The study of mathematical expression recognition and the embedded system design

Abstract

Mathematical expression recognition is an important problem about pattern recognition, because mathematical expression is an essential part of scientific literature. This paper focused on two aspects, first of all, About the recognition of on-line handwritten mathematical expressions, this paper elaborates the structural analysis strategy from the perspective of the general mathematical expression, we bring forward a feasible solution to the overall structural analysis by using some pretreatment methods, such as dominating relationships, hard-regulations, function merging, and so on. The results show that this method can overcome some shortcomings of traditional structure analysis methods, and also it can dispose some recognition errors that often appear in the space structure and logical structure of the on-line handwriting recognition system effectively, such as symbol overlapping and handshaking, complex layer structure, the table structure and so on. Secondly, on this basis, We have contrived an online handwriting recognizing system of mathematical expression based on an embedded equipment, which supports in the use of the languages of the minority, such as Uyghur and Kazakh languages .For the features of mathematical expression recognizing, and from the point of ensuring the system’s efficiency , this paper has constructed an embedded platform, on which the capability of the system has been tested, and it has compared the data gained from the test with those in PC , which offers a gist to improve the algorithm and perfect the system.