Gray Gradient Co-occurrence Matrix Based On The Maximum Entropy Estimate Of Blind Video Watermarking Algorithm

With the development of mulitimedia and Internet technology, there are more and more digital products for our entertainment and it can be easily copied and delivered on Internet without any loss in quality. More and more pirates strongly hurt the producer and author's interests, so effective method is required to protect the digital products. Digital watermarking is such an effective method and it attracts many researchers, especially digital video watermarking. In recent years there are many digital watermarking techniques for popular video compression standards.In this paper, some images and video watermarking techniques are introduced, including their development and general principle. Comparison and analysis of image and video watermarking is also made. This paper deals with streaming video watermarking and its exclusive requirement and proposed a blind video watermarking scheme for object-based video coding standard MPEG-4 based on maximum entropy estimation.The original video image is preprocessed by GGCM maximum entropy estimation to dynamically select the most suitable frame, location and intensity for embedding digital watermark, and the DCT transform of watermarking is used for all selected video frames to improve watermark capacity, and keep robust to MPEG-4 compression and decompression, .