The study on the detection and recognition of the printed formula
Yonghua Li, Kejun Wang, Qinchang Guo, Junfei Wu · 2008
At present, the widely applied OCR system has a high recognition for printed texts, however it doesn't have a good recognition schema for formulas. In order to locate and recognize printed formulas accurately, this paper conducted researches according to the following several aspects: use Adaboost method to locate formulas automatically, image processing to obtain thinned formula area, moment method to revise formulas, domain method to segment formulas, template feature method to recognize characters, and two dimensional structural analysis based on character feature to recognize formulas. The method presented in this paper can achieve a good recognition.