.png){kind=icon}
返回
A video watermark algorithm based on tensor feature map
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-022-14299-5.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Video has become one of the main ways of information transmission with the development of the Internet. Video copyright protection becomes an urgent task. Video watermark technology embeds copyright into the redundant information of the carrier, and video copyright protection is achieved. However, most video watermark algorithms do not use the correlation and redundancy among adjacent frames of a video and are weak to resist frame attacks. In order to make up this shortage and improve robustness, a video watermark algorithm based on a tensor feature map is proposed. A grayscale video segment with the same scene is selected and represented as a 3-order tensor, a high-order singular value decomposition is performed on the video tensor to obtain a stable core tensor and three factor matrices. A feature tensor is obtained by the mode-3 product of the video tensor with the transpose of the factor matrix that contains a time axis. It is called a tensor feature map. Since the tensor feature map contains the main information of each frame of a video, the watermark is distributed in each frame of a video by embedding the watermark into the tensor feature map. The first-order discrete wavelet transform and discrete cosine transform are used to embed the watermark into the tensor feature map. The experimental results show that the proposed watermark algorithm based on the tensor feature map has better transparency and is robust to common video attacks.
Keyword:
Video watermark
High-order singular value decomposition
Tensor feature map
Feature tensor
期刊
IF:
3
论文数:
1.9W
被引数:
3.2W
机构
引用论文
.png){kind=icon}
An Evaluation of an Unstructured and Structured Approach to Increasing Recess Physical Activity
PREAI
分享收藏Technique for gray-scale visual light and infrared image fusion based on non-subsampled shearlet transform基于非下采样shearlet变换的灰度可见光与红外图像融合技术
PREAI
分享收藏Tensor Completion via Generalized Tensor Tubal Rank Minimization Using General Unfolding通过使用一般展开的广义张量输卵管秩最小化来完成张量
PREAI
分享收藏
查看更多内容