Digital Information Hiding Techniques

Information hiding, a form of watermark, embeds data into digital media for the purpose of identification and copyright. Several constraints affect this process: the size of data to be hidden, the need for robustness of these data under conditions where a host_signal is subject to distortions, for e.g., lossy_compression, and the degree to which the data must be immune to interception, modification, or removal by a third person. Here, we explore two techniques (DC watermarking scheme and Time Domain watermarking technique) for addressing the data-hiding process and evaluate these techniques in light of the copyright protection application. The measures (SNR, PSNR, NRMSE) were used to improve the results. Besides that the Matlab were used as a programming language in this paper. ةيمقرلا تامولعملا ءافخإ تاينقـت قفوم ءايلع ديجملا دبع دمحم نعم ةيدان دعاسم سردم دعاسم سردم ساحلا مولع مسق و ب ، ساحلا مولع ةيلآ و تايضايرلاو ب ، لصوملا ةعماج صلختسملا يمقرلا طئاسولا تا فلم لخاد تا نايبلا رمطب موقي يذلا ةيئاملا ةملاعلا جذومن ،تامولعملا ءافخإ ة و مدختسي يذلا ة يكلملا قو قح ى لع ظا فحلاو ةيوهلا ديدحتو فيرعتلا ضرغل ، تباوثلا نم ددع كا نه ةيلمعلا هذه ىلع رثؤت يتلا : تا نايبلا هذ ه ةمواقم ىلا ةجاحلا ،اهئافخا دارملا تانايبلا مجح ) ةبلاصلا ( ىلا ةيسيئرلا ةراشلاا ضرعتت ثيح ،ةيجراخلا تاريغتملل ةفلتخملا فورظلا تحت لاثمو مئاد شيوشت ةعانملا ةجرد نوكتل تانايبلا اهلمحت نا بجي يتلا ةجردلا كلذآو ،نادقفلا ةيعونب تانايبلل سبكلا كلذ ةثلاث ةهج وا صخش لبق نم فذحلا وا رييغتلا ،عطقلا دض ةيلاع . تينقت قيبطت مت ،ثحبلا اذه يف ي ) بولساب ةيئاملا ةملاعلا DC حتلا بولساب ةيئاملا ةملاعلا و ل يو ينا مزلا ( ق يبطت ءوض ي ف تا ينقتلا هذ ه ةءا فآ تا بثلاو تا نايبلا ءا فخا ة يلمع ة نونع ضرغل ك لذو ةيكلملا قوقح ةيامح . سييا قملا مادخت سا م ت ) SNR ، PSNR ، NRMSE ( ا هتءافآو جئا تنلا ةح ص تا بثا ضر غل دامتعا مت ،كلذ ىلا ةفاضلااب Matlab ثحبلا اذه يف ةيجمرب ةغلآ .


Introduction
According to the spread of the internet, multimedia becomes treat digital contents, which can be copied easily without any loss in quality and contents.This poses a big problem for the protection of Intellectual Property (IP) rights of the copyright owners and the copyright of these digital media has become a lot of more difficult to manage.As a result, a technique called digital watermarking is introduced to protect the ownership of these contents.Digital watermarking can be realized by many different methods.
In common to all of those methods, digital watermarking is a technique of embedding a digital signal or pattern on a digital document.The digital document may be text, audio, image or video.When the digital document is in the form of an audio signal, the embedding technique is called audio watermarking.[1] [2] The digital media that carries the watermark is called a cover signal or host signal.The watermark is embedded into the host signal by a watermark embedder and is detected by a watermark detector.A watermark key prevents unauthorized watermark embedding and watermark detection.[3] Data hiding in audio signals is especially challenging, because the Human Auditory System (HAS) operates over a wide dynamic range.The HAS perceives over a range of power greater than one billion to one and a range of frequencies greater than one thousand to one.While the HAS has a large dynamic range, it has a fairly small differential range.As a result, loud sounds tend to mask out quiet sounds.Additionally, the HAS is unable to perceive absolute phase, only relative ignored by the listener in most cases.[4][5]

Audio Watermarking Goals
Audio watermarking is implemented to satisfy four goals: [6] a.The original intention of watermarking is for copyright protection.Therefore, the most obvious purposes are the needs for proof of ownership and the enforcement of usage policy.b.Maximizing the difficulty of removing the watermark without destroying the audio.c.Minimizing the perceptual effect of the watermark.d.Maximizing the information, which can be encoded per second of original audio.

Types of Watermarking
Three types of watermarking system can be classified by the information required in the watermark extraction process: [4] [7] • Private Watermarking Private watermarking is also called none blind watermarking system requires at least the original data, and watermarks are needed to verify the presence of a watermark.A secret key used in embedding may also be needed.
It can be used as authentication and content integrity mechanisms in a variety of ways.It may contain information for identifying the license or to prove ownership in disputes.

• Semiprivate Watermarking
Semi-private watermarking is also called semi-blind watermarking.The original secret key and watermark are needed in order to identify a watermark.In applications such as Digital Versatile Disc (DVD) where the disc reader needs to know whether it is allowed to play the content or not, and fingerprinting where the goal is to identify the original recipient of pirated copies.

• Public Watermarking
Public watermarking is also called blind watermarking.The original signal is not required.Only requires the secret key used in embedding for a watermark to be extracted.These public watermarks should not detectable or removable by a third party.It usually contains copyright or licensing information, such as the identifier of the copyright holder, the creator of the material.The receiver (licensee) of copyrighted material retrieves a public watermark.
Although public watermarking is not secure it is much more difficult to remove than a visible label.Moreover, the failure of detection of the public watermark indicates that the image has been significantly tampered with and the user can be informed of the alteration.

