We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
High capacity, secure audio watermarking technique integrating spread spectrum and linear predictive coding.
- Authors
Korany, Noha O.; Elboghdadly, Namat M.; Elabdein, Mohamed Z.
- Abstract
This paper introduces a method adjustable audio watermarking approach with high auditory quality by exploiting the Spread Spectrum (SS) and the psychoacoustic modeling-based Discrete Wavelet Packet Transform (DWPT). The psychoacoustic model is used to shape the amplitude of the watermark signal during the embedding phase, ensuring a high level of imperceptibility. The approach based DWPT carefully mimics the multi- resolution properties of the human ear and incorporates simultaneous and temporal auditory masking. This paper aims to increase data payload and security. The watermark bits positions in the watermarking domain are determined by the logistic chaotic map in a random manner ensuring a high security level. The watermark used is an audio signal in parametric representation using Linear Predictive Coding (LPC) to accomplish a higher data payload. The LPC technique predicts a small number of coefficients, which represent different speech parameters, that are then applied in digital filters to create a synthetic version of the original speech signal. Various host audio signals are examined under various watermarking threats to assess the performance of the watermarking system. Experimental results expose that the introduced scheme provides a good measure of imperceptibility and robustness which are evaluated by the Objective Difference Grade (ODG) and the Bit Error Rate (BER%) respectively. The ODG's values were between -0.74 and -2.75, while the BER was in the range of 0 to 7.54%. The comparison introduced between the standard psychoacoustic model-1 (Standard PM) and the enhanced psychoacoustic model-1 based DWPT (Enhanced PM) for audio watermarking shows that Enhanced PM enhances the status of insufficient time-frequency resolution. A comparison of the proposed method to other state-of-the-art audio watermarking algorithms shows that it is a feasible alternative.
- Subjects
LINEAR predictive coding; DIGITAL watermarking; DISCRETE wavelet transforms; BIT error rate; AUDITORY masking
- Publication
Multimedia Tools & Applications, 2024, Vol 83, Issue 17, p50645
- ISSN
1380-7501
- Publication type
Article
- DOI
10.1007/s11042-023-17630-w