A Method for Recognizing and Processing Geometric Inconsistencies in Multi-source Identical Polygonal Vector Entities Based on Machine Learning
In the fusion and updating of multi-source vector data,there are significant geometric positional differ-ences between identical entities,leading to a low degree of automation in geometric inconsistencies recognition and processing.A geometric inconsistency recognition and processing method for multi-source identical polygonal vector entities is proposed in this paper.Firstly,an in-depth analysis is conducted on the classification of geometric in-consistency features,and a recognition model based on geometric inconsistency feature indices of identical polygo-nal entities is constructed by using machine learning methods.Then,the point set registration algorithm is intro-duced for entity alignment to achieve geometric consistency processing of identical polygonal entities.The effective-ness of the proposed method for multi-source hydrological data are tested in Zhoushan area.The results demonstrate that this method has high accuracy in identifying geometric inconsistencies,and the geometric consistency process-ing results are superior to those of existing direct overlay operation methods.Additionally,this method can effec-tively reduce positional errors and improve geometric consistency.
万方数据
维普
