Max-Min Approach for Watermark Text Extraction in Documents

Abstract

Online digital content authentication and copyright protection is a crucial issue. A Digital watermarking concept offers a comprehensive authentication and copyright protection for online documents. In contrast to images, audio, and video, text is the most common form of Internet communication and needs to be completely protected. Digital watermark text is a process of adding the watermark text information on online documents to protect from unauthorized access. The extraction of watermark from documents helps to know the quick information about respective document. In this research work the max-min clustering algorithm is employed to increase the semantic gap between the watermark text pixels and background text pixels. Detection process determines the watermark text and actual location of the watermark text in the documents. Thresholding system and post processing heuristic technique is employed to extract the actual watermark text region from the image. The results depicts that proposed model is promising and encouraging.