Segmentation of Connected Symbols in Online Handwritten Chemical Formulas

In this paper, we present a novel approach to separate connected symbols in online handwritten chemical formulas. The proposed method combines both the chemical structural characteristics and online information. The main process can be divided into three stages: preprocessing, location of connected point, and segmentation. In addition, organic ring, a common chemical structure, is much different from other chemical symbols. We adopt the Freeman chain code to separate ring from chemical bond when they are connected. Experiment results show that the proposed method can not only satisfy the real-time demand, but also achieve a high correct segmentation rate of 92.7%.