期刊名称：Walailak Journal of Science and Technology (WJST)
印刷版ISSN：2228-835X
出版年度：2017
卷号：14
期号：6
页码：527-546
DOI：10.14456/vol14iss4pp%p
语种：English
出版社：Institute of Research and Development, Walailak University.
其他摘要：The capability of a computer system to interpret intelligible handwritten data input and analyze it for many automated process system is the core of a handwriting character recognition system. Handwritten recognition plays a major role in many applications, such as cheques verification, office automation, natural human-computer interaction, as well as mail sorting. Procedures of handwritten numeral recognition have showed a remarkable contribution in order to improve the automation process and have developed human and machine interaction in several fields and applications. In this paper, the proposed approach is based on extraction of both local and global geometric features. The features are further quantified into a set of facts or conditions that are used for classification based on a set of rules as an expert system. In addition, sample handwritten images are tested with the proposed technique and the results are stated and accuracy rate is calculated based on a confusion matrix. The output showed that the recognition error rate in terms of False Rejection Rate (FRR) is 4 % or the overall successful accuracy is 96 %.
关键词：Arabic numeral character recognition; image processing; pattern recognition; feature extraction; object segmentation; expert system
