Unsupervised Video Frame Clustering Approach for Video Watermarking Using SIFT Features and Contourlet Transform

Resumen

With the remarkable development of the Internet and the exchange and transmission of data
through digital imaging means, it has become necessary to preserve the property rights of this
data. Watermarks have become an ideal solution to preserve ownership. Hence, watermarks can
be defined as a seal, signature, or mark placed within the data such as images, audio, and video
clips to show the rights of ownership of the data. Unlike many technologies that seek to hide or
encrypt content, the watermark appears as an icon embedded within the multimedia so that it
does not affect its quality and guarantees its accuracy, and is not affected by attempts to delete,
steal or copy. There are many areas in which the watermark can be integrated, both visible and
invisible. In this research, a sensitive technique was applied to hide digital watermarks inside
videos to achieve randomness in hiding and comprehensiveness of all parts of the video to protect
them from attack. The watermark should be robust against different image-processing attacks. In
this paper, we introduced robust video watermarking algorithms for copyright protection.
Elementary models of algorithms that are used for video watermarking are shown, and the
assignment measurements for both robustness and imperceptibility are also calculated. To help
new researchers to understand the current watermarking algorithms rapidly, approximately basic
quantitative measurements of several watermarking features are analyzed and compared. Last,
we explore the problems in the study of robust video watermarking systems and propose a new
path of development.