Character Recognition Based on Attribute-Dependent Programmed Grammar
IEEE Transactions on Pattern Analysis and Machine Intelligence
Twenty Years of Document Image Analysis in PAMI
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.14 |
We report a pattern description and generation method of structural characters, and show some examples of Chinese and Korean character patterns actually generated. We also consider its educational and graphical applications. In this method, any character is regarded as a composite pattern constructed by several simpler subpatterns, and is described in terms of them by introducing three kinds of positional relationships among them. A composite pattern can become a subpattern, too. We call these patterns blocks. Syntactic grammar is defined to encode a pattern expression by two code strings, that is, a string of blocks and a string of production rules. They are used in generating patterns, namely, derivation of pattern expression from the code strings is defined as a process of pattern generation. By this description and generation method, we can encode structural characters without much memory, and generate natural shapes of patterns.