data segmentation error and wrong fitting geometric elements
are evaluated respectively
and then a method to decrease the uncertainties is proposed for the curve fitting of parts. Firstly
based on Monte Carlo method
the uncertainty of fitting parameters is studied by adding noise to fitting data
using 0-means Gaussian distribution as segmentation error model
and using circular arc to fit Bezier curve. The results show that segmentation error and misuse of geometric elements have a great influence on the uncertainty
which is much greater than the influence of data noise. Secondly
in order to segment the data sequence accurately
high-dimensional vector mapping is proposed to test the constancy of local features of elements
identify the geometric elements in the data sequence
and find out the range of the data segment. Simulation and experiment results show that the proposed method can obtain more accurate data segmentation and fitting than traditional methods.
TANG Hao, DENG Zhaohui, WAN Linlin, et al. Application of cam lift fitting method based on alternative-point interpolation in grinding process [J]. Journal of Mechanical Engineering, 2012, 48(23): 191-199.
FORBES A B. Parameter estimation based on least squares methods [M]∥ Data Modeling for Metrology and Testing in Measurement Science. Boston, USA: Birkhäuser Boston, 2009: 1-30.
YU Jianwu, LUO Hong, HU Junzhi, et al. Reconstruction of high-speed cam curve based on high-order differential interpolation and shape adjustment [J]. Applied Mathematics and Computation, 2019, 356: 272-281.
LU Lizheng, ZHAO Shiqing. High-quality point sampling for B-spline fitting of parametric curves with feature recognition [J]. Journal of Computational and Applied Mathematics, 2019, 345: 286-294.
PEI S C, HORNG J H. Circular arc detection based on Hough transform [J]. Pattern Recognition Letters, 1995, 16(6): 615-625.
HE Wenhua, LIU Zhijing, QU Jianming. A visual tracking method based on global constrained Hough model [J]. Journal of Xi'an Jiaotong University, 2014, 48(12): 69-73.
ROTH G, LEVINE M D. Geometric primitive extraction using a genetic algorithm [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994, 16(9): 901-905.
XU Guangshen, GAO Xiaoding, XU Weixing. A new method for computing the radius of curvature of a cam profile with oscillating flat-faced follower [J]. Mechanical Science and Technology, 2002, 21(2): 204-206.
TIRANDAZ H, NASRABADI A, HADDADNIA J. Curve matching and character recognition by using B-spline curves [J]. International Journal of Engineering and Technology, 2011, 3(2): 183-186.
GÁLVEZ A, IGLESIAS A. Memetic improved cuckoo search algorithm for automatic B-spline border approximation of cutaneous melanoma from macroscopic medical images [J]. Advanced Engineering Informatics, 2020, 43: 101005.
LIU Feng, ZHANG D, SHEN Linlin. Study on novel curvature features for 3D fingerprint recognition [J]. Neurocomputing, 2015, 168: 599-608.
ALEXANDRU M B. Fitting data using optimal Hermite type cubic interpolating splines [J]. Applied Mathematics Letters, 2012, 25: 2047-2051.
WEI Jinwen. The dimensionality reduction of feature vectors by generalized cross product [C]∥2008 International Symposium on Computer Science and Computational Technology. Piscataway, NJ, USA: IEEE, 2008: 281-284.
FU Huan, LIANG Li, WANG Fei, et al. A point cloud segmentation algorithm using local convexity and octree [J]. Journal of Xi'an Jiaotong University, 2012, 46(10): 60-65.