文章基本信息

标题：Skew Detection and Correction in the Document Image
本地全文：下载
作者：A.Sakila ; Dr. S.Vijayarani
期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2017
卷号：6
期号：8
页码：17457
DOI：10.15680/IJIRSET.2017.0608274
出版社：S&S Publications
摘要：Document image processing has become a growingly important technology in the field of imageprocessing; it is generally used in historical studies and digital libraries. Document image analysis mainly used forimproving the human readability and the OCR quality of the document. When a paper documents are captured usingcameras or scanned using scanners for digitizing, it often suffers from some level of skew or gradient.After the skewcorrection is done successfully then the character segmentation of the word will be more perfect and as a consequencethe percentage of the correct word recognition will be higher.This paper compared four existing skew correctiontechniques, they are Hough Transformation, Cross Correlation, K- Nearest Neighbour and Fast Fourier transformation.It has been observed that the Fast Fourier transformtechnique correct the skew in the document image and it haveproduced higher accuracy and less execution time compared with other techniques.
关键词：Skew Detection and Correction; Hough Transformation; Cross Correlation; K- Nearest Neighbour; Fast;Fourier transformation.