Research On The Approaches Of Digital Audio Watermarking

The rapid development of computers and Internet and the maturity of video and audio compression technologies make it possible to release and distribute a variety of digital products expediently on the Internet. The communicative means of digital information approach to achieve an unprecedented breadth and depth. But at the same time, the illegal modifying, illegal copying and illegal distribution of digital information has also become easier. The secure storage, secure transmission and secure access of digital information are facing new challenges. The phenomenon that the legitimate interests of the copyright owners of digital products are violated is increasingly serious. Thus, in this day and age. with high infomatization and complete internetization. how to make the network resources fully utilized and the copyright of digital products effectively protected is becoming more and more important.Digital audio products are an important part of digital multimedia products. The increasingly broad scope of its use makes the copyright protection of digital audio products become an urgent problem to be solved. Audio digital watermarking, as an important means of copyright protection of audio products, in-depth study of the technology has great practical significance. Copyright owner can embed the original audio data is not visible or not the police's image, serial number, signature or copyright text and other information, such information is embedded in close connection with the original audio data, and not separable.Audio digital watermarking technology has been in the international academic and business circles attracted wide attention, as the current field of information security characteristics of a technology that can effectively protect the audio legitimate interests of copyright owners.According to the h.iman auditor system (HAS) characteristics, and the human visual system (HVS) compared to the sensitivity of the image changes, it changes the sound sensitivity is much higher, and therefore the signal in a static image than digital watermark embedded in the embedding diaital watermarking audio data is much more difficult. At present, domestic and international research on digital watermarking technology mainly as an embedded object still image signal, and audio data based watermarking research is rare.In this article, the realization of digital audio watermarking scheme has been studied, in order to meet the robust watermark, transparency, blind detection and blind extraction requirements, this article and improved audio digital watermarking to achieve the following programs:(1) Energy comparison based digital audio watermarking implementations. In this scenario, as the audio signal embedded object is divided into several sections, each section contains the same sampling point, every section has been divided into several sections, each section also contains the same sampling point. The previous two sections for each piece of the energy of the comparison, digital watermark bits, and the human auditory system (HAS) of the masking effect of the combination. through the audio signal does not change or reduce the energy method of audio digital watermark embedding.(2) Based on amplitude comparison of audio watermarking implementations. In this scenario, as the audio signal embedded object is divided into several segments, each segment contains the same sampling point. For each segment for the first two sampling points amplitude comparison, digital watermark bits, and the human auditory system (HAS) of the masking effect of the combination does not change or decrease by the amplitude of the audio signal the method of audio watermarking embedded.(3) Wavelet-based digital audio watermarking implementations. In this scenario. as the embedded object audio signal is divided into a number of frames, each frame contains the same sampling point, each frame has been divided into several sections. each section also contains the same sampling point. For each frame before the two wavelet transform, the fine component for energy comparison, digital watermark bits, and the human auditory system (HAS) of the masking effect of the combination does not change or reduce the fine by means of energy components. Digital watermark is embedded in the audio component in fine.(4) Based on Cepstrum audio digital watermark implementations. In this scenario, as the embedded object is divided into a number of audio frames, each frame contains the same sampling point, the designated frame complex cepstrum transformation, more complex cepstral mean and threshold, according to the watermark sequence is "0 "or" 1 ", by no change or decrease, increasing the average number of complex cepstrum method, the audio digital watermark embedded in the complex cepstrum coefficients.By using MATLAB to achieve the above audio watermarking scheme, simulation experiment proves that the digital watermark implementations Timbre smooth by including low pass filtering, resampling, adding Gaussian white noise, for lossy compression and audio format conversion, etc. common signal processing and performance testing to withstand attacks that can better meet the robustness, transparency and Watermarking, blind extraction requirements.