الفهرس | Only 14 pages are availabe for public view |
Abstract Our community has seen significant changes as a result of digital information transfer storage, and processing. Information can be stolen, duplicated, or interacted with. Actions might be random or planned. The digital information may be audio, image, or video. Robust image watermarking seeks to include invisible information into videos, generally for copyright protection purposes. Watermarking is the technique of embedding data, a watermark, within audio, image, or video object. Text, image, or icon may all be used as watermarks. The embedded watermark can later be recognised or extracted to authenticate the host object’s identification. The video watermark system is designed to insert potentially invisible information, often for copyright protection applications into videos.This thesis aims to develop a robust watermarking algorithm. The watermark should be resistant to a wide variety of image and video processing attacks. An approach to robust image watermarking techniques termed the second generation watermarking started from an idea from existing watermarking systems, but extended to further research to reach better results and satisfy the needs of watermarking systems. This class of watermarking schemes increases robustness against some types of image and video processing attacks by using the Contourlet Transform. Furthermore, watermarking algorithm is implemented with the aid of the Scale Invariant Feature Transform (SIFT) and K-Means Clustering due to many advantages relevant to image processing applications, especially in Key-Frames extraction that is used to embed the watermarking to reduce the data capacity of the embedded data. The combination between CT, SIFT, and K-Means Clustering has succeeded to embed and extract the watermark correctly will be discussed in full detail. This work achieves the need of maintaining good video quality and introduces solutions to video watermarking applications. |