基于向量线段的特殊字符识别

VECTOR SEGMENT-BASED SPECIAL CHARACTER RECOGNITION

传统的OCR技术在某些特定方面,如印刷体数学公式中特殊字符的识别存在困难和限制,其识别结果的准确率不尽如人意。为此,重点介绍基于向量线段的特殊字符识别算法规则和识别系统的设计。提出通过提取字符中的向量线段进行特征比较的分析方法,并将噪点去除算法融入其中。实验表明,该方法对于特殊字符的分析识别具有较好的准确性和应用前景。

The accuracy rate of recognition outcome of traditional OCR technology is still unsatisfactory in some special areas,for example there are difficulties and limitations in recognising special characters in typeset mathematics formulae.Therefore,this paper deliberately introduces the rule of special character recognition algorithm based on vector segment and the design of recognition system.The analysis approach presented in this paper is to compare the features of vector segments extracted from the characters,and the noise removal algorithm is fused into it.Experiments indicate that this method has good accuracy and application prospects for the analysis and recognition of special characters.