Abstract Many computer graphics and computer-aided design (CAD) applications require mathematical models to be generated from measured coordinate data. The three-dimensional surface model of a complex object can be created by fitting the data with low-order parametric surface patches, and adjusting the control parameters such that the constituent patches meet seamlessly at their common boundaries. Most neural networks that are designed for function approximation produce a solution that is not easily transferable to CAD software. An adaptive technique to reconstruct a smooth surface from Bezier patches is presented in this paper. The approach utilizes a neural network that performs a weighted summation of Bernstein polynomial basis functions. The number of basis neurons is related to the degree of the Bernstein polynomials and is equivalent to the number of control parameters. Free-form surfaces are reconstructed by simultaneously updating networks that correspond to the separate patches. A smooth transition between adjacent Bezier surface patches is achieved by imposing positional and tangential continuity constraints on the weights during the adaptation process. The final weights of the various networks correspond to the control points of the stitched Bezier surface, and can therefore be used directly in many commercial CAD packages. This method has been used to create a closed surface by adaptively fitting two adjacent patches to synthetic range data and a complex open surface by fitting four patches to experimentally measured range data.
[1]
Simon Haykin,et al.
Neural Networks: A Comprehensive Foundation
,
1998
.
[2]
Josef Hoschek,et al.
Fundamentals of computer aided geometric design
,
1996
.
[3]
Hyungjun Park,et al.
An adaptive method for smooth surface approximation to scattered 3D points
,
1995,
Comput. Aided Des..
[4]
Pramod N. Chivate,et al.
Review of surface representations and fitting for reverse engineering
,
1995
.
[5]
Chia-Hsiang Menq,et al.
Smooth-surface approximation and reverse engineering
,
1991,
Comput. Aided Des..
[6]
Xue Yan,et al.
Neural network approach to the reconstruction of freeform surfaces for reverse engineering
,
1995,
Comput. Aided Des..
[7]
Ralph R. Martin,et al.
Computer Vision, Models and Inspection
,
1992,
World Scientific Series in Robotics and Intelligent Systems.
[8]
G. W. Vickers,et al.
Free-form surface reconstruction for machine vision rapid prototyping
,
1993
.
[9]
D.R. Hush,et al.
Progress in supervised neural networks
,
1993,
IEEE Signal Processing Magazine.
[10]
David F. Rogers,et al.
Mathematical elements for computer graphics
,
1976
.
[11]
Jake K. Aggarwal,et al.
Model-based object recognition in dense-range images—a review
,
1993,
CSUR.
[12]
Ramesh C. Jain,et al.
Segmentation through Variable-Order Surface Fitting
,
1988,
IEEE Trans. Pattern Anal. Mach. Intell..