Computer-aided reverse engineering using simulated evolution on NURBS

Abstract

The current study is related to the issue of computer-aided reverse engineering. Although the proposed technique has been suggested and adopted for the image-based planar objects, it is extendable to the objects in three dimensions (3D) with some modifications. The objective of the present study is to provide a robust and automated solution and methodology for the applications of reverse engineering technology in manufacturing and other industries. By extracting nearly exact geometric data of the models, the key points (one may call them 'corner points') from the data are detected using an efficient technique. Then curve or surface models are fitted by using non-uniform rational Bsplines (NURBS). In curve or surface fitting problems, the selection of appropriate parameters, in order to obtain an optimized shape design, is a well-known problem. For large data, this problem needs to be dealt with by optimization algorithms avoiding possible local optima and at the same time reaching the desired solution in an iterative fashion. Many evolutionary optimization techniques including genetic algorithm and simulated annealing have been applied to the problem. This paper presents an application of an evolutionary heuristic technique known as 'simulated evolution' (SimE) to the curve and surface fitting problems using NURBS. NURBS have become the standard for representing curves and surfaces. This paper presents a method for approximating NURBS using digitized data. Instead of using the classical optimization approach to solve the problem, a new method based on SimE is developed. The shape parameters, in the description of NURBS, have been targeted to be optimized in a best possible way. The paper describes the mapping scheme of the problem to SimE followed by the proposed algorithm outlines with the interesting results obtained.

A hybrid surfacing methodology for reverse engineering

“…The patches (or loops) can be drawn on the polygon model manually, semi-automatically or automatically based on a target patch count and the curvature of a model. Although most rapid surfacing packages provide an automatic method to generate patch structures quickly, manual organisation of the patch structure is always required to obtain the optimal patch layout because, in many cases, features such as edges and primitives cannot be recognised and the patches do not adequately define the surface curvature (Varady and Facello 2005).…”

“…Here we will categorise the commercial reverse modelling software from a different viewpoint, i.e., the modelling methodology; and according to the different reverse modelling methodologies used, the software for RE can be classified into two categories: classic surfacing and rapid surfacing. A classic surfacing package creates surfaces in the traditional point-curve-surface manner (Clark 2002, Sarfraz 2006, as shown in Figure 2. Feature lines must be extracted from the point cloud manually based on the user's judgement first, to segment the point cloud into several partitions.…”

Intelligent Approaches for Vectorizing Image Outlines

“…Designing and modeling some appropriate curve scheme [9][10][11][12][13][14][15][16][17][18][19][20][21][22][23] is one of the important phases of capturing and vectorizing outlines of images. It plays a significant role in various applications.…”

“…For example, scaling, shearing, translation, rotation and clipping operations can be performed without any difficulty. Although a good amount of work has been done in the area [8][9][10][11][12][13][14][15][16][17][18][19][20], it is still desired to proceed further to explore more advanced and interactive strategies. Most of the up-to-date research has tackled this kind of problem by curve subdivision or curve segmentation.…”

Gravitational Search Algorithm for NURBS Curve Fitting

“…To solve the problem, a novel method based on simulated evolution was developed instead of classic optimization approach [24].…”