PCA and FCM clustering algorithm-based reversible watermark embedding and extracting method for three-dimensional model

Abstract

The invention relates to a PCA and FCM clustering algorithm-based reversible watermark embedding and extracting method for a three-dimensional model. According to the invention, the traditional watermark embedding method is combined, and innovation and optimization are carried out on the basis, so that the efficiency is greatly improved. According to the method, coordinate system reconstruction iscarried out on a three-dimensional model through a traditional PCA algorithm, partitioning processing is carried out on vertexes through FCM clustering, watermarks are embedded for multiple times through grouping, and the purpose of the method is achieved. Experimental results prove that the method can be used for effectively embedding and extracting the watermark of the three-dimensional model.Besides, compared with a traditional method, the method has the advantages that reversibility is better processed, translation resistance, rotation resistance, scaling resistance and clipping resistance of the three-dimensional model are greatly improved and are greatly superior to those of a previous algorithm, and data capacity is improved by nearly one order of magnitude.

Description

technical field [0001] The invention aims at the field of embedding and extracting the watermark of the three-dimensional model, especially proposes a method based on the PCA algorithm and the FCM algorithm for embedding and extracting the watermark. For the three-dimensional model with a large amount of data, the method has higher processing efficiency. Background technique [0002] With the development of computer graphics, games, virtual reality and other fields, the application of three-dimensional objects is becoming more and more extensive, followed by more and more infringements of illegal possession, copying, modification and dissemination of unauthorized products. Thereby people's legal rights such as the ownership of their products, the right to modify, etc. are severely damaged. As a new information security protection measure, digital watermarking technology has become a research hotspot in the field of three-dimensional objects. [0003] Due to the complexity a...

AI Technical Summary

Problems solved by technology

Due to the huge workload of building the grid itself, coupled with the huge amount of point model data, the embedding and extraction algorithm of the 3D point model watermark has disadvantages such as large amount of calculation, slow embedding and extraction speed, etc.

Second, it is difficult to transform the 3D point model into the frequency domain for operation

Some operations do not make any changes to the properties of the 3D model, but seriously affect the detection of watermarks, such as translation, rotation, uniform / non-uniform scaling, vertex reordering, resampling, etc.

These attack operations bring great difficulties to 3D model watermark detection

Images