Image data encoding system and image input device

Description: TECHNICAL FIELD [Detailed description of the invention]

【0001】

[Technical field to which the invention belongs]

The present invention relates to the field of digital images, and particularly to an image data encoding system for embedding identification data (electronic shaving data) having special information in a digital image.

【0002】

The present invention also relates to an image input device such as a personal computer, and more particularly to an image input device having an illegal copy protection function.

【0003】

[Conventional technology]

In recent years, illegal duplication of digital images has become a problem.

【0004】

In order to prevent illegal duplication, a system in which digital image data is encrypted and only a reproduction system having a valid decryption key can reproduce the encrypted digital image data is considered. However, once the code is broken, further duplication cannot be prevented.

【0005】

Conventionally, the method of preventing unauthorized copying of this type of image input device has been aimed at preventing copying itself.

【0006】

FIG. 9 is a block showing an example of a conventional image input device having an unauthorized copy protection function. The input image is converted into compressed image data such as an MPEG data stream via the imaging means 901, the analog-digital conversion means 902, the conversion means 903, the quantization means 904, and the variable length coding means 905, and then the scramble means 906. It becomes scrambled compressed image data. The scrambled compressed image data can only be played back by a device that has a specific reverse scramble function.

【0007】

In this way, in the past, scrambling was used to prevent unauthorized copying.

【0008】

This conventional technique has a problem that once the scramble is broken, unauthorized copying cannot be prevented thereafter.

【0009】

Further, for example, in order to prevent illegal copying of banknotes, securities, etc., for identification as shown in JP-A-4-351164, JP-A-6-22062, and JP-A-6-22119. A method of embedding information in the pixel component of an image has been proposed.

【0010】

The method of embedding the identification information in the pixel component of the image has a problem that the identification information can be easily removed by falsification.

【0011】

Therefore, in order to prevent unauthorized use and duplication of the digital image, a method of embedding special information (hereinafter, this information is referred to as electronic watermark data) in the digital image itself has been considered.

【0012】

Two types of electronic watermark data for such digital images are considered: visible electronic watermark data and invisible electronic watermark data.

【0013】

Visible electronic watermark data is obtained by synthesizing special characters or symbols with an image so that it can be visually perceived, which causes deterioration of image quality, but for users of digital images. It has the effect of visually appealing to prevent unauthorized distribution.

【0014】

An example of embedding visible electronic watermark data is shown in Japanese Patent Application Laid-Open No. 8-241403. In this method, when synthesizing visible electronic watermark data with respect to the original image, only the brightness of the pixel corresponding to the opaque portion of the electronic watermark data is changed, and the color component is not changed. The electronic watermark data is combined with the original image. At this time, the scaling value for changing the brightness component of the pixel is determined by the color component, the random number, the pixel value of the electronic watermark data, and the like.

【0015】

Invisible electronic watermark data is created by embedding electronic watermark data in an image in consideration of not degrading the image quality, and it is characterized by being visually undetectable because there is almost no deterioration in image quality. is there.

【0016】

By embedding special information that can identify the author as this electronic watermark data, it is possible to identify the author by detecting this electronic watermark data even after illegal copying. Is. In addition, if the non-copyable information is embedded, for example, when the playback device detects the non-copyable information, the user is notified that the data is copy-prohibited, or the copy prevention mechanism in the playback device is operated. , VTR, etc. can be restricted.

【0017】

One method of embedding invisible electronic watermark data in a digital image is to embed special information as electronic watermark data in the part of the pixel data that has little effect on image quality, such as the least significant bit (LSB). There is a way. However, for this method, it is easy to erase the electronic watermark data from the image. For example, if a low-pass filter is used, the LSB information of the pixel will be lost. Further, since the image compression process aims to reduce the amount of data by removing the redundant information of the portion having little influence on the image quality, the electronic watermark data is lost by the image compression process. Therefore, there is a problem that it becomes difficult to rediscover the electronic watermark data of the image that has undergone the image compression process.

【0018】

Therefore, a method of frequency-converting an image and embedding electronic watermark data in the frequency spectrum has been proposed (Nikkei Electronics 1996.4.22 (no.660), page 13). In this method, since the electronic watermark data is embedded in the frequency component, the electronic watermark data is not lost even for image processing such as compression processing and filtering. Furthermore, by adopting random numbers that follow a normal distribution as the electronic watermark data, it is possible to prevent interference between the electronic watermark data and make it difficult to destroy the electronic watermark data without significantly affecting the entire image.

