Optical Character Recognition for Digital Aut...

Optical Character Recognition for Digital Authentication Using Unicode Mapping

Author Name : S. Mahalakshmi, Sharon

ABSTRACT Optical Character Recognition (OCR) refers to the process of converting printed text documents into software translated Unicode of the Text. In this paper, we propose a system which processes the given text using the OCR system. Accordingly, the scanned document is stored as an image file and as part of the preprocessing phase the image file is skewed, corrected by a simple rotation technique in the appropriate direction. Then the image is binarized after noise elimination. The preprocessed image is segmented using an algorithm which decomposes the scanned text into paragraphs using special space detection technique (RLSA) and then into lines using vertical histograms, and lines into words using horizontal histograms, and words using horizontal histograms into character picture glyphs. After each image glyph is passed through various routines which extract the characteristics of the glyph, all image glyphs are considered for recognition using Unicode mapping. The image thus obtained along with the Unicode is used to provide security against Cross-site Scripting (XSS), which is the top most vulnerability in the web applications, by using it for authentication purpose. To test the effectiveness of this approach, the vulnerable web inputs listed in research sites are considered. The proposed security system was tested on those vulnerable inputs and has been found that the approach is very effective as it addresses the vulnerabilities at a granular level of tags and attributes, in addition to addressing the XSS vulnerabilities.