Content Region Detection and Feature Adjustment for Securing Genuine Documents
Cu Vinh Loc, Tran Cao De, Jean-Christophe Burie, Jean-Marc Ogier · 2020
Motivated by increasing the tampering of genuine documents during a transmission over digital channels, we focus on developing a data hiding framework for determining whether a received document is genuine or falsified. The input document is transformed into a standard form to minimize geometric distortions. Fully convolutional networks (FCN) is utilized to detect document's content regions. Next, we construct hiding patterns used for hiding a secret information. Modifying the pixel values of these patterns for carrying secret bits depends on the edge and comer features of document content, and the connectivity of their neighboring pixels. The hiding process is then conducted by changing the ratio between the number of edge features and the number of comer features of subregions within the content regions. The experiments are performed on various binary documents, and our approach gives competitive performance compared to state-of-the-art approaches.