【0019】

With reference to FIG. 10, the method of embedding electronic watermark data in this method is as follows. First, the original image is converted into frequency components using the discrete cosine transform means 1020. Select n pieces of data showing high amplitude in the frequency domain and use them as f (1), f (2), ..., f (n). Electronic watermark data w (1), w (2), ... w (n) are selected from a normal distribution with an average of 0 variances of 1, and the electronic watermark data embedding means 1030 is used. F (i) = f (i) + α | f (i) | * w (i) Is calculated for each i. Here, α is a scaling element. Finally, the image data in which the electron watermark data is embedded from the frequency component in which f (i) is replaced with F (i) is obtained by the inverse discrete cosine transform.

【0020】

The electronic watermark data is detected by the following method. In this detection method, the original image and the electronic watermark data candidate w (i) (where i = 1, 2, ..., n) must be known.

【0021】

Referring to FIG. 11, first, the image containing the electronic watermark data is converted into a frequency component by using the discrete cosine transform means 1120, and the frequency components F (1), F (2), ... F (n) are obtained. .. The original image data is also converted using the discrete cosine transform means 1110 to obtain frequency components f (1), f (2), ... f (n). The electronic watermark data estimate W (i) is calculated by f (i) and F (i). W (i) = (F (i) -f (i)) / f (i) Calculate and extract by. Next, the statistical similarity between w (i) and W (i) is calculated by the inner product calculation means 1140 by C = W * w / (WD * wD) using the inner product of the vectors. Where W = (W (1), W (2), ..., W (n)), w = (w (1), w (2), ..., w (n)), WD = Absolute value of vector W, wD = Absolute value of vector w. When C is equal to or higher than a certain value, the statistical similarity determination means 1160 determines that the corresponding electronic watermark data is embedded.

【0022】

Embedding electronic watermark data in an image using this method is effective when the author who owns the original image performs detection processing on digital image data that seems to be an illegal copy. FIG. 12 shows a conventional example of an image data encoding system incorporating this electronic watermark embedding means. The original image data is orthogonally converted from the time domain to the frequency domain by the discrete cosine transform means 1201. The electronic shaving data embedding means 1202 inserts the electronic shaving data 1203 into the data converted into the frequency domain by the discrete cosine transforming means 1201. The data into which the electronic shaving data is inserted is quantized by the quantization means 1204, and the quantized data is encoded by the coding means 1205 to become MPEG data.

【0023】

[Problems to be Solved by the Invention]

The conventional encoding system described above always embeds electronic watermark data. Since the electronic watermark data is embedded in the frequency component, the deterioration of the image quality is small, but it is not completely eliminated. Therefore, when encoding data that does not require embedding of electronic watermark data, especially when the image quality is particular, encode with another system that does not have electronic watermark data embedding means in order to avoid unnecessary deterioration of image quality. There was a problem that it had to be done.

【0024】

An object of the present invention is to provide an image input device that can identify even if image data is illegally copied, and creates image data whose identification information cannot be removed by falsification of the image data. To do.

【0025】

Another object of the present invention is to provide an encoding system for embedding electronic shaving data, which can also encode data without embedding electronic shaving data.

【0026】

[Means for solving problems]

The image data encoding system according to the present invention is an encoding system in which identification data (hereinafter referred to as electronic data) is embedded in an original image, and is converted by a discrete cosine transform means for discrete cosine transform of the original image and a discrete cosine transform means. It is characterized by having an electronic squeeze embedding means for inserting electronic shaving data into data, and a data selection means for selecting the output of the discrete cosine transform means and the output of the electronic shaving embedding means.

【0027】

Further, the image data encoding system according to the present invention further includes a flip-flop connected to the data selection means, and the data selection means uses the information stored in the flip-flop to electronically display the output of the discrete cosine transform means. It is characterized by selecting the output of the embedding means.

【0028】

Further, the image data encoding system according to the present invention is characterized in that the data selection means selects the output of the discrete cosine transform means and the output of the electronic squeeze embedding means by an external signal.

【0029】

Further, the image data encoding system according to the present invention is a digital encoding system that applies a predetermined coding process to the digital image data and outputs the digital image data, and is selected from a plurality of electronic shaving data for the digital image data. It is provided with a shaving data inserter for inserting electronic shaving data, and at least one of the plurality of electronic shaving data is such that the digital image data is invariant even if it is inserted into the digital image data. It is characterized in that it is a predetermined electronic squeeze data.

