Title :
A novel method for recognition of handwritten chemical symbols based on MQDF
Author :
Yang, Jufeng ; Wang, Kai ; Li, Mingda ; Xu, Jing
Author_Institution :
Coll. of Inf. Tech. Sci., NanKai Univ., Tianjin, China
Abstract :
Chemical symbols recognition is a base component in online handwritten chemical formulas recognition which has been paid more and more attention. This paper adopts 108 chemical symbols which are comprised of inorganic chemical symbols, organic ring structures, chemical bonds and laboratory apparatus. The recognizing process is divided into four steps. The first is adjusting and optimizing the handwritten data. Then, eight-directional features are extracted from each trajectory point. In the third stage, the features are optimized by maximum filter and Gaussian filter. The MQDF classifier is used for recognition of handwritten chemical symbols at last. Experimental results show that the proposed method is robust and flexible.
Keywords :
Gaussian processes; chemical engineering computing; feature extraction; filtering theory; handwriting recognition; handwritten character recognition; image classification; Gaussian filter; MQDF classifier; chemical bonds; eight directional feature; handwritten chemical symbol recognition; handwritten data; inorganic chemical symbol; laboratory apparatus; maximum filter; online handwritten chemical formula recognition; organic ring structures; Character recognition; Chemicals; Educational institutions; Feature extraction; Handwriting recognition; Structural rings; Writing; MQDF; eight-directional features; handwritten chemical symbols;
Conference_Titel :
Image and Signal Processing (CISP), 2011 4th International Congress on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-9304-3
DOI :
10.1109/CISP.2011.6100458
