Nowadays virtually all multimedia production and distribution is digital. The advantages of digital media, for creation, processing and distribution are all well known: superior quality, quicker and easier to edit and modify possibility of software processing and may be the most important advantage is the unlimited copying of digital data without any loss of quality, whatsoever. However, this later advantage is not at all desired by the media producers and the content provider, in fact is perceived as a major threat to their intellectual property right and may cause them considerable financial losses. Once the digital technology is widely available to the public, piracy becomes a major issue. This generates the need of protecting the copyrighted material against piracy. One approach to address the problem is ‘digital watermarking’.
2. DIGITAL WATERMARKING
Digital watermarking is the process of embedding information into a digital signal which may be used to verify its authenticity or the identity of its owners. It is the imperceptible marking of multimedia data to brand ownership. The process of Digital Watermarking involves the modification of the original multimedia data to embed a watermark containing key information such as authentication or copy right codes. The embedding data must leave the original data perceptually unchanged, yet should impose modifications which can be detected by using an appropriate extraction algorithm [1, 4]. Some characteristics that a watermark data exhibits are [5, 6]: * IMPERCEPTIBILITY: Watermark should not be noticeable to the viewers nor should it degrade the quality of the content. * ROBUSTNESS: Watermark must be robust to common signal processing methods, geometric distortions on the image & video and to attacks like collusion and forgery. Watermarking must be robust enough to withstand all kinds for signal processing operations, “attacks”. * UNAMBIGUOUSNESS: Retrieval of the watermark should unambiguously identify the content owner. * UNDELETEABLE: The watermark should be undeletable by the hackers. * UNDETECTABLE: The watermark should be statistically undetectable.
There are two types of watermarking namely:
* VISIBLE WATERMARK: The information is visible in the picture or video. Typically, the information is text or a logo, which identifies the owner of the media. Example: When a television broadcaster adds its logo to the corner of transmitted video, this also is a visible watermark. * INVISIBLE WATERMARK: Information is added as digital data to audio, picture, or video, but it cannot be perceived as such. It also is possible to use hidden embedded information as a means of covert (hidden or secret) communication between individuals. * ROBUST: A watermark is robust if the embedded information can be detected reliably from the marked signal, even if degraded by any number of transformations. It helps to trace the one who is making illegal copies. * FRAGILE: A watermark is fragile if it fails to be detectable after the slightest modification. Fragile watermarks are commonly used for tamper detection or integrity proof. * SEMI-FRAGILE: A digital watermark is semi-fragile if it resists benign transformations, but fails detection after malignant transformations. Semi-fragile watermarks
commonly are used to detect malignant transformations.
Watermarking plays an important role in many fields. Some of the applications of watermarking are:
* Digital watermarking technology for authentication and ownership protection. The primary reason as stated before for using watermark is to identify the owner of the content by an invisible hidden “mark” that is imprinted into the video. * Digital watermarking technology for tracing of distribution channels.In many cases watermark is used in addition to the content encryption, where the encryption provides the secure distribution method while the watermark offers opportunities for the content owners to trace the contents and detect the unauthorized use or duplications. * Digital watermark technology for authentication and tamper proofing. The objective is to provide a method to authenticate the image or a video and assure the integrity of it .
In this paper, we developed a new method for watermarking of still images in a video sequence. Still images are embedded into the video in discrete wavelet transform domain. The main application of this method is to provide copy control, video authentication and copyright protection. Invisible watermarking is used in this approach, so that the watermarked image can be obtained only during the retrieval process. In the embedding process, we transform the video sequence to the DWT domain and replace the DWT co-efficient by the DWT co-efficient of the still image. Thus we obtain a video sequence, which is watermarked with a still image.
3. DESCRIPTION OF BASIC WATERMARKING TECHNIQUE
Digital watermarking is a process of embedding a secondary data namely the watermark in the primary data, which can be an image, audio or video.
Embedding and Extraction Process
The primary data is given as the input. Based on the technique used, the features are extracted from the data. The extracted and the watermark are combined by the proposed technique. Watermarked data is obtained as the output. Extraction process is the reverse of the embedding process .
WATERMARKED EXTRACTION FEATURE
Fig: (a) embedding process
Fig: (b) extraction process
Invisible Watermarking Techniques
* Spatial Domain Method
* Least significant bit modification.