【0030】

Further, the image data encoding system according to the present invention is a digital encoding system that outputs digital image data by subjecting it to a predetermined coding process, and has a plurality of electronic shaving data for identifying the digital image data. A data table, a sushi data selector that selects one of the plurality of sushi data tables, and a sushi data inserter that inserts electronic sushi data selected for the digital image data are provided. At least one of the plurality of shaving data tables is characterized by having predetermined electronic shaving data such that the digital image data does not change even when inserted into the digital image data. To do.

【0031】

Further, the image data encoding system according to the present invention includes a conversion means for frequency-converting digital image input data and outputting it to the squid data inserter, and data in which electronic squid data is inserted by the squid data inserter. It is characterized by having a quantizer that quantizes the data and a variable length encoder that outputs the output data of the quantizer by variable length coding.

【0032】

Further, the image data encoding system according to the present invention is characterized in that the predetermined electronic shaving data is other than the random numbers of the generation algorithm using the normal distribution.

【0033】

The image input device according to the present invention has an imaging means for obtaining an analog image signal, an analog digital conversion means for analog-digitally converting the analog image signal obtained by the imaging means to obtain image data, and a first image data. A conversion means for converting into frequency region data, a storage means for temporarily storing the first frequency region data, an identification data holding means for holding the identification data, and the identification data are added to the first frequency region data. The second frequency region data is generated and output, and the selection means for selectively outputting one of the first frequency region data and the second frequency region data is provided.

【0034】

Further, the image input device according to the present invention is characterized in that the conversion means is an orthogonal converter.

【0035】

Further, the image input device according to the present invention is further provided with a compression means for compressing and coding the output of the selection means.

【0036】

Further, the image input device according to the present invention has an imaging means for obtaining an analog image signal, an analog digital conversion means for analog-digitally converting the analog image signal obtained by the imaging means, and obtaining image data, and the image data. A conversion means for converting into frequency region data of 1, a storage means for temporarily storing the image data, an identification data holding means for holding identification data, and the identification data are added to the first frequency region data, and the first One of the means for generating and outputting the frequency region data of 2, the inverse conversion means for inversely converting and outputting the second frequency region data into the time region data, and the output of the inverse conversion means and the output of the storage means. It is characterized by including a selection means for selective output.

【0037】

Further, the image input device according to the present invention is characterized in that the conversion means is an orthogonal converter and the inverse conversion means is an inverse orthogonal converter.

【0038】

Further, the image input device according to the present invention is further provided with a compression means for compressing and coding the output of the selection means.

【0039】

BEST MODE FOR CARRYING OUT THE INVENTION

Next, embodiments and examples of the present invention will be described in detail with reference to the drawings.

【0040】

[Embodiment 1] With reference to FIG. 1, the image data encoding system according to the first embodiment has a discrete cosine conversion means 101 for converting the input original image data from a time region to a frequency region, and a discrete cosine conversion means. Output of the electronic shaving data embedding means 102 that inserts the electronic shaving data into the data converted into the frequency region in 101, the electronic shaving data 103 that is inserted by the electronic shaving data embedding means 102, and the discrete cosine conversion means 101. Data selection means 106 that selects either 107 or the output 108 of the electron squeezing data embedding means 102, quantization means 104 that quantizes the data selected by the data selection means 106, and quantization means 104. It has a coding means 105 that encodes the obtained data and generates MPEG data.

【0041】

Next, the operation of the circuit of FIG. 1 will be described with reference to the drawings.

【0042】

The original image data is orthogonally converted from the time domain to the frequency domain by the discrete cosine transform means 101. The electronic watermark data embedding means 102 inserts the electronic watermark data 103 into the data converted into the frequency domain by the discrete cosine transform means 101.

【0043】

The output signal 108 of the electronic watermark data embedding means 102 serves as one input of the data selection means 106. Further, the output signal 107 of the discrete cosine transform means 101 serves as an input of the electronic watermark data embedding means 102 and also as the other input of the data selection means 106. The data selection means 106 selects the signal 108 when inserting the electronic watermark data into the original image data, and selects the signal 107 when the electronic watermark data is not inserted into the original image data.

【0044】

The data selected by the data selection means 106 is quantized by the quantization means 104, and the quantized data is encoded by the coding means 105 to become MPEG data.

【0045】

[Example 1] Next, a first embodiment corresponding to the first embodiment of the present invention will be described in detail with reference to the drawings.

