Stroke Correspondence Based on Graph Matching for Detecting Stroke Production Errors in Chinese Character Handwriting
People may make mistakes in writing a Chinese character. In this paper, we apply error-tolerant graph matching to find the stroke production errors in people’s handwriting of Chinese characters. A set of edit operations to transform one graph into another
- PDF / 518,913 Bytes
- 10 Pages / 430 x 660 pts Page_size
- 19 Downloads / 177 Views
Department of Computer Science and Technology, University of Science & Technology of China, Hefei, China 2 Joint Research Lab of Excellence, CityU-USTC Advanced Research Institute, Suzhou, China 3 Department of Computer Science, City University of Hong Kong, Hong Kong, China [email protected], [email protected], [email protected]
Abstract. People may make mistakes in writing a Chinese character. In this paper, we apply error-tolerant graph matching to find the stroke production errors in people’s handwriting of Chinese characters. A set of edit operations to transform one graph into another are defined for achieving this purpose. The matching procedure is denoted as a search problem of finding the minimum edit distance. The A* algorithm is used to perform the searching. Experiments show that the proposed method outperforms existing algorithms in identifying stroke production errors. The proposed method can help in Chinese handwriting education by providing feedback to correct users who have stroke production errors in writing a Chinese character. Keywords: Chinese handwriting education, stroke production error, graph matching, graph edit distance, moment function.
1 Introduction Chinese characters have been part of the Chinese culture for several thousands of years. Chinese characters are ideograms instead of letters in an alphabetic system. Each Chinese character has its own structure formed by the strokes that should be written in the correct position, proportion and order. While people write the Chinese characters they may have some handwriting errors [1]. The major stroke production errors can be divided into four types (Fig. 1): (1) Missing stroke error; (2) Concatenated stroke error; (3) Extra stroke error; (4) Broken stroke error. Although there are other kinds of handwriting errors such as stroke order error or badly written strokes etc., in this paper we focus on detecting the stroke production errors. In order to learn how to write a Chinese character properly, the student should be corrected if he/she makes handwriting errors. It is thus very important to know whether and where exactly the errors are. The nature of this Chinese character handwriting education problem makes it different from Chinese character recognition mentioned in [2]. The character recognition is focused on finding the similarity between the input character and a set of candidate characters, and then classifying the input into one of the H.H.S. Ip et al. (Eds.): PCM 2007, LNCS 4810, pp. 734–743, 2007. © Springer-Verlag Berlin Heidelberg 2007
Stroke Correspondence Based on Graph Matching
735
candidate character with the highest similarity. On the other hand, for the Chinese handwriting education system, it is necessary to find a detailed matching between the input character and a known template character in order to find out the exact difference.
Missing (a) Template character
Extra Broken (b) Errors in sample character
Concatenated
Fig. 1. Four types of production errors
There are three main types of Chinese handwr
Data Loading...
