作者: Ming Chang , Shi Han , Dongmei Zhang
关键词: Graph (abstract data type) 、 Representation (mathematics) 、 Directed graph 、 Symbol (chemistry) 、 Search problem 、 Computer science 、 Theoretical computer science 、 Context model 、 Set (abstract data type) 、 Graph
摘要: Chemical expressions have more variant structures in 2-D space than that math equations. In this paper we propose a unified framework for recognizing handwritten chemical including both inorganic and organic expressions. A set of novel statistical algorithms is presented two key components framework: symbol grouping structure analysis. Non-symbol modeling inter-group are proposed to achieve better result, bond group the special symbols framework. graph-based representation (CESG) defined representing generic expressions, analysis problem formulated as search CESG over weighted direction graph. Experiments on database 35,000 were conducted results presented.