Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: International Conference on Knowledge - Based and Intelligent Information and Engineering Systems, KES 2017
Год издания: 2017
Идентификатор DOI: 10.1016/j.procs.2017.08.019
Ключевые слова: arc, interpolation, keypoints, line segment, Local warping, motion level, video stabilization
Аннотация: Digital watermarking algorithms have been developed rapidly as a response on the challenges caused by various internet attacks that are distorted the content of the host image and watermark partially or fully. In this paper, the issues of texture analysis with a goal to detect the most suitable image areas for embedding are discussПоказать полностьюed. The statistical and model-based methods are investigated as a trade-off between the computational cost and quality of the detected areas, where the embedded bits of a watermark could be the most invisible for a human vision. The criteria for detection of such areas based on the textural, contrast, illumination, and color coherence of the host image and watermark are formulated. The experiments show that the statistical methods based on the gradient oriented Local Binary Patterns (LBP) provide better computational time regarding to fractal estimation of textural image areas. © 2017 The Author(s).
Журнал: Procedia Computer Science
Выпуск журнала: Vol. 112
Номера страниц: 1460-1469
ISSN журнала: 18770509
Издатель: Elsevier B.V.