摘要
动态图象的知识产权保护越来越重要 .动态图象所固有的特点使其水印技术存在很多问题 .为此提出动态图象运动与非运动区域分块分割的多分辨率自适应水印嵌入和提取算法 ,运动区域与非运动区域分别嵌入不同类型的水印 ,使得水印嵌入过程适应原始动态图象的内容 .水印嵌入和提取的仿真实验表明 ,用该算法嵌入的水印具有很好的鲁棒性和透明性 .无论采用何种类型的动态图象 ,该水印技术对于原始动态图象都是自适应的 ,经各种典型的攻击后 ,至少有一种水印 (独立的或统一的 )可有效地生存 .
Proposes an adaptive watermark embedding algorithm for protection of intellectual property right of motion pictures, which divides motion and motionless regions into non overlapping segments, and embed motion and motionless regions with different kinds of watermarks, and points out this algorithm remains adaptive to the original pictures no matter what type of motion pictures is used, and at least one type of watermark will survive different types of attacks, and concludes from experimental results that the proposed algorithm has excellent robustness and transparency.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2001年第4期499-504,共6页
Journal of Harbin Institute of Technology
