Image analysis – Applications
Reexamination Certificate
1999-11-19
2004-01-06
Mehta, Bhavesh M. (Department: 2721)
Image analysis
Applications
C713S176000, C380S205000
Reexamination Certificate
active
06674874
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to data processing apparatus and method for handling digital watermark information in digital contents and also relates to a storage medium which stores such a method.
2. Related Background Art
Hitherto, a technique called a digital watermark has been known as a method of preventing contents such as motion images, still images, audio sound, or the like from being illegally copied. The digital watermark is a technique for embedding additional information into the contents so that a human being cannot perceive it.
According to the digital watermark embedding technique, digital watermark information is embedded in a state as it is into digital contents or watermark information is once converted into digital information and, thereafter, the digital watermark information is embedded into analog contents.
Hitherto, a system for embedding digital watermark information by performing an arithmetic operation to a lightness value and a chromaticity value of each pixel of a still image is known. For example, in U.S. Pat. No. 5,636,292, a system whereby digital contents are divided into blocks and a “watermark pattern” constructed by values of +1 and −1 is added to every block is used.
As another embedding method, a method whereby a frequency transformation (Fourier transformation, discrete cosine transformation, wavelet transformation, etc.) is performed to a still image, watermark information is added into a frequency region, and thereafter, a reverse frequency transformation is performed is known.
According to a method of using the Fourier transformation, a PN line is added to input contents and the resultant contents are diffused and, thereafter, divided into blocks. The Fourier transformation is performed every block and digital watermark information of 1 bit is embedded into one block. A reverse Fourier transformation is performed to the block in which the digital watermark information was embedded and the same PN line as that used at the first stage is again added to the resultant contents, so that the contents in which the digital watermark was embedded is derived. A detailed technique of the above method has been disclosed in, for example, Onishi, Oka, and Matsui, “Method of Watermark Signature into Image by PN line”, Records of Symposium about Encryption and Information Security, SCIS 97-26B, 1997.
According to a method of using the discrete cosine transformation, input contents are divided into blocks and a discrete cosine transformation is performed every block. After information of 1bit was embedded into one block, the reverse transformation is performed, thereby forming contents in which a digital watermark was embedded. A detailed technique of the above method has been disclosed in, for example, Nakamura, Ogawa, and Takashima, “Digital Watermark System in Frequency Region for Protection of Copyright of Digital Image”, Records of Symposium about Encryption and Information Security, SCIS 97-26A, 1997.
According to a method of using the wavelet transformation, there is no need to divide input contents into blocks. A detailed technique of the above method has been disclosed in, Ishizuka, Sakai, and Sakurai, “Experimental Examination Regarding Safety and Reliability of Digital Watermark Technique using Wavelet Transformation”, Records of Symposium about Encryption and Information Security, SCIS 97-26D, 1997.
As for a technique for embedding digital watermark information into audio data as well, the digital watermark information can be embedded by a method whereby input contents are divided into blocks and, thereafter, a frequency transformation is performed, digital watermark information is embedded, and a reverse frequency transformation is performed in a manner similar to the case of an image.
A method whereby audio data is wavelet transformed and digital watermark information is embedded into obtained wavelet coefficients has also been known. Such a kind of technique has been disclosed in detail in, for example, Inoue and Sugi, “Encryption System of Digital Watermark—Encryption System in Multimedia Age”, Maruyama Gakugei Tosho Co., Ltd.
As a technique for embedding a digital watermark into motion image data, a method of embedding it into a motion vector or a method of combining motion images which are obtained from two cameras for photographing an object from slightly different angles has been known.
For example, according to MPEG as one of the motion image compression systems, by obtaining a difference between frames, a redundancy in the time base direction is reduced, the obtained difference is DCT transformed and variable length encoded, and a redundancy in the spatial direction is reduced, thereby realizing high efficient encoding. In this case, by paying attention to that there is a high correlation between continuous frames, a difference between the target frame to be encoded and the frame that is time-preceding or subsequent to the target frame is obtained, thereby reducing the redundancy. In MPEG, one unit (GOP) is formed by one I-picture (intra encoding image), four P-pictures (forward predictive encoding images), and ten B-pictures (bidirectional predictive encoding images), and a combination such that the I-picture is arranged at the head and the B-picture and P-picture are repetitively arranged is recommended. By arranging the I-picture at a predetermined period, a special reproduction such as a reverse reproduction or the like and a partial reproduction using the GOP as a unit are enabled and an error propagation is prevented. A motion compensation is performed in MPEG. That is, a difference between the target block and the macroblock close to the corresponding block of the preceding or subsequent frame is obtained on a unit basis of predetermined blocks (macroblock) obtained by collecting four blocks with respect to luminance data and two blocks with regard to color difference data, in which each block consists of (8 pixels×8 pixels), and the macroblock of the smallest difference is searched, thereby detecting a motion vector. The detected motion vector is encoded as data. Upon decoding, corresponding macroblock data of the preceding or subsequent frame is extracted by using the detected motion vector, thereby decoding the encoding data which was encoded by using the motion compensation. In the motion compensation as mentioned above, after the time-preceding frame was once encoded, it is again decoded, and the resultant decoded frame is set to a preceding frame, and the motion compensation is performed by using the macroblock in such a preceding frame and the macroblock of the frame to be encoded. In the method of embedding the digital watermark information into the motion vector of the motion image which is encoded as mentioned above, the movement of one motion vector is made to correspond to one bit of a bit train of the watermark information. That is, when the user wants to set a value of such one bit to “1”, the vector is moved by a distance of an extent such that it cannot be visually recognized. When the user wants to set a value of such one bit to “0”, the vector is not moved. All of the digital watermark information can be embedded by performing such a process to many motion vectors.
Subsequently, the method of embedding the digital watermark information by photographing the object from slightly different angles by using two cameras will now be described. Since the angles are very small, a difference between the two images photographed by two cameras cannot be visually discriminated. The two cameras are now assumed to be A and B. The images photographed by the cameras A and B are divided every frame and the divided images are assumed to be (a
1
, a
2
, . . . , an) and (b
1
, b
2
, . . . , bn), respectively. The images are selected in order showing the digital watermark information from the frames of the images photographed by the cameras A and B, thereby constructing motion image data. For example, one frame is made to correspond to one bit constructing the digit
Iwamura Keiichi
Yoshida Jun
Canon Kabushiki Kaisha
Choobin Barry
Fitzpatrick ,Cella, Harper & Scinto
Mehta Bhavesh M.
LandOfFree
Data processing apparatus and method and storage medium does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Data processing apparatus and method and storage medium, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Data processing apparatus and method and storage medium will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3265776