期刊名称：International Journal of Engineering and Computer Science
印刷版ISSN：2319-7242
出版年度：2015
卷号：4
期号：4
页码：11233-11236
出版社：IJECS
摘要：Optical Character recognition which is the original method of character recognition many times gives poor recognition rate dueto error in character segmentation. Segmentation is a very important task in every OCR system. OCR system separates the scannedhandwritten image text documents into lines, words and characters. The accuracy of OCR system mainly depends on segmentationalgorithm and noise removal technique being used. Segmentation of handwritten Devnagari text is difficult when compared with printedDevnagari or printed English or any other printed document, because of its structural complexity and increased character set. It containsvowels, consonants and half form consonant. This system addresses the segmentation of handwritten Devnagari text document, the mostpopular script of Indian sub – continent into lines, words and characters. Mainly artificial neural network technique is used to design topre-process, segment and recognize Devnagari characters.
关键词：Devnagari Character Recognition; Offline Handwritten Recognition; Segmentation; Feature Extraction
