A Case-Based Reasoning Method for Generalizing Depth Contours considering Both Navigational Safety Assurance and Line Shape Preservation

AbstractTo address the problem that the parameters are relatively fixed in the existing automatic methods of depth contour generalization, a case-based reasoning method for generalizing depth contours is proposed considering navigational safety and line shape. First, the structured description of depth contours before and after cartography generalization is made to form case samples. Then, driven by the training samples, the machine learning of BP neural network model is constructed, to obtain the simplification degree taking into consideration the preservation of contour shapes. Finally, the generalization parameters are flexibly adjusted based on the simplification degree obtained through the case-based reasoning, so that depth contours can be adaptively generalized for various complex situations. The experimental results demonstrate that: (1) The case-based reasoning method can make the generalization of depth contours comply with the principle of navigational safety; (2) The case-based reasoning method has a stronger applicability maintaining the shape of depth contour, and is more suitable for the automatic generalization of depth contours, compared with the rolling circle method and the triangulation method. Generally, the case-based reasoning method has the potential to improve cartographic quality meeting the requirements of IHO specification, supporting the automatic production of ENC and nautical chart product.Keywords: Chart generalizationdepth contour generalizationnavigational safety assuranceline shape preservationcase-based reasoning AcknowledgmentsWe would like to thank the editor and anonymous reviewers for their valuable suggestions. We have made some modifications according their suggestions.Disclosure statementNo potential conflict of interest was reported by the authors.This paper is supported by National Natural Science Foundation of China (41901320, 41871369).Additional informationFundingThis paper is supported by National Natural Science Foundation of China (41901320, 41871369).