【0046】

Referring to FIG. 2, the selection of the output signal 107 of the discrete cosine transformer 101 and the output signal 108 of the electronic squeeze data embedding 102 is made by the selector 110 operated by the information stored in the flip-flop 111. The flip-flop 111 is set with a logical value of 0 when the electronic watermark data is not inserted into the image data, and a logical value of 1 when the electronic watermark data is inserted into the original image data.

【0047】

The original image data is orthogonally converted from the time domain to the frequency domain by the discrete cosine transformer 101. The electronic watermark data embedding device 102 inserts the electronic watermark data 103 into the data converted into the frequency domain by the discrete cosine transformer 101.

【0048】

The output signal 108 of the electronic watermark data embedding device 102 serves as one input of the selector 110. Further, the output signal 107 of the discrete cosine transformer 101 serves as the input of the electronic watermark data embedding device 102 and also as the other input of the selector 110. The selector 110 selects the signal 107 when the information of the flip-flop 111 has a logical value of 0, and selects the signal 108 when the information of the flip-flop 111 has a logical value of 1.

【0049】

The data selected by the selector 110 is quantized by the quantizer 104, and the quantized data is encoded by the encoder 105 to become MPEG data.

【0050】

[Example 2] Next, a second embodiment corresponding to the first embodiment of the present invention will be described in detail with reference to the drawings.

【0051】

Referring to FIG. 3, the selection of the output signal 107 of the discrete cosine transformer 101 and the output signal 108 of the electronic squeeze data embedding device 102 is performed by the external signal 112 in the selector 110. When the electronic shaving data is not inserted in the image data, the logical value 0 is set in the external signal 112, and when the electronic shaving data is inserted in the original image data, the logical value 1 is set in the external signal 112.

【0052】

The original image data is orthogonally converted from the time domain to the frequency domain by the discrete cosine transformer 101. The electronic watermark data embedding device 102 inserts the electronic watermark data 103 into the data converted into the frequency domain by the discrete cosine transformer 101.

【0053】

The output signal 108 of the electronic watermark data embedding device 102 serves as one input of the selector 110. Further, the output signal 107 of the discrete cosine transformer 101 serves as an input of the electronic watermark data embedding device 102 and also as one input of the selector 110. The selector 110 selects signal 107 when the external signal 112 has a logical value of 0, and selects signal 108 when the external signal 112 has a logical value of 1.

【0054】

The data selected by the selector 110 is quantized by the quantizer 104, and the quantized data is encoded by the encoder 105 to become MPEG data.

【0055】

[Embodiment 2] Next, the image data encoding system according to the second embodiment will be described with reference to the drawings.

【0056】

FIG. 4 is a block diagram showing a configuration of an image data encoding system according to the present embodiment. In FIG. 4, the image data encoding system according to the present embodiment includes a discrete cosine conversion means 402 that performs discrete cosine conversion processing on the original image stream 401 to be encoded, and a plurality of electronic shaving data tables having electronic shaving data. 408 (0), 408 (1), 408 (2), ..., 408 (n), electronic shaving data selector 407 that selectively selects the electronic shaving data table, and discrete cosine conversion means. Electronic shaving data embedding means 404 that inserts electronic shaving data into the data output from 402 and temporarily held in buffer 404, and quantum that quantizes the data output by electronic shaving data embedding means 404. It includes a conversion means 405 and a coding means 406 that encodes the data output by the quantization means 405 in a variable length and outputs it as MPEG data 409.

【0057】

Of the multiple electronic shaving data tables 408 (0) to 408 (n), the electronic shaving data table 408 (0) is not a random number of the generation algorithm using a normal distribution, but can be inserted into digital image data. It is a table having electronic data such that the digital image data is invariant. On the other hand, the electronic watermark data tables 408 (1) to 408 (n) are random numbers of the generation algorithm using the normal distribution.

【0058】

Next, the operation of the image data encoding system according to the present embodiment will be described.

【0059】

First, a case of inserting ordinary electronic watermark data will be described. The original image data 401 is taken out in blocks of 8 × 8 pixels based on a normal MPEG compression process. The discrete cosine transform means 402 performs a discrete cosine transform process on the extracted data to perform frequency conversion. The electronic watermark data selection means 107 selects the electronic watermark data from the electronic watermark data tables 408 (1) to 408 (n) excluding the electronic watermark data table 408 (0), and the electronic watermark data. Output to the embedding means 404. The electronic watermark data embedding means 404 inserts selected electronic watermark data into the data that has been subjected to the discrete cosine transform process and converted into frequency components. The quantization means 405 performs a quantization process on the data output from the electron squeezing data embedding means 404. The coding means 406 encodes the data output by the quantization means 405 and outputs it as MPEG data 409.

