point cloud and the elevation sequence of the elevation image are obtained by using the grid method. Then, feature points of the free-form surface are extracted by the discrete stationary wavelet transform. Finally, using the extracted feature points, high accuracy NURBS free-form surface is realized. Compared with the conventional NURBS method and the NURBS method based on dual-tree complex wavelet transform, the experimental results showed that the root mean squared errors of fitting results of free-form surface were deduced 77.64% and 23.16%, respectively. Therefore, the proposed method has satisfactory fitting precision.