Research on Precise Constraint Reconstruction of Section Data

Abstract

Since the features are not prominent and the algorithm is complex during whole reconstruction of section data, a step-by-step optimization reconstruction method is proposed. The order of reconstruction is optimized: line features are reconstructed firstly; Then arc features are reconstructed (constraints are satisfied between the arc and known lines); finally freeform features are reconstructed (constraints are satisfied between the B-spline and known lines/arcs). In this way, the reconstruction accuracy of the line/arc features is ensured in the first. Since freeform features have more freedom, it is convenient to be adjusted to meet the more constraints. Linear boundary constraints are constructed and the algorithm becomes optimization problem of the quadratic objective function under the linear constraints. The examples show that the reconstruction accuracy is improved greatly under satisfying constraints; the expected goal is achieved in real application.