Digital Audio Watermarking
Watermarking digital media has received a great interest in the literature and research community.Most watermarking schemes focus on image and

A) Frequency Domain Audio Watermarking
Audio watermarking techniques, that work in frequency domain, take the advantage of audio masking characteristics of HAS to embed an inaudible watermark signal in digital audio.Transforming audio signal from time domain to frequency domain enables watermarking system to embed the watermark into perceptually significant components.This will provide the system with a high level of robustness, because of that any attempt to remove the watermark will result in introducing a serious distortion in original audio signal fidelity.The input signal is first transformed to frequency domain where the watermark is embedded, the resulting signal then goes through inverse frequency transform to get the watermarked signal as output.See Figure (1) 5) Where, 'α' is a scaling factor.Here, the value of α=0.2 .w (i) is the watermark bit (0 or 1).f. Compute the Inverse Discrete Cosine Transform (IDCT).

Result of DC Method using Symphony
The implementation of the DC method was tested on the symphony signal of size 1.21 MB with sampling rate 8 KH Z and resolution of 8 bits/sample, and the watermark is a speech signal of size 10.63 KB of sampling rate 8 KH Z and resolution of 8 bits/sample.Figures (3)(4)(5)( 6) show a sample of watermark message that to be hidden inside original message before and after applying watermarking algorithm.

Results of DCT Method using Symphony
The implementation of the DCT method was tested on the symphony signal of size 1.21 MB with sampling rate 8 KH Z and resolution of 8 bits/sample, the watermark is a speech signal of size 10.63 KB of sampling rate 8 KH Z and resolution of 8 bits/sample.Figures (7) (8) show a sample of watermark message that to be hidden inside original message before and after applying watermarking algorithm.The implementation of the DC method was tested on the speech signal of size 493 KB with sampling rate 8 KH Z and resolution of 8 bits/sample.The watermark is speech signal of size 10.63 KB with sampling rate 8 KH Z and resolution of 8 bits/sample.Figures (9) (10) show a sample of watermark message that's hide inside original message before and after applying watermarking algorithm.

Results of DCT Method using Speech
The implementation of the DCT method was tested on the speech signal of size 493 KB with sampling rate 8 KH Z and resolution of 8 bits/sample.The watermark is speech signal of size 10.63 KB with sampling rate 8 KH Z and resolution of 8 bits/sample.Figures (11)(12) show a sample of watermark message that to be hidden inside original message before and after applying watermarking algorithm.

Summary
After applying the (DC and DCT) methods, a summary can be given for the last results using the following performance measures (SNR, PSNR, NRMSE) as shown in Table (1).13. Conclusions a.After executing the 2 above methods (DC, DCT), it concluded that the DC method is better and has a good performance than the DCT method for embedding the watermark data of type (speech) in both types of signal (Symphony, Speech), as it's clear from table (1).b.Some audio watermarking systems require the original audio signal, or any information derived from it, to be presented in detection process.This will leads to a large number of original works have to be stored and searched during detection.Here, the applied methods don't required the original signal to detect the watermarked signal.c.The methods mentioned in this paper are very useful to hide data of type sound signals.d.The measures (SNR, PSNR) well be increase in each time the frame size is increased, but they well be decreased in each time the size of hidden data is increased.So, it's important to choose the suitable size of frame and the data to be hide.
‫ﻟﺴﻨﺔ‬ ‫ﻭﺍﻟﺮﻳﺎﺿﻴﺎﺕ‬ ‫ﺍﳊﺎﺳﻮﺏ‬ ‫ﻟﻌﻠﻮﻡ‬ ‫ﺍﻟﺮﺍﻓﺪﻳﻦ‬ ‫ﳎﻠﺔ‬ ٢٠١٠‫ﺍﳌﻌﻠﻮﻣﺎﺕ‬ ‫ﺗﻘﺎﻧـﺔ‬ ‫ﰲ‬ ‫ﺍﻟﺜﺎﻟﺚ‬ ‫ﺍﻟﻌﻠﻤﻲ‬ ‫ﺍﳌﺆﲤﺮ‬ ‫ﻭﻗﺎﺋﻊ‬ 29-30/Nov./2010‫ﻭﺍﻟﺮﻳﺎﺿﻴﺎﺕ‬ ‫ﺍﳊﺎﺳﻮﺏ‬ ‫ﻋﻠﻮﻡ‬ ‫ﻛﻠﻴﺔ‬ -‫ﺍﳌﻮﺻﻞ‬ ‫ﺟﺎﻣﻌﺔ‬ ٤٦٤video watermarking.A few audio watermarking techniques have been reported.Digital audio watermarking is the process of embedding a watermark signal into audio signal.Audio watermarking is a difficult process because of the sensitivity of Human Auditory System (HAS).The requirements mentioned earlier are common to both image and audio watermarking techniques.Despite their similarities, audio and still image watermarking systems exhibit significant differences.First of all, the fact that images are two-dimensional signals provide attackers with more ways of introducing distortions that might affect watermark integrity e.g.scaling, rotation or removal of rows/columns.Audio watermarking methods need not to deal with such attacks, as audio is a one-dimensional signal.Due to the difference between HAS and Human Visual System (HVS), different masking principles should taken into account in each case.Digital audio watermarking techniques can be classified according to the domain where the watermarking takes place.The following sections will discuss audio watermarking techniques and classify them to four categories:[3]

Table ( 1
) Results of applying the performance measures