* Correlation based technique.
* Spread spectrum.
* Frequency Domain Method
* Discrete Fourier transform
* Discrete cosine transform
* Discrete wavelet transform
4. DISCRETE WAVELET TRANSFORM
The transform of a signal is just another form of representing the signal.
It does not change the information content present in the signal. The Wavelet Transform provides a time-frequency representation of the signal. It can also be used to analyze transient signals. The Wavelet Transform uses multi-resolution technique by which different frequencies are analyzed with different resolutions.
A wave is an oscillating function of time or space and is periodic. Wavelets are localized waves. They have their energy concentrated in time or space and are suited to analysis of transient signals. The Wavelet Transform uses wavelets of finite energy.
Fig: Demonstration of (a) A Wave (b) A Wavelet.
The DWT is computed by successive low pass and high pass filtering of the discrete time-domain signal. In the figure, the signal is denoted by the sequence x[n], where n is an integer. The low pass filter is denoted by G0 while the high pass filter is denoted by H0. At each level, the high pass filter produces detail information; d[n], while the low pass filter associated with scaling function produces coarse approximations, a[n].
Water marking scheme overview
The DWT divides the image into 4 sub-bands
* LL – Lower resolution version of image
* LH – Horizontal edge data
* HL – Vertical edge data
* HH – Diagonal edge data
In the figure above we can see as we apply DWT to the image we get high resolution diagonal edge data in HH part, vertical edge data in HL part, horizontal edge data in LH part and Lower resolution version of the image in LL part. Again further DWT is applied to the each sub band .
5. EMBEDDING TECHNIQUE
In this approach discrete wavelet transform based algorithm is used to embed
the still image in the video sequence. Steps:
* The video sequence, which has to be watermarked, is divided into frames. * The frames are selected sequentially.
* Discrete wavelet transform is applied to each frame, which is divided into 5 levels. * The d5 level DWT coefficients of the video sequence are selected. * The still image, which is to be embedded in the video sequence, is wavelet transformed into 10 levels. * The DWT coefficients of the still image replace the DWT coefficients of the video sequence. * Inverse discrete wavelet transform is applied to these new coefficients. Thus, we get the watermarked video sequence, which is robust in nature .
6. EXTRACTION TECHNIQUE
The extraction of watermark requires the watermarked video and the knowledge of the technique used for the embedding process. Steps:
* The watermarked video sequence is divided into frames.
* The frames are selected sequentially.
* Discrete wavelet transform is applied to each frame, which is divided into 5 levels. * The d5 level DWT coefficients of the video sequence are selected. These coefficients are extracted from the video. * Inverse discrete wavelet transform is applied to these coefficients. Thus the still image can be obtained from the video .
In this paper, we have presented an overview about the digital watermarking as a tool for authentication and data hiding in multimedia applications. We then described an approach, in which a still image is watermarked in a video sequence using the embedding and extraction method of discrete wavelet transform (DWT).Such a technique can successfully withstand very powerful watermark attacks. An extension of this technique could be watermarking of audio and image in a video.
Digital watermark is an important step toward management of copyrighted and secured documents. Any kind of digital data like video (consists of image frames), audio, text can easily watermarked. The most obvious application is to use watermarks to encode information. Information hiding is possible with the help of watermarking which is the need of today’s piracy or illegal use of images. The images can be copyrighted & patented with the help of watermarking technique to prevent unauthorized access. Visible watermarking technique is used by the Television channels to restrict the illegal transmission & broadcasting of T.V. programs by inserting the watermark at the corner of the channel signals. Illegal use of the watermarking technique can be done by the terrorists or other persons who are against the community by transferring the hiding information through watermarks. But there are watermarking agents which act as a detective through the transmission & reception of data/information. So, the watermarking technique is very much popular in these days, only thing we need now days is recovery of Digital watermark if there is attack on it.
 K. Anusudha & A. Sangeetha “Wavelet domain approach to watermarking of still image in video” “International Journal of Secure Software Engineering and Technology”. Vol1 No1 (January-June 2011). F.A.P. Petitcolas, et al., “Information Hiding – A Survey”, Proc.IEEE Trans. In Image Watermarking, 87(7), 1062-1078, July 1999. WEBSITES