This paper proposes an adaptive watermarking algorithm, in which the watermark is a binary image such as a logotype related directly to the image owner. Obviously the information amount of this type of watermark is much larger than the amount of a pseudorandom pattern generated by secret key. This fact makes it difficult to embed a bi-dimensional watermark code into a given image in an imperceptible way. To achieve this goal, we propose an imperceptible adaptive embedding algorithm using the JND (Just Noticeable Difference) criterion. The evaluation results show the desirable features of proposed algorithm, such as the watermark imperceptibility and its robustness against attacks such as: JPEG compression, filtering, and impulsive noise.
CITATION STYLE
Corona-Ramirez, B., Nakano-Miyatake, M., & Perez-Meana, H. (2004). Adaptive watermarking algorithm for binary image watermarks. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3034, pp. 207–216). Springer Verlag. https://doi.org/10.1007/978-3-540-24681-7_23
Mendeley helps you to discover research relevant for your work.