论文部分内容阅读
本文提出了一种用于字符预分类的模糊逻辑分析法,对文本字符作印刷结构分析,给出了一个带有客差分析的文本行字符基线精确测定算法,其它有效参考线则通过聚类分析获得.模糊逻辑用于确定各字符类的隶属值以保证字符的正确预分类.实验结果表明,我们的模糊印刷字符预分类法在SUN4/490工作站上每秒可有效处理104以上字符,并对不同大小的字符有满意的结果.
In this paper, we propose a fuzzy logic analysis method for pre-classification of characters, make a printed structure analysis of text characters, give an accurate baseline determination algorithm for text lines with passenger analysis, and other effective reference lines through clustering Analysis obtained. Fuzzy logic is used to determine the membership of each character class to ensure the correct pre-classification of characters. The experimental results show that our fuzzy typographic pre-tax classification method can effectively process more than 104 characters per second on the SUN4 / 490 workstation and have satisfactory results for characters of different sizes.