A blind audio watermarking scheme based on neural network and psychoacoustic model with error correcting code in wavelet domain

Digital watermarking consists in embedding imperceptible information into a host signal. It has been proposed to solve problems as varied as the protection of the copyright, content authentication, fingerprinting and broadcast monitoring. This paper presents a new approach for audio watermarking using the QR factorization in wavelet domain. This approach is based on embedding a watermark binary image in the R matrices of low frequency blocks DWT coefficients of audio signal. In this algorithm, the watermark is embedded by applying a Quantization Index Modulation (QIM) process on the determined optimal sample for each matrix R. The watermark can be blindly extracted without the knowledge of the original audio signal. Experimental results show that the proposed audio watermarking scheme maintains high quality of the audio signal. Signal to Noise Ratio (SNR), Log Spectral Distortion (LSD) and Mean Opinion Score (MOS) are about 40 dB, 0.37 dB and 4.84, respectively. Moreover, the scheme is quite robust against common signal processing attacks such as noise addition, filtering and MP3 compression. In addition, this method ensures a secure extraction process by using a private key, making it suitable for secure applications such as copyright protection.

The technique used to hide copyright information (watermark) into the digital audio signal is termed as audio watermarking. Audio watermarking is an excellent approach to provide a solution to mitigate challenges that occur from easy copying and distribution of audio files that are being downloaded or uploaded through the web. The audio watermarking algorithms proposed earlier were implemented by image or binary logo or a unique pattern as watermark. To tackle the difficulty of online musical piracy, unethical attempt to alter purposefully copyrighted audio data and unauthorized distribution and tampering of audio files an efficient audio watermarking method is suggested here. In this method the copyright information (watermark) which is an audio signal of shorter length is imperceptibly added into the original audio signal with greater duration. Here two audio watermarking methods using DCT (Discrete Cosine Transform) and then with DWT-SVD (Discrete Wavelet Transform — singular value decomposition) methods are proposed with audio signal encrypted with secure hash algorithm as watermark. A comparative study on performance evaluation of both DCT and DWT-SVD methods are made in terms of PSNR (Peak-Signal to Noise Ratio) and MSE (Mean Square Error). The results show that transparency is best for each methods and DWT-SVD method is more efficient against signal attacks. The elementary notion of this document is to provide a beneficial and satisfactory method for secure audio file transmission by providing copyright protection.

Audio watermarking is a method that embeds inaudible information into digital audio data. This paper describes an audio watermarking scheme using the Least Significant Bit (LSB) in the multiresolution domain exploiting the perceptual evaluation of audio quality algorithm (PEAQ) and MP3 compression characteristics. To guarantee perfect inaudibility propriety, we adapt the PEAQ algorithm to select the adequate audio band used to hide the watermark in the LSB of the multiresolution coefficients. Additionally, we use some MP3 compression characteristics to increase robustness of the scheme and we duplicate the watermark bits in the whole audio signal blocks assuring then an important ratio property without affecting the inaudibility property. The described method provides good robustness results against compression and StirMark attacks. Besides, it allows blind retrieval of embedded watermark which does not need the original audio.

Many effective watermarking algorithms have been proposed and implemented for digital images and digital video. However, a few algorithms have been proposed for audio watermarking. This is due to the fact that, human auditory system (HAS) is far more complex and sensitive than human visual system (HVS). In this research work, a new method of embedding image data into the audio signal and additive audio watermarking algorithm based on Discrete Cosine Transformation (DCT) domain is proposed. First, the original audio is transformed into DCT domain. The DCT coefficients are divided into a fixed number of subsections and the energy of each subsection is calculated. Next, watermark is generated from image by image processing algorithm. Watermarks are then embedded into selected peaks of highest energy subsection. Experimental results demonstrate that the watermark is inaudible and this algorithm is robust to common operations of digital audio signal processing, such as noise addition, re-sampling, requantization and so on. To evaluate the performance of the proposed audio watermarking method, subjective and objective quality tests including Bit Error Rate (BER) and Signal to Noise ratio (SNR) are conducted.

