Previous |  Up |  Next


mathematical discourse; OCR
We present a summary of our work in progress related to mathematical formulae recognition. Our approach is based on the structural construction paradigm and two-dimensional grammars. It is a general framework and can be successfully used in the analysis of images containing objects exhibiting rich structural relations. In contrast to most of all other known approaches, the method does not treat symbols segmentation and structural analysis as two separate processes. This allows the system to solve arising ambiguities more reliably. We have already implemented pilot studies for the off-line as well as on-line mathematical formulae recognition showing that the proposed method can be effectively implemented and practically used.
1. Chan, K. F., Yeung, D. Y.: Mathematical expression recognition: a survey. International Journal of Document Analysis and Recognition 3 (2000) 3–15.
2. Schlesinger, M., Hlaváč, V.: Ten lectures on statistical and structural pattern recognition. Volume 24 of Computational Imaging and Vision. Kluwer Academic Publishers, Dordrecht, The Netherlands (2002). Zbl 1087.62505
3. Průša, D.: Two-dimensional Languages. Ph.D. thesis, Faculty of Mathematics and Physics, Charles University, Prague (2004).
4. Průša, D., Hlaváč, V.: 2D context-free grammars: Mathematical formulae recognition. In: Holub, J., Ždárek, J., eds.: Proceedings of the Prague Stringology Conference ’06, Prague, Czech Republic, Prague Stringology Club, CTU (2006) 77–89.
5. Průša, D., Hlaváč, V.: Mathematical formulae recognition using 2D grammars. In: Proceedings of the 9$th$ International Conference on Document Analysis and Recognition. Volume II, Curitiba, Brazil (2007) 849–853.
6. Průša, D., Hlaváč, V.: Structural construction for mathematical on-line formulae recognition. In: Proceedings of the 13th Iberoamerican Congress on Pattern Recognition, Havana, Cuba (2008) accepted, to appear.
7. Savchynsky, B., Schlesinger, M., Anochina, M.: Parsing and recognition of printed notes. In: Proceedings of the conference Control Systems and Computers, Kiev, Ukraine (2003) 30–38, in Russian, preprint in English available.
8. Kiyko, V.: Recognition of objects in images of paper based line drawings. In: Third International Conference on Document Analysis and Recognition, Montreal (1995) 970–973.
Partner of
EuDML logo