Accuracy improvement of point data reduction with sampling-based methods by Fuzzy logic-based decision-making

Abstract

Modern 3D-digitization systems, which are employed by reverse engineering (RE), feature the ever-growing scanning speed, with the ability to generate large quantity of points in a unit of time. Generally speaking, that is advantageous for the quality and efficiency of RE modelling. However, huge number of point data, generated in the course of 3D-digitization, can turn into a serious practical problem, later on, when the CAD model is generated. Having this in mind substantial research effort has been focused towards development of methodologies for reduction of point data which is the result of 3D-digitization. The analysis of point data reduction using sampling methods revealed several problems among which the most prominent two are lack of feedback information on the effect of the reduction of a particular point data on the deviation of the resulting cross-sectional curve and insufficient efficiency of decision-making procedures. With this in mind, proposed in this article is an approach which is based on analysis of deviation (caused by point data reduction) and fuzzy logic-based decision-making. This approach significantly advances the quality of sampling-based point data reduction, which was demonstrated on comparative analysis of a obtained experimental results.