摘要
针对二值文本纹理复杂、信息隐藏的免疫力差的特点,提出了一种基于DCTDWT的二值文本频率域数字水印算法。算法结合了DCT和DWT二者的特点,将水印信号进行分块DCT变换,按照从低频到高频的顺序对各块系数进行重新排列,对应嵌入到原始图像的DWT分解子带中,强化了不可见性。为了验证算法的鲁棒性和不可见性,利用相似度原理设计了水印提取检测器,通过将嵌入水印后的图像进行JPEG压缩、加入高斯噪声、椒盐噪声以及将其剪切掉一块的方法来模仿图像在实际网络环境下可能遭受的各种攻击和干扰,仿真实验证明,在各种情况下,检测器均有较好的检测效果。
Digital watermarking techniques have been considered as an effective solution to protecting the copyright of multimedia data. Considering that binary texts featured complicated texture and weak immunity, the problem of how to embed the watermark into the binary text is investigated and a new algorithm that combines DWT (Discrete Wavelet Transform) and DCT (Discrete Cosine Transform) domain is proposed in this paper. The watermark signal is transformed into the DCT domain and its coefficients of every block are rearranged according their frequency from high to low. Then those rearranged coefficients are inserted into the sub bands of original images DWT domain correspondingly. The experimental results show that the watermarks are invisible and robust against noise and commonly used image-processing methods such as Guassian, Salt, Pepper noise, JPEG, crop procession etc. The algorithm offers a feasible method to protect and authenticate the binary texts such as check, bargain, certificate, ID (identification) and other valuable documents.
出处
《系统仿真学报》
CAS
CSCD
2004年第3期521-524,共4页
Journal of System Simulation
基金
陕西省教育厅产业化培育项目(02JC40)
陕西省自然科学研究计划基金资助(2001X29)。
