Intelligent Multiple Watermarking Schemes for the Authentication and Tamper Recovery of Information in Document Image

Abstract

Existing works on document image watermarking provide same level of protection for information present in the source document image. However, in a document image the distribution of the information contents influences on the level of protection required. This necessitates application of multiple watermarking techniques on the source document image. In this paper, novel intelligent multiple watermarking techniques are proposed. The source document image is divided into blocks of the same dimension. For each block, appropriate type of watermarking is decided based on the type of block which is determined automatically using gradient binarized technique. The blocks with regeneratable information are protected using semi-fragile watermarking and blocks with non-regeneratable information are protected using fragile watermarking. Experimental results reveal the accurate identification of type of the block. The performance results reveal that multiple watermarking schemes have reduced the capacity of embedding and consequently improved perceptual quality of the watermarked image.