【0060】

Next, a case where encoded data based on the original data is required will be described. Similar to the above-mentioned normal case, the original image data 401 is taken out for each block of 8 × 8 pixels based on the normal MPEG compression processing. The discrete cosine transform means 402 performs a discrete cosine transform process on the extracted data to perform frequency conversion. Then, the electronic watermark data selection means 407 selects electronic watermark data from the electronic watermark data table 408 (0) so that the digital image data does not change even when inserted into the digital image data, and the electronic watermark data is electronically watermarked. Output to data embedding means 404. The electronic watermark data embedding means 404 inserts selected electronic watermark data into the data that has been subjected to the discrete cosine transform process and converted into frequency components. The quantization means 405 performs a quantization process on the data output from the electron squeezing data embedding means 404. The coding means 406 encodes the data output by the quantization means 405 and outputs it as MPEG data 409.

【0061】

[Embodiment 3] Next, the basic configuration of the image input device according to the third embodiment will be described in detail with reference to the drawings. Referring to FIG. 5, the imaging means 501 for capturing an image of the outside world, the analog digital conversion means 502 for converting the captured analog image data into digital image data, and the image data being converted from the spatial region to the frequency region. The conversion means 503, the storage means 507 that temporarily stores the image data converted into the frequency region, the identification data holding means 510 that holds the identification data, and the identification data in the image data converted into the frequency region. Identification data embedding means 509 to be embedded, output of storage means 507 and data selection means 508 to select and output the output of identification data embedding means 509, quantization means 504 to quantize image data, and quantized image data. The coding means 505 for variable length coding is provided.

【0062】

Next, the operation of the image input device according to the present embodiment shown in FIG. 5 will be described. The imaging means 501 captures an image of the outside world and outputs image data. The analog-to-digital conversion means 502 converts the captured analog image data into digital image data and outputs the data. The conversion means 503 orthogonally converts the image data from the spatial domain to the frequency domain and outputs the image data. The storage means 507 temporarily stores the image data converted into the frequency domain. The identification data holding means 510 holds and outputs the identification data. The identification data embedding means 509 embeds the identification data in the image data converted into the frequency domain and outputs the identification data. The data selection means 508 selects and outputs the output of the storage means 507 or the output of the identification data embedding means 509. The quantization means 504 quantizes the image data and outputs it. The coding means 505 encodes the quantized image data in a variable length and outputs the compressed image data.

【0063】

[Example 3] Next, a third embodiment corresponding to the third embodiment of the present invention will be described in detail with reference to the drawings. Referring to FIG. 6, the CCD image sensor 601 captures an image of the outside world and outputs image data. The analog-to-digital converter 602 converts the captured analog image data into digital image data and outputs the data.

【0064】

The discrete cosine transformer 603 orthogonally converts the image data from the spatial domain to the frequency domain and outputs the image data. The buffer 607 temporarily stores the image data converted into the frequency domain. The identification data table 610 holds and outputs the identification data. The identification data embedding device 609 embeds the identification data in the image data converted into the frequency domain and outputs the identification data.

【0065】

The selector 608 selects and outputs either the output of the buffer 607 or the output of the identification data embedding device 609. Here, when the output of the buffer 607 is selected, the original image data is output, and when the output of the identification data embedding device 609 is selected, the image data in which the identification data is embedded is output.

【0066】

The quantizer 604 quantizes and outputs image data. The variable length encoder 605 encodes the quantized image data in a variable length and outputs the MPEG data. The generated MPEG data is transmitted to a device (not shown), for example, a personal computer, a storage media processing device such as an optical magnetic medium, a network processing device that performs transmission processing to a network line, or a wireless line. It is sent to a wireless media processing device.

【0067】

Next, the method of embedding the identification data will be described with reference to FIG. 7. When the image data is orthogonally transformed from the spatial domain to the frequency domain by the discrete cosine transformer 603, a frequency spectrum is generated. This spectrum can be shown as shown in 701 in FIG. The identification data table 610 outputs identification data of a similar frequency spectrum, which spectrum can be shown as 704 in FIG. When the frequency spectrum 701 of the original image and the frequency spectrum 704 of the identification data are added by the adder 702, the frequency spectrum 703 in which the identification data is embedded is obtained. The above is the description of the method of embedding the identification data.

