基于空间认知的等高线树的构建

Spatial Cognition Oriented Contour-Tree Building

等高线空间关系的建立是地貌综合的关键因素之一。等高线之间的邻近、层次等空间关系通常采用等高线树来表达,目前已存在的等高线树的构建方法均是在假设等高线没有断裂的情况下进行,但在实际应用中效果并不理想。文章提出了一种基于空间认知的新的等高线树构建方法,利用空间认知中的抽象思维,通过对等高线之间的相邻、嵌套、层次等关系的分析、比较、综合,抽取出等高线描述地形的规则,并找出断线连接的辅助边,由此确定等高线断线唯一的一种逻辑闭合连接,进而建立等高线树。实验证明,该方法较好解决了在等高线存在断线的情况下无法合理有效建立等高线树的问题,较已有的等高线树构建方法具有更广的适用性和更强的可靠性。

Establishing spatial relationships of contour line is one of the key techniques of automatic generalization. In the past, the neighbor and hiberarchy relationship of contour lines were expressed with contour-tree. However, most current algorithms build contour-trees under the supposition that all contours are well linked up, which are usually noneffective in fact. A new spatial cognition oriented contour-tree building algorithm is put forward in this paper, which picks up the rules of expressing terrain, and finds the auxiliary sides of unconnected lines, so as to confirm the only logical relationships of contours and establish the contour-tree based on the spatial cognition and abstraction etc. Expeiments illustrate that this method is more advanced in building contour-tree under the kind of existing disconnected contours.