改进的重采样算法研究

Research on Improved Re-Sampling Algorithm

针对逆向工程中点云数据过多导致拟合曲面光顺性、精度等方面难以达到要求的问题,通过分析对比多种重采样算法的特点,提出一种改进的重采样算法--在区域划分基础上的混合采样法。该算法首先利用曲率估算算法计算各点曲率,然后根据曲率特征将点云数据划分为特征区域和平坦区域,接着对特征区域进行曲率采样,对平坦区域进行均匀采样。通过实验表明,该算法能够有效地减少点云数据,保留细节特征防止出现空白区域,最终获得较高曲面拟合精度。

In reverse engineering, too much cloud point will make it diffwult to meet the requirement of manufacturing, like the smooth and the accuracy of the fitting surface and so on. Aimed at solving these problems, an improved re-sampling algorithm is put forward by analyzing and comparing with different kind of re -sampling algorithm, which is named mixed sampling method based on regional division. Firstly, it is need to use curvature estimation algorithm to calculate the curvature of each point. Secondly, cloud point will be divided into two regions namely feature region and fiat region, because of different characteristic about curvature. Finally, feature region will be sampled with curvature sampling algorithm, fiat region will be sampled with uniform sampling algorithm. According to the experiment, it is proved that the improved re-sampling algorithm is effective in reducing the number of cloud point and saving detail characteristic of mold and avoiding the appearance of blank area and attaining better accuracy.