BSC PROJECT SCREENING FORM

BSc Project Screening Form

BSc Project Screening Form

Student Name

Student Number

Degree Pathway

Supervisor Name

Title of Project

Surface Fitting using Artificial Neural Networks

Description of your artefact

In this paper, a neural network-based algorithm is proposed to explore the sequence of the measured point data for surface fitting. In CAD/CAM, the ordered data serves as the input to fit smooth surfaces so that a reverse engineering system can be established for 3D sculptured surface design. The geometry feature recognition capability of back-propagation neural networks is also explored. Scan number and 3D coordinates are used as the inputs of the proposed neural networks to determine the curve which a data point belongs to and the sequence number of the data point on the curve. In the segmentation process, the neural network output is segment number; while the segment number and sequence number on the same curve are the outputs when sequencing those points on the same curve. After evaluating a large number of trials, an optimal model is selected from various neural network architectures for segmentation and sequence. The proposed model can easily adapt for new data measured from the same part for a more precise fitting surface. In comparison to Lin et al.'s [Lin, A. C., Lin, S.-Y., & Fang, T.-H. (2008). Automated sequence arrangement of 3D point data for surface fitting in reverse engineering. Computer in Industry, 35, 149- 173] method, the presented algorithm neither needs to calculate the angle formed by each point and its two previous points nor causes any chaotic phenomenon of point order.

There is a review on data acquisition techniques, characterization of geometric models and related surface representations, segmentation and fitting techniques in reverse engineering (Varady, Martin,& Cox, 2010: 255). From theCMMpoint of view, it is difficult to fetch surface information rapidly through CMM and the massive point data obtained can barely be processed directly (Yau et al., 2009: 236). The initial point data acquired by a measuring device generally require pre-processing operations such as noise filtering, smoothing, merging and data ordering for subsequent operations. Using the pre-processed point data, a surface model can be generated by a curve fitting and surface fitting method.

A set of heuristics is used to break compound features into simple ones using an NN (Nezis & Vosniakos, 2010: 144). Freeform surfaces from Bezier patches are reconstructed by simultaneously updating networks that correspond to the separate patches (Knopf & Kofman, 2009: 125). A back-propagation network is used to segment surface primitives of parts by computing Gaussian and mean curvatures of the surfaces (Alrashdan, Motavalli, & Fallahi, 2009: 179). A neural network self-organizing map (SOM) method is employed to create a 3D parametric grid and reconstruct a B-spline surface (Barhak & Fisher, 2009, 2008: 105).

What methodology (structured process) will you be following to realise your artefact?

In reverse engineering, a designer firstly places the existing product in either a CMM or a laser scanner for surface configuration measurement. And the acquired point data can be input into CAD software to establish curves and ...