【0068】

When extracting the identification data, the identification data can be easily obtained by subtracting the original frequency spectrum 701 from the frequency spectrum 703 in which the identification data is embedded by a subtractor (not shown) to obtain the frequency spectrum 704 of the identification data. Can be extracted.

【0069】

[Example 4] Next, a fourth embodiment corresponding to the third embodiment of the present invention will be described in detail with reference to the drawings. Referring to FIG. 8, the CCD image sensor 601 captures an image of the outside world and outputs image data. The analog-to-digital converter 602 converts the captured analog image data into digital image data and outputs the data. The buffer 807 temporarily stores the digitally converted image data.

【0070】

The discrete cosine transformer 603 orthogonally converts the image data from the spatial domain to the frequency domain and outputs the image data. The identification data table 610 holds and outputs the identification data. The identification data embedding device 609 embeds the identification data in the image data converted into the frequency domain and outputs the identification data. The inverse discrete cosine transformer 811 converts the image data from the frequency domain to the spatial domain and outputs it. Here, the transformation between the spatial domain and the frequency domain may be a fast Fourier transform in addition to the discrete cosine transform.

【0071】

The selector 808 selects and outputs either the output of the buffer 807 or the output of the inverse discrete cosine transformer 811. Here, when the output of the buffer 807 is selected, the original image data is output, and when the output of the inverse discrete cosine transformer 811 is selected, the image data in which the identification data is embedded is output.

【0072】

Since the method of embedding the identification data is the same as that of the third embodiment, the description thereof will be omitted. When extracting the identification data, the image data is orthogonally converted from the spatial domain to the frequency domain by the discrete cosine transform means, and then the original frequency spectrum is used from the frequency spectrum in which the identification data is embedded by a subtractor (not shown). The identification data can be extracted by subtracting the above to obtain the frequency spectrum of the identification data.

【0073】

[Effect of the invention]

As described above, according to the present invention, encoding of both image data that needs to be embedded with electronic shaving data and image data that does not need to be embedded can be easily realized by one system, so that the encoding is performed according to the image data. There is no need to prepare two systems, and the hardware scale can be significantly reduced.

【0074】

Further, according to the present invention, even if the image data is illegally copied, it can be identified. The reason is that it is possible to elucidate the route by investigating the embedded identification data.

【0075】

Further, if the image data is modified for the purpose of deleting or destroying the identification data, the original image itself is significantly deteriorated, so that it is possible to prevent the image data from being illegally copied by falsification.

[Simple explanation of drawings]

[Figure 1]

It is a figure which shows the structure of the image data encoding system by Embodiment 1 of this invention.

[Figure 2]

It is a figure which shows the structure of the image data encoding system according to Example 1 of this invention.

[Fig. 3]

It is a figure which shows the structure of the image data encoding system according to Example 2 of this invention.

[Fig. 4]

It is a figure which shows the structure of the image data encoding system by Embodiment 2 of this invention.

[Fig. 5]

It is a figure which shows the structure of the image input device according to Embodiment 3 of this invention.

[Fig. 6]

It is a figure which shows the structure of the image input device according to Example 3 of this invention.

[Fig. 7]

It is a figure explaining the embedding of the identification data in the frequency domain in Example 3 of this invention.

[Fig. 8]

It is a figure which shows the structure of the image input device according to Example 4 of this invention.

[Fig. 9]

It is a block diagram which shows an example of the conventional image input device.

[Fig. 10]

It is a figure explaining the method of embedding electronic watermark data of a conventional example.

[Fig. 11]

It is a figure explaining the detection method of the electronic watermark data of the conventional example.

[Fig. 12]

It is a figure which shows the structure of the image data encoding system by a conventional example.

[Explanation of symbols]

101, 402 Discrete cosine transform means 102, 404 Electronic watermark data embedding method 103 Electronic watermark data 104, 405, 504 Quantization means 105, 406, 505 coding means 106, 508 Data selection method 407 Electronic Watermark Data Selector 408 (0), 408 (1), 408 (2), ..., 408 (n) Electronic watermark data table 501 Imaging Means 502 Analog-to-digital conversion means 503 Conversion means 507 Memory means 509 Identification data embedding method 510 Identification data retention means