Robust crypto-watermarking approach based on spherical harmonics and AES algorithm for 3D mesh safe transmission
The wide range of applications of 3D meshes and the birth of very high-speed networks for storing 3D models in remote multimedia databases have made the copyright protection of these objects a necessity. In order to contribute to safe sharing, transfer and indexing of 3D multiresolution meshes, a ne...
Gespeichert in:
Veröffentlicht in: | Multimedia tools and applications 2022-11, Vol.81 (27), p.38543-38567 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The wide range of applications of 3D meshes and the birth of very high-speed networks for storing 3D models in remote multimedia databases have made the copyright protection of these objects a necessity. In order to contribute to safe sharing, transfer and indexing of 3D multiresolution meshes, a new crypto-watermarking approach is proposed based on spherical harmonics and AES encryption algorithm. Data to be inserted, in this case, is a grayscale image already encrypted using AES256 algorithm. The host mesh, to be watermarked, is decomposed using spherical harmonics to extract frequency coefficients and to compute optimal order for reconstruction. These coefficients undergo two watermarking iterations. During each round, modulation, insertion using quantification method and demodulation steps are implemented. Finally, reconstruction of watermarked mesh using a minimal number of coefficients is conducted. The results established show clearly that our algorithm protects the mesh quality even with the insertion of a large amount of information (a whole image) while ensuring real time execution. The application of various attacks (noise addition, coordinate quantization, smoothing, translation, rotation, uniform scaling and compression) to a watermarked mesh did not prevent the correct retrieval of inserted information. Our approach presents then an enhancement compared to existing ones. |
---|---|
ISSN: | 1380-7501 1573-7721 |
DOI: | 10.1007/s11042-022-13113-6 |