In order to improve the robustness and imperceptibility in practical application, a novel audio watermarking algorithm with strong robustness is proposed by exploring the multi-resolution characteristic of discrete wavelet transform (DWT) and the energy compaction capability of discrete cosine transform (DCT). The human auditory system is insensitive to the minor changes in the frequency components of the audio signal, so the watermarks can be embedded by slightly modifying the frequency components of the audio signal. The audio fragments segmented from the cover audio signal are decomposed by DWT to obtain several groups of wavelet coefficients with different frequency bands, and then the fourth level detail coefficient is selected to be divided into the former packet and the latter packet, which are executed for DCT to get two sets of transform domain coefficients (TDC) respectively. Finally, the average amplitudes of the two sets of TDC are modified to embed the binary image watermark according to the special embedding rule. The watermark extraction is blind without the carrier audio signal. Experimental results confirm that the proposed algorithm has good imperceptibility, large payload capacity and strong robustness when resisting against various attacks such as MP3 compression, low-pass filtering, re-sampling, re-quantization, amplitude scaling, echo addition and noise corruption.

In this paper, we present a novel audio watermarking scheme based on spread spectrum techniques that embeds a digital watermark within an audio signal using the instantaneous mean frequency (IMF) of the signal. Audio watermarking offers a solution to data piracy and helps to protect the rights of the artists and copyright holders. Our content-based algorithm aims to satisfy and maximize both imperceptibility and robustness of the watermark. In addition, our technique uses the short-time Fourier transform of the original audio signal to estimate a weighted IMF of the signal. Based on the masking properties of the psychoacoustic model, we then derive the required sound pressure level of the watermark. From these results, modulation is performed to produce a signal-dependent watermark that is imperceptible. The proposed method allows 25 bits to be embedded and recovered within a 5 second sample of an audio signal. Experimental results have shown that our scheme is robust to common signal processing attacks including filtering, MP3 compression and noise addition.

The technology of embedding image into the audio signal has been researched and an algorithm of additive audio watermarking based on SNR to determine a scaling parameter /spl alpha/ is proposed. The visually recognizable binary watermark image could be embedded in the original signal wavelet domain in the algorithm. The intensity of embedded watermarks on the different audio segments would be modified by adaptively modulating the scaling parameter /spl alpha/. The experimental result demonstrate that the watermark is imperceptible and the algorithm is robust to many operations to digital audio signal, such as resampling, cropping, low pass filter, MP3 compression and so on. The algorithm is better than that without using SNR, and has a reference value to blind watermarking.

This paper presents an audio watermarking scheme that ensures the efficient of robustness-distortion ratio and a pleasing amount of capacity. Quantization Index Modulation (QIM) method synergism along with Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT), Singular Value Decomposition (SVD), and Cartesian to Polar Transformation (CPT) to emphasize copyright protection for harmonical work of art. DWT synergism along with DCT, SVD, and CPT to emphasize copyright protection for harmonical work of art. The system establishes with DWT that divides the signal within a specific range of frequency, alter it to the frequency domain by DCT, partite with mathematical processing SVD, and transform the peculiar value of matrix-S to polar value using CPT, furthermore, QIM embed the watermark into the angular value from the CPT. The system performance is astonishing, tested using 4 different audio host signal and evaluated considers to desirable Bit Error Rate (BER) <; 0.025, Signal-to-Noise Ratio (SNR)>21dB, a great amount of data payload that is 525 bit per second, and the noise caused by watermark is inaudible. All of the experiments, trials, and results show that the designed audio watermarking system is robust and imperceptible.

In this paper, we propose an intelligent audio watermarking approach based on QR factorization in wavelet domain by exploiting Multi-objective Particle Swarm Optimization (MOPSO). First, the audio signal is decomposed into low and high frequency ranges by using Discrete Wavelet Transform (DWT). The low frequency part is then segmented into 2-D blocks and QR factorization is applied to each one. The watermark bits are embedded by quantizing the coefficients of resulting R matrices. To achieve best compromise between the conflicting objectives, namely imperceptibility and robustness, the quantization step is optimized using MOPSO process. The watermark can be blindly extracted without the knowledge of the original audio signal. Experimental results show that the proposed method not only maintains high quality of the audio signal, but also is robust against common signal processing manipulations. These results confirm the effectiveness of this method for secure applications such as copyright protection.

