JP2000022930A - Data embedding device and data generating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To relate main data and subordinate data to each other without increasing the number of files. SOLUTION: The data embedding device includes devices 40 and 42 for inputting the main data, devices 48 and 50 for inputting the subordinate data, and devices 44 and 52 which divides the subordinate data by specific bits and embeds them as electronic watermark information in respective data blocks while dividing the inputted. main data into the data blocks and performing a signal processing. The data generating device includes a device 78 which decodes the respective blocks of the main data with the electronic watermark into the subordinate data by the specific bits in order and devices 72, 74, and 76 which restores the respective data blocks to the main data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for adding sub data to main data such as image data, and more particularly to an apparatus for adding sub data such as voice and text data to a part of image data. Device.

[0002]

2. Description of the Related Art In a technology called the world wide web of the Internet, an image is displayed on a display surface of a personal computer, and by specifying a part of the image, information on a region of the specified image is obtained. There is a data search method in which another information previously linked to the site is obtained from a remote place or a data file of the personal computer or a network to which the personal computer belongs, and the information is presented. The information to be presented includes another image, sound, moving image, text data, and the like.

[0003] Using such a technique, for example, a method of storing all necessary information data files in a CD-ROM and distributing the data files is not limited to the Internet. There is a need for technology to attach.

[0004]

However, in such a technique, since an image and data linked to the image are separate, for example, there is a problem that the number of necessary data files increases. When an image is displayed on a computer, additional information can be obtained through the computer because a link to the additional information is held by the computer. In such a case, there is a problem in that additional information cannot be obtained because the link information is lost.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a data embedding device and a data generating device capable of associating main data with sub data without increasing the number of files.

Another object of the present invention is to provide a data embedding device and a data generating device capable of associating main data with voice data or character data without increasing the number of files.

Yet another object of the present invention is to provide a method for associating main data with sub data without increasing the number of files, and when such main data is once fixed on a medium other than an electronic medium. It is another object of the present invention to provide a data embedding device and a data generating device capable of restoring sub data.

[0008]

According to a first aspect of the present invention, there is provided a data embedding apparatus, comprising: a main data fetching means for fetching main data; a sub data fetching means for fetching sub data. A digital watermark for dividing the main data into a plurality of predetermined data blocks and performing predetermined signal processing while dividing the sub-data into predetermined bits and embedding each data block as digital watermark information. Embedding means.

[0009] By dividing the sub-data for each predetermined bit and embedding it in the data block of the main data, a relatively large amount of sub-data as a whole can be embedded as a digital watermark.

According to a second aspect of the present invention, in addition to the configuration of the first aspect, the digital watermark embedding means divides the sub-data into predetermined bits to form a plurality of sub-data. It is characterized by including means for embedding digital watermark information only in a part of a data block.

Since the sub data can be divided into predetermined bits and embedded in some of the main data blocks, the sub data can be embedded in only a part of the main data, Or other sub-data can be embedded in the area of

A data embedding device according to a third aspect of the present invention is characterized in that, in addition to the configuration of the first or second aspect, the sub-data includes audio data.

A data embedding device according to a fourth aspect of the present invention is characterized in that, in addition to the configuration of the first or second aspect, the sub-data includes character data.

According to a fifth aspect of the present invention, there is provided a data generating apparatus including a plurality of data blocks subjected to a predetermined process, and each data block includes sub-data divided for each predetermined bit. Main data preparation means for preparing main data with a digital watermark embedded as information; digital watermark decoding means for decoding sub-data for each predetermined bit from the main data with a digital watermark; Data processing means for restoring main data by performing a process determined corresponding to a predetermined process on each of the data blocks subjected to the predetermined process.

The sub data is divided into predetermined bits, embedded in the data blocks of the main data, and decoded, whereby a relatively large amount of sub data can be restored as a whole.

According to a sixth aspect of the present invention, in addition to the configuration of the fifth aspect of the present invention, the main data with the electronic watermark is included in only a part of the plurality of data blocks. The digital watermark decoding means includes sub data, and the digital watermark decoding means includes means for decoding the sub data from a part of the plurality of data blocks for each predetermined bit.

