The Arabic Language has a very rich vocabulary. More than 200 million peoplespeak this language as their native speaking, and over 1 billion people use it in severalreligion-related activities. In this paper a new tec...The Arabic Language has a very rich vocabulary. More than 200 million peoplespeak this language as their native speaking, and over 1 billion people use it in severalreligion-related activities. In this paper a new technique is presented for recognizing printedArabic characters. After a word is segmented, each character/word is entirely transformed into afeature vector. The features of printed Arabic characters include strokes and bays in variousdirections, endpoints, intersection points, loops, dots and zigzags. The word skeleton is decomposedinto a number of links in orthographic order, and then it is transferred into a sequence of symbolsusing vector quantization. Single hidden Markov model has been used for recognizing the printedArabic characters. Experimental results show that the high recognition rate depends on the number ofstates in each sample.展开更多
文摘将HMM(Hidden Markov Model)用于手写数字脱机识别,特征提取是一个关键问题.本文首先提出一种新的FT(FourierTransform)特征提取方法一基于和差的一维FT特征,然后将其与另外几种特征进行组合,来提高系统性能,并对多种特征组合中出现的问题进行了研究.在银行票据OCR(Optical Character Reader)系统中的应用表明,本文提出的基于和差的一维FT特征提取方法及多特征组合的方法是有效的.
文摘The Arabic Language has a very rich vocabulary. More than 200 million peoplespeak this language as their native speaking, and over 1 billion people use it in severalreligion-related activities. In this paper a new technique is presented for recognizing printedArabic characters. After a word is segmented, each character/word is entirely transformed into afeature vector. The features of printed Arabic characters include strokes and bays in variousdirections, endpoints, intersection points, loops, dots and zigzags. The word skeleton is decomposedinto a number of links in orthographic order, and then it is transferred into a sequence of symbolsusing vector quantization. Single hidden Markov model has been used for recognizing the printedArabic characters. Experimental results show that the high recognition rate depends on the number ofstates in each sample.