The technology of embedding image data into the audio signal and additive audio watermarking algorithm based on SNR to determine a scaling parameter was researched. The audio based on DWT. The intensity of embedded watermarks on the original audio signal is modified by adaptively modulation of the scaling parameter. Experimental result demonstrate that the watermark is inaudible and this algorithm is robust to many operations of digital audio signal processing, such as, resampling, lowpass filter and so on. The extracted watermark image quality is shown by considering normalize correlation (NC) value with a suitable scaling parameter for embedding.

To analyze the robustness performance of direct sequence spread spectrum (DSSS) based audio watermarking algorithm theoretically, the fundamental principle of spread spectrum watermarking is analyzed firstly. The communication model of DSSS-based audio watermarking system is designed according to the spread spectrum communication theory. The interrelation between bit error rate (BER) and watermark power and processing gain is revealed by utilizing definition of the optimal baseband transmission system under ideal channel. Finally, Monte Carlo simulation method is used to prove the BER formula. Experiments reveal that the expression can calculate the BER of detected watermarking under cover audio and additive white Gaussian noise (AWGN) interferences accurately.

This chapter discusses the robustness of digital audio watermarking algorithms against digital to analogue and analogue to digital conversions. This is an important challenge in many audio watermarking applications. We provide an overview on distortions caused by converting the signal in various scenarios. This includes environmental influences like background noise or room acoustics when taking microphone recordings as well as sound sampling effects like quantization noise. The aim is to show the complexity of influences that needto be taken into account. Additionally, we show test results of our own audio watermarking algorithm with respect to analogue recordings using a microphone which proves that a high robustness in this area can be achieved. To improve even the robustness of algorithms, we briefly introduce strategies against downmixing and playback speed changes of the audio signal.

Audio watermarking is a relatively new technology that aims to embed copyright information in audio signals as a proof of their ownership. The driving force behind this technology has been the ever-increasing illegal manipulation of genuine audio products released by the music industry. In this paper, we propose an effective, robust, and inaudible audio watermarking algorithm. The effectiveness of algorithm has been brought by virtue of applying a cascade of two powerful mathematical transforms; the discrete wavelets transform (DWT) and the singular value decomposition (SVD). Reported experimental results demonstrate the effectiveness of the proposed algorithm.

Advancement of digital signal processing and networking has raised many security and copyright concerns, thus it is very important to protect the authentication of digital data. In this work, an audio watermarking algorithm has been proposed which can be efficiently used for tamper detection and is also robust against reasonable attacks. Also, the watermarks are inaudible. The proposed algorithm can easily detect tampering as the watermarks are embedded at each frame without causing any audio degradation. In the proposed technique, first the audio signal is compressed using Graph Based Transform (GBT), for which watermarks are embedded into Line Spectral coefficients (LSFs) that are derived from linear prediction (LP) analysis with dither modulation-quantization index modulation (DM-QIM). Watermarks thus embedded in all frames are not only inaudible to the Human auditory system but also potentially provide robustness against meaningful attacks. This work also focuses on Blind tamper detection which is made effortless due to the proposed embedding algorithm. To measure the robustness of the algorithm, general processing of watermarked signals was done along with fragility testing. Quality of the audio was measured using Perceptual Evaluation of Speech Quality (PESQ) and Short-time objective intelligibility (STOI). The maximum PESQ score and STOI score of 2.8781 and 0.8150 respectively was observed without any attack on the audio signal. Tamper detection and quality measurement are the major contributions of this work. Detailed metric evaluation for attacks such as Scaling, Resampling, Filtering, Compression and Addition of White Gaussian noise (AWGN) has been computed and compared. The proposed technique makes tamper identification easier and gives framewise security.

This chapter gives an overview of the digital audio watermarking systems, including description of recently developed watermarking algorithms and an overview on the existing applications that use different audio watermarking methods. Audio watermarking algorithms are characterised by six essential properties, namely: perceptual transparency,watermark bit rate, robustness, blind/informedwatermark detection, security and computational complexity. Psychoacoustic models of the HAS that are exploited in order to preserve the subjective quality of the watermarked audio during the watermarking process are shortly reviewed. The most common watermark embedding techniques ranging from the simple LSB scheme to the various spread spectrum methods are presented. Application areas that have recently been developed and possible future applications areas are listed as well.