Video Mining Watermarking Using Spectrum Nework

Watermarking VIDEO MINING files has recently become much focused due to the wide availability on the Internet leading to digital proliferation of digital VIDEO MINING files. Watermarking may give the record companies the ability to enforce the copy right protection of their products to identify the origin, author, usage rights and distributor or authorized user of the VIDEO MINING clip even if the clip has been processed or distorted. Proof of ownership watermarks may help to determine rightful file ownership perhaps in the court of law. Enforcement of usage watermarks could provide instructions or copyright information to consumer applications, which could refuse to duplicate or play music in violation of a usage policy. There are also variations in watermarks, namely, “Fingerprint” watermarks and “Fragile” watermarks.

1.    Introduction: Digital watermarking is a technique, which allows you to add hidden copyright messages or data in to digital VIDEO MINING, video, image signals or documents. Such hidden message is a group of bits describing information pertaining to the signal. VIDEO MINING watermarking embeds a sequence of data such as additional information into an VIDEO MINING file. his additional information is called as watermark. The watermark is inaudible because it exploits the characteristics of Human Auditory System (HAS). If the signal is maintained below the threshold of sensitivity then the watermark will be inaudible. Thus, since the HAS sensitivity threshold is 2 KHz, small changes in spectrum above 2 KHz are less likely to be noticed by a human listener.

Tools for Digital Watermarking:
There are two different categories of watermarking tools available:
•    First is based on fingerprint binary information (FBI) that identifies documents by hidden numbers.
•    Second is SYSCOP (System for Copyright Protection), can encode additional identification information such as author’s name or ISBN number of a book.
•    Direct sequence and frequency hopping spread spectrum techniques are major watermarking embedding methods used in existing tools.
•    Hiding secret messages in least significant bit of same pseudorandom frequencies or pixels of an image is a common approach.
•    The direct sequence technique adds noise to every element of the document whereas, frequency hopping spread spectrum technique selects a pseudorandom subset of data to be watermarked.
•    Dig marc and FBI use direct sequence methods to superimpose a watermark over an image by modulating a noise pattern of the same size into an image.
•    SYSCOP uses a secret key to pseudorandomly select the blocks and frequencies that are modulated within the block.

2.    Modern Techniques in Frequency Domain:
•    Modern techniques use two different spread spectrum techniques called Direct sequence spread spectrum & Frequency Hopping Spread Spectrum (FHSS).
•    Direct Sequence Spread Spectrum (DSSS) hides the information by phase modulating the data signal witha pseudorandom number sequence known to both sender and receiver.
•    Frequency Hopping Spread Spectrum (FHSS) divides the available bandwidth into multiple channels and hops between these channels.
•    Modifications of original signal’s noise caused by moderate levels of wideband noise are not visible or audible to any intruder.

3.    Watermark Extraction: A watermark must be extractable even from the degraded documents. The degraded document has to be normalized into its original format before extraction of original watermark.
•    The retrieval process normally needs either the original, unwatermarked data or the added noise (watermark) for comparison with the watermarked data/document.
•    But in case the original data is not available, as in most of the cases, it is sill possible to extract the watermark.
•    The algorithm instead of comparisons detects specific properties & patterns from the watermarked document. These patterns can be represented as signal shapes or the cross correlation between certain document elements. The retrieval is more efficient in real-time systems.

4.    Scope & Applications Of The Proposed Scheme:
    Digital watermarks find a wide range of applications when it comes to authentication.
    However, copyright is the motivating factor for developing watermarking technique.
    Digital VIDEO MINING watermarking aims to identify the origin, author, owner, usage rights, distributors or authorized user of an VIDEO MINING clip.
    The watermark should also be retrievable if the clip is processed or distorted.

5.    Problem Definition
Digital Watermarking For VIDEO MINING Files:
VIDEO MINING piracy has become a real problem for the VIDEO MINING recording industry. Watermarking schemes are most commonly designed for copyright protection to resolve piracy disputes.


Any VIDEO MINING watermarking technique has to adhere to various parameters. These parameters are as follows:
1.    Robustness: It describes the reliability of the watermark detection after it has been through various signal processing operations.
2.    Security: Security reflects how difficult it is to remove a watermark. A scheme is truly secure if knowing the exact embedding algorithm do not help the user to detect or extract the hidden data.
3.    Transparency: Transparency relates to human ability to hear VIDEO MINING watermark. Usually if nor always complete transparency (inaudibility) is desired.
4.    Complexity: Complexity of encoding scheme might be an important reason to choose one algorithm over the other. The complexity should also consider the processing power of consumer devices.
5.    Capacity: It describes how many information bits can be reliably embedded.

