数字水印
算法
计算机科学
零(语言学)
无损压缩
零阶
整数(计算机科学)
人工智能
谐波
图像(数学)
数学
数据压缩
一级
应用数学
物理
哲学
量子力学
语言学
程序设计语言
作者
Zhiqiu Xia,Xingyuan Wang,Chunpeng Wang,Changxu Wang,Bin Ma,Qi Li,Mingxu Wang,Tingting Zhao
出处
期刊:Applied Intelligence
[Springer Science+Business Media]
日期:2021-05-07
卷期号:52 (1): 607-621
被引量:61
标识
DOI:10.1007/s10489-021-02476-2
摘要
The continuous orthogonal moments of images have become a main tool used in zero-watermarking algorithms in recent years. However, there is a problem that should not be neglected for existing zero-watermarking algorithms, which is known as numerical instability. Because the continuous orthogonal moments used in existing zero-watermarking algorithms are integer-order moments, this problem affects the computational accuracy of such moments and thus affects the performance of zero-watermarking algorithms. Therefore, in this paper, the integer-order radial harmonic Fourier moments (IoRHFMs) are firstly extended to fractional-order radial harmonic Fourier moments (FoRHFMs) to effectively mitigate the problem of numerical instability and improve the computational accuracy of IoRHFMs, and then a FoRHFM-based zero-watermarking algorithm for medical images is proposed to achieve lossless copyright protection of medical images. The experimental results show that this algorithm is highly robust to geometric attacks and common attacks, and is superior to IoRHFM-based and other zero-watermarking algorithms.
科研通智能强力驱动
Strongly Powered by AbleSci AI