Curve and Surface Fitting Algorithm for Measurement Data

doi:10.12068/j.issn.1005-3026.2021.03.016

Abstract

Abstract: The moving least squares (MLS) method is widely used in curve and surface fitting due to its good approximation performance. However， the fitting accuracy is extremely unstable when processing data with gross error.In order to reduce the effect of gross error on the fitting accuracy， a moving least trimmed squares (MLTS)method was proposed. In this method， the least trimmed square (LTS) method was introduced in the influence domain to replace the least square (LS) method， and the optimal group of nodes without abnormal data was selected among all the nodes to determine the local fitting coefficient.Assigning weights or setting threshold values artificially is unnecessary， which avoids the influence of subjective operations. Numerical simulation and experimental data processing showed that the gross error of measurement data can be handled effectively， and the fitting results of the MLTS method are better than those of the MLS method， which has good fitting accuracy and robustness.

Key words: curve and surface fitting; gross error; moving least squares(MLS); least trimmed square (LTS); local fitting

CLC Number:

TP391.9

GU Tian-qi， LUO Zu-de， HU Chen-jie， LIN Shu-wen. Curve and Surface Fitting Algorithm for Measurement Data[J]. Journal of Northeastern University(Natural Science), 2021, 42(3): 408-413.

References

[1]崔青玲，李长生，刘相华，等.模拟平板轧制的新方法——无网格伽辽金法［J］.东北大学学报(自然科学版)，2005，26(5):463-466.(Cui Qing-ling，Li Chang-sheng，Liu Xiang-hua，et al.A new method to simulation of plate rolling:element-free Galerkin method (EFGM)［J］.Journal of Northeastern University(Natural Science)，2005，26(5):463-466.)
[2]Mirzaei D.A greedy meshless local Petrov-Galerkin method based on radial basis functions［J］.Numerical Methods for Partial Differential Equations，2016，32(3):847-861.
[3]Liu H S，Xing Z W，Yang Y Y.Simulation of sheet metal forming process using reproducing kernel particle method［J］.International Journal for Numerical Methods in Biomedical Engineering，2010，26(11):1462-1476.
[4]Lancaster P，Salkauskas K.Surface generated by moving least square methods［J］.Mathematics of Computation，1981，37:141-158.
[5]程玉民，李九红.弹性力学的复变量无网格方法［J］.物理学报，2005，54(10):4463-4471.(Cheng Yu-min，Li Jiu-hong.A meshless method with complex variables for elasticity［J］. Acta Physica Sinica，2005，54(10):4463-4471.)
[6]Cheng Y M，Li J H.A complex variable meshless method for fracture problems［J］.Science China Physics，Mechanics & Astronomy，2006，49(1):46-59.
[7]崔小曼，于凤芹.利用Tikhonov正则化改进移动最小二乘的图像变形算法［J］.激光与光电子学进展，2019，56(23):100-106.(Cui Xiao-man，Yu Feng-qin.Moving least squares based image deformation algorithm improved with Tikhonov regularization［J］.Laser and Optoelectronics Progress，2019，56(23):100-106.)
[8]Wang Q，Zhou W，Cheng Y G，et al.Regularized moving least-square method and regularized improved interpolating moving least-square method with nonsingular moment matrices［J］.Applied Mathematics and Computation，2018，325:120-145.
[9]Wang F J，Qu W Z，Li X L.Augmented moving least squares approximation using fundamental solutions［J］.Engineering Analysis with Boundary Elements，2020，115:10-20.
[10]秦亚光，罗周全，汪伟，等.采空区三维激光扫描点云数据处理技术［J］.东北大学学报(自然科学版)，2016，37(11):1635-1639.(Qin Ya-guang，Luo Zhou-quan，Wang Wei，et al.Cavity three-dimensional laser scanning point cloud data processing technology［J］.Journal of Northeastern University(Natural Science)，2016，37(11):1635-1639.)
[11]朱广堂，叶珉吕.基于曲率特征的点云去噪及定量评价方法研究［J］.测绘通报，2019(6):105-108.(Zhu Guang-tang，Ye Min-lyu.Research on the method of point cloud denoising based on curvature characteristics and quantitative evaluation［J］.Bulletin of Surveying and Mapping，2019(6):105-108.)
[12]李睿，林海荣，吴小燕.基于稳健移动最小二乘法的点云数据拟合［J］.测绘与空间地理信息，2017，40(5):122-124.(Li Rui，Lin Hai-rong，Wu Xiao-yan.Point cloud data fitting based on robust moving least square method［J］.Mapping and Spatial Geographic Information，2017，40(5):122-124.)
[13]Amir A，Levin D.Quasi-interpolation and outliers removal［J］.Numerical Algorithms，2018，78(3):805-825.
[14]Fleishman S，Cohenor D，Silva C T，et al.Robust moving least-squares fitting with sharp features［J］.ACM Transactions on Graphics，2005，24(3):544-552.
[15]Mili L，Cheniae M G，Rousseeuw P J，et al.Robust state estimation of electric power systems［J］.IEEE Transactions on Circuits and Systems Irregular Papers，1994，41(5):349-358.
[16]Torti F，Perrotta D，Atkinson A C，et al.Benchmark testing of algorithms for very robust regression:FS，LMS and LTS［J］.Computational Statistics & Data Analysis，2012，56(8):2501-2512.
[17]Beliakov G，Kelarev A V，Yearwood J，et al.Derivative-free optimization and neural networks for robust regression［J］.Optimization，2012，61(12):1467-1490.
[18]Mount D M，Netanyahu N S，Piatko C D，et al.A practical approximation algorithm for the LTS estimator［J］.Computational Statistics & Data Analysis，2016，99:148-170.
[19]Adedia D，Adebanji A，Labeodan M，et al.Ordinary least squares and robust estimators in linear regression:impacts of outliers，error and response contaminations［J］.British Journal of Mathematics & Computer Science，2016，13(4):1-11.
[20]Tong G W，Chen H Y，Liu S，et al.Memetic reconstruction algorithm for the ECT［J］.Iet Science Measurement & Technology，2018，12(7):917-924.
[21]Koch I E，Veronez M R，Silva R M，et al.Least trimmed squares estimator with redundancy constraint for outlier detection in GNSS networks［J］.Expert Systems with Applications，2017，88:230-237.
[22]Zhang H Q，Guo C X，Su X F，et al.Measurement data fitting based on moving least squares method［J］.Mathematical Problems in Engineering，2015(6):1-10.