By dividing the sub-data into predetermined bits and embedding it in some of the data blocks of the main data, the sub-data can be decoded from only a part of the main data, It is possible to decode another sub-data from the area of.

According to a seventh aspect of the present invention, in addition to the configuration of the fifth or sixth aspect of the present invention, the sub-data includes audio data.

An eighth aspect of the present invention is directed to the data generation apparatus according to the fifth aspect, wherein the sub-data includes character data in addition to the configuration of the fifth or sixth aspect.

[0020]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In order to solve the above-mentioned problem,
It is conceivable to use a so-called digital watermark technique.
Digital watermarking is a technique for preventing image data of works of art, such as photographs, digitized in accordance with the trend of digitization of data in recent years, from being illegally copied, diverted or distributed.

In the digital watermark technique, watermark information having a small data amount is embedded in the entire image for copyright protection.
Just by looking at the image after embedding, it cannot be recognized that such watermark information is embedded therein, but the watermark information can be restored by performing electronic processing. Therefore, a method of embedding additional information in an image using a digital watermark technique is conceivable.

However, since the digital watermarking technique is mainly used for copyright protection, a small amount of watermark information is embedded in the entire screen so that the watermark information can be taken out wherever the picture is taken out. . Therefore, when the additional information has a relatively large data amount, such as audio data, it is difficult to embed such additional information in the image data. Even if the data can be embedded, the amount of additional information is large, so that noise in the image data increases and the image deteriorates.

Therefore, in the present invention, JPEG (Joint
Photographic Experts Group) DCT for data in image processing technology such as image compression represented by the standard
In a system in which orthogonal transform processing such as (discrete cosine transform) is performed, this problem is solved as follows, focusing on the fact that data is divided into a large number of blocks (called DCT blocks in the case of DCT). did. In addition, an FFT (Fast Fourier Transform) and the like are often used in addition to the DCT as the orthogonal transform processing.

Referring to FIG. 1, a case where audio data 20 is embedded in image data 22 will be considered. The image data 22 is divided into a plurality of DCT blocks. The audio data 20 also
Divided into blocks A, B, C, etc. in a fixed bit unit,
These are sequentially embedded as watermark data a, b, c, etc. in the DCT block of the image data 22 in block units.

Referring to FIG. 2, the following is performed at the time of restoration. The image data 22 is divided into a plurality of DCT blocks. So, from each of these DCT blocks,
The blocks a, b, c, etc. of the audio data embedded as the watermark data are extracted, and the blocks A,
The original audio data 20 is restored by arranging B, C,. Note that the order of embedding is not necessarily the order from the beginning.
If not restricted by resources such as a memory, each block of the sub data may be replaced in a predetermined order and then embedded in the main data. In this case, it is necessary to reverse the order of the sub data at the time of restoration, contrary to the embedding.

FIG. 3 shows a block diagram of a data embedding device 30 according to an embodiment of the present invention. In the following description, a case where audio data is embedded in image data will be described as an example. However, the present invention is not only applicable to such a combination, but generally, data which is secondary to main data is The present invention can be applied to any data that is added and that divides the main data into predetermined blocks when encoding the main data. For example, the sub data may be character data, or may be image data different from the image of the main data. However, since the amount of data that can be embedded is smaller than the amount of main data, the size of the sub data is naturally limited.

Referring to FIG. 3, data embedding device 30
A data embedding unit 32 for embedding audio data in image data, a memory device 36 for storing image data in which audio data is embedded, and a memory device 36
And a data restoration unit 34 for performing a process of restoring the original image and the audio data as the additional data from the image data stored in.

The data embedding unit 32 includes an image capturing device 40 such as a scanner and a camera for capturing an image as an electric signal, and a signal processing device 42 for performing predetermined signal processing on the electric signal. JPEG compression unit 44 for performing image compression processing according to the JPEG standard
And a memory writing device 46 for writing the compressed image data to the memory device 36, and a switch 54 for switching functions as described later. The data embedding unit 32 further includes a voice capturing device 4 such as a microphone.
8 and the electric signal output from the voice capturing device 48 is A / D
A / D device 50 for (analog / digital) conversion
And a digital watermark writing device 52 for embedding the digitized audio data in the image data as a digital watermark.

