An Audio Watermarking Method Based On Molecular Matching Pursuit

Mathieu Parvaix (1) , Sridhar Krishnan (2) , Cornel Ioana (3)

1 GIPSA-MPACIF - GIPSA - Machines Parlantes, Agents Communicants & Interaction Face-à-face
2 SAR - Signal Analysis Research Group
3 GIPSA-SIGMAPHY - GIPSA - Signal Images Physique

In this paper we introduce a new watermarking model combining a time frequency representation using the Molecular Matching Pursuit algorithm and a psychoacoustic model. We take advantage of the notion of structure of the signal introduced by the MMP to get a precise representation of audio signals, and then by using a psychoacoustic model we can embed a watermark efficiently on the signal. By selecting atoms of time frequency components that are not perceptible by the human ear we ensure the security and imperceptibility of the watermark. Then by a thinner selection of the watermark host spots we ensure the robustness of our watermark to main kind of attacks, especially lossy compression attacks such as mp3 compression. We show the efficiency of this method thanks to a non-blind watermark recovery algorithm which confirms this method is robust and particularly adapted to long audio signals.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Mathieu Parvaix, Sridhar Krishnan, Cornel Ioana. An Audio Watermarking Method Based On Molecular Matching Pursuit. ICASSP 2008 - IEEE International Conference on Acoustics, Speech and Signal Processing, Mar 2008, Las Vegas, Nevada, United States. pp.1721 - 1724, ⟨10.1109/ICASSP.2008.4517961⟩. ⟨hal-00349541⟩