A representation model and modeling method of 3D reconstruction object supporting product redesign

In this paper, an approach is proposed to solve the redesign problem of 3D reconstructed object in reverse engineering. The conventional approach is to adopt simple analytical surface, B-splines surface, or NURBS directly fitting to measured points of object. However this representation is difficult in changing object shape, or making a redesign. Our approach is to select parameterization features, to represent the reconstructing model. The reconstruction processes is composed of three steps. In the first step, geometric features are recognized and extracted based on a features-matching algorithm, and simultaneously the constrriints between these features are also detected. In the second step, these features are adjusted according to constraints relationship between features to obtain an optimal fitting of the measured data points. Finally, the features are applied to parameterized representation, and a cooperative deform relationship of the whole shape is established. Thus, the reconstructing model, composed of geometric features, can be randomly changed by change feature size (size drive) to meet our requirement of making product shape redesign and innovative design. The proposed approach is illustrated with examples.