6.    Security against VIDEO MINING Watermarking Attacks:
An operation that may decrease watermarking performance is called “attack”. Attacks can be categorized as follows:
1.    Removal Attacks: Removes the watermark without understanding the watermarking scheme used.
2.    Geometric Attacks:Distort watermarks detection through receiver’s desynchronisation.
3.    Cryptographic Attacks: Cracks the watermarking scheme itself
4.    Protocol Attacks: It exploits the invertible watermarks to cause ownership ambiguity.
Thus the difficulties in watermarking VIDEO MINING have to be overcome. The difficulties lie in both the device desire to pressure file quality and the need for watermark to remain intact after a number of possible damaging file operations. The available technologies for VIDEO MINING watermarking are effective in their own ways. But also have certain drawbacks. However, new technologies are being developed to enable efficient VIDEO MINING watermarking algorithms.

Proposed System: The proposed system enables to utilize the features of existing technologies. But since the existing techniques are not freely available for common user, the proposed system aims at generating the watermarking schemes for common user.
:
1.    Embedder
2.    Detector

Embedder Subsystem: The owner of the file uses the embedder. The owner of the file is assumed to have all the copyrights and all the ownership rights. The file is assumed legally to belong only to the owner.

Database Subsystem: The database is required to store the corresponding record of file’s watermark time & timestamp, the watermark of the file (encrypted) and the corresponding owner of the file. Thus, we can say that the ‘embedder subsystem’ & ‘Detector subsystem’ will require a common database if the watermark is not self-revealing about the owner’s identity.

Detector Subsystem: The ‘detector subsystem’ is used to identify the watermark from the already watermarked file. It will detect whether the file is watermarked or not before extracting the watermark. 

7.    Reference
1.    P. Bassia and I. Pitas, (1998)“Robust noise watermarking in the time domain”, in 9th European Signal Processing Conference (EUSIPCO’98), Island of Rhodes, Greece.
2.    J. G. Beerends, J. A. Stemerdink, (1992)“A perceptual noise quality measure based on a psychoacoustics sound representation,” J. Noise Eng. Soc., vol. 40, pp. 963-978.

3.    W. Bender,  D. Gruhl, N. Morimoto and A. Lu, (1996) “Techniques for data hiding”, IBM Systems Journal, vol. 35, No. 3&4, pp. 313-336.
4.    D. V. S. Chandra, (2002) “Digital image watermarking using singular value decomposition”, Proceedings of 45th IEEE Midwest Symposium on Circuits and Systems, Tulsa OK, pp. 264-267.
5.    B. Chen and G. W. Wornell, (2001)“Quantization index modulation: a class of probably good methods for digital watermarking and information embedding”,  IEEE Trans. on Information Theory, vol. 47,  no: 4, pp. 1423-1443.
6.    I. J. Cox, J. Kilian, F. T. Leighton, and T. Shamoon, (1997) “Secure spread  spectrum watermarking for multimedia”, IEEE Trans. on Image Process., vol. 6, no: 12, pp. 1673-1687.
7.    J. Dittmann, M. Steinbach, T. Kunkelmann, and L. Stoffel, (2000)  “Watermarking for media: Classification, quality evaluation, design improvements”, In Proceedings ACM Multimedia 2000, Marina Del Rey, USA, pp. 107-110.
8.    R. A. Garcia, (1999) “Digital watermarking of noise signals using a psychoacoustic auditory model and spread spectrum theory”, 107th Convention: Noise Engineering Society, New York.
9.    V. I. Gorodertski, L. J. Popyack, V. Smailov, and V. A. Skormin, (2001) “SVD-based approach to transparent embedding data into digital images”, MMM-ACNS2001, St. Petersbug, Russia.
10.    D. Gruhl, A. Lu, and W. Bender, (1996) “Echo hiding”, Information Hiding 1st International Workshop.
11.    S. Katzenbeisser and F. A. P.  Petitcolas, (2000) “Information Hiding techniques for steganography and digital watermarking, Artech House.
12.    D. Kirovski and H. Malvar, “Spread-spectrum watermarking of noise signals”, IEEE Transaction on Signal Processing, vol. 51, No.4, pp. 1020-1033.
13.    A.N. Lemma, J. Aprea, W. Oomen, and L. Kerkhof (2003) “A temporal domain noise watermarking technique”, IEEE Transaction on Signal Processing, Vol. 51, No.4, pp. 1088-1097.
14.    R. Liu and T. Tan, (2002) “A SVD-based watermarking scheme for protecting rightful ownership”, IEEE Transaction on Multimedia, 4(1), pp. 121-128, March.
15.    C. S. Lu, H. Mark, and L. H. Chen, (2000) “Multipurpose noise watermarking”, Proc. of the 15th IAPR : Int. Conf. on Pattern Recognition, Barcelona.