采用Logistic混沌系统的三维流程工厂模型数字水印算法

Digital Watermarking for 3D Process Plant Models Based on Logistic Chaotic System

针对已有水印算法在精确保持CAD模型几何和拓扑信息能力方面的不足,基于改进的Logistic混沌系统,提出一种面向流程工厂参数化实体模型的鲁棒性盲水印算法.该算法利用模型的拓扑表示选择对偶点为嵌入载体;利用改进的Logistic混沌映射对对偶点进行分组,通过修改对偶点的大小对同一分组重复嵌入相同信息;利用对偶点对中对偶点的相对大小关系实现水印的盲检测.理论分析和实验结果表明,该算法无须修改模型的几何和拓扑信息,对剪切、添加、缩放和网格简化等攻击具有较强的鲁棒性.

This paper presents a novel and robust blind watermarking algorithm for 3D parametric process plant models based on an improved logistic chaotic system. Instead of modifying the CAD model＇s geometry or topology information, which is mostly used by existing algorithms, our approach embeds the watermarks into the model＇s dual points by making full use of the model＇s topology information. Group operation of dual points is processed based on the improved logistic chaotic system, and the same information is embedded into the points of the same group by slightly modifying the point＇s size. The watermarks can be blindly extracted according to the relationship between the sizes of each pair of dual points. Theoretical analysis and experimental results show that the algorithm can resist many attacks such as cropping, adding, scaling and mesh simplification without modifying the model＇s geometry and topology information.