一种允许误差的异构图形语义匹配技术

Algorithm for Error-Tolerant Semantic Matching of Heterogeneous Graphics

为解决二维和三维异构图形数据匹配问题,以流程工厂协同设计为背景,提出了基于允许误差的最大语义匹配算法(MSMGE).该算法首先用类无向图描述二维和三维图形数据的工程属性和拓扑关系,消除了研究对象的异构性;然后,用语义表达式表示类无向图顶点与边的语义关系,将二维和三维异构图形数据的匹配转化为允许误差最大公共序列的匹配;最后,通过采用基于工程语义的类无向图语义分割、语义规整和语义裁剪等方法降低MSMGE的复杂度.研究表明,用此算法,匹配精度能够达到90%以上.

In order to realize matching of 2D and 3D heterogeneous graphic data,an algorithm,named as MSMGE,for error-tolerant semantic matching of heterogeneous graphic data was presented based on the computer supported cooperative work in plant design.With this algorithm,the undirected-graph-like is used to represent the engineering properties and topological relations of 2D and 3D graphic data to smooth away the heterogeneity of the graphic data.And then,the semantic expression is adopted to describe the semantic relations among vertexes and edges in the undirected-graph-like to transform the semantic matching of heterogeneous graphic data into largest common subsequence error-tolerant matching.Lastly,the graphic partitioning based on engineering semantics,the semantic normalization and the semantic pruning are adopted in order to reduce the complexity of MSMGE.The research result shows that with the presented algorithm,the matching precision can exceed 90%.