A Generalized Deep Neural Network Approach for Digital Watermarking Analysis

Technology advancement has facilitated digital content, such as images, being acquired in large volumes. However, requirement from the privacy or legislation perspective still demands need for intellectual content protection. In this paper, we propose a deep neural network (DNN) based watermarking method to achieve goal. Instead of training protecting specific image, train on an image dataset and generalize trained model protect distinct test images bulk manner. Respective evaluations both subjective objective aspects confirm generality practicality our proposed method. To demonstrate robustness general approach, commonly used attacks are applied watermarked examine corresponding extracted watermarks, which retain sufficient recognizable traits some occasions. Testing distinctive shows satisfying generalization method, practice loss function adjustment can cater capacity complicated watermark. We also discuss model, incur vulnerability JPEG compression attack. remedy seeking potentially open window understand underlying working principle DNN future work. Considering its performance economy, it is concluded that subsequent studies work utilizing protection might be promising research trend.