Linking Broken Characters in Online Handwritten Chemical Formulas

In this paper, we present an effective approach to link the broken characters in online handwritten chemical formulas. By observing a large number of formula samples, we classify the common broken problems in online handwritten chemical formulas into three types, ldquopoint-likerdquo broken type, liner-type and arc-type. The proposed method combines the strokes characteristic of handwritten chemical characters, online information and structural characteristics of chemical formulas. The main process can be divided into three stages: preprocessing, location of broken strokes and linking. Experiment results indicate that the proposed method can not only satisfy the real-time demand, but also achieve a high linking accuracy of 94.0%.