[0029] The JPEG compression section 44 is a DPEG for performing a DCT operation on a data block (DCT block).
A CT operation device 60, a quantization device 62 for performing a quantization process on the DCT-processed data block, a switch 66 for switching the connection according to the function, and D
A Huffman encoding device 64 for performing a Huffman encoding process on the data block subjected to the CT process.

The data restoration unit 34 includes a memory reading device 70 for reading data from the memory device 36, a JPEG decompression unit 72 for performing JPEG decompression processing on the read image data, and a restored image data. Display device 74 for performing data processing for displaying the image, a monitor device 76 for displaying an image signal output from the image display device 74, a switch 84 for switching connections according to functions, a memory device A digital watermark reading device 78 for extracting and restoring watermark information from the image data read out from the image data 36, and a D / A for digital / analog (D / A) conversion of the restored watermark information (audio data). Device 80 and an audio output device 82 such as a speaker that reproduces an analog signal output from the D / A device as audio.
And

The JPEG decompression unit 72 includes a Huffman decoding device 90 for performing a Huffman decoding process on a data block of a read image, a switch 96 for switching connections according to functions, and a Huffman decoding device. Inverse quantization device 9 for performing an inverse quantization process on the obtained data
2 and ID for the data block subjected to the inverse quantization
IDCT for performing CT (Inverse Discrete Cosine Transform) processing
And an arithmetic unit 94.

The operation of the data embedding device 30 shown in FIG. 3 will be described with reference to FIGS. 4 and 5.
First, an image is captured by the image capturing device 40 (100). Subsequently, the image processing is performed on these image signals by the signal processing device 42 (102), and the JPE
The additional information related to the digital watermark is written in the G header (104). The data is further divided into DCT block units and transferred to the JPEG compression unit 44 (106).

Hereinafter, processing relating to JPEG compression will be described later. In parallel with the image capturing process, voice capturing by the voice capturing device 48 is also performed (1).
14). The sound and the image may be prepared at the same time, for example, when the image and the sound are simultaneously captured by an electronic still camera, or may be used when adding a memo by a user's voice to commercially available image data. As such, it may be prepared separately. In this case, it is not necessary to capture the image and the audio image in parallel.

The acquired audio data is converted into digital data by the A / D device 50 (116), and the digitized audio data is converted into n-bit units (n is a predetermined constant. For example, n = 8). ) To the electronic watermark writing device 52. This data is the JPEG described below.
Used in compression processing and embedded in image data.

In the processing by the JPEG compression section 44, the DCT operation device 60 first performs a DCT operation on the transferred DCT block (108). The data after the DCT operation is quantized by the quantization device 62 and the switch 66
Given to. When the digital watermark embedding process is performed, the switch 66 gives the output of the quantization device 62 to the digital watermark writing device 52 via the terminal 66C (“ON” in step 112 of FIG. 4), and does not perform the embedding process. In this case, the signal is supplied to the Huffman encoding device 64 via the terminal 66a ("OFF" in step 112 of FIG. 4). When the embedding process is not performed, the process is the same as the normal process.

In this case, the digital watermark writing device 52
In one DCT block of image data output from the quantization device 62, n bits of audio data transferred from the A / D device 50 are embedded as watermark information (1).
20). Huffman encoding processing is further performed on the image data including the audio data embedded as the watermark information by the Huffman encoding device 64 (122), and the Huffman encoded image data is stored in the memory writing device 46.
Is written to the memory device 36. Processed DC
It is determined whether the T block is the last block (12).
4) If it is the last block, the process ends. If the processed DCT block is not the last block, the control is returned to step 108, and the above-described processing is repeated for the subsequent DCT block.

Thus, the audio data divided into n-bit blocks is embedded in each DCT block of the image data, and image data having additional data is created and stored in the memory device 36.

Next, a description will be given of a process of restoring audio from image data having audio data as additional data (digital watermark). Referring to FIGS. 3 and 5,
First, the memory reading device 70 fetches image data from the memory device 36 (140). This image data includes audio data as additional data (digital watermark) as described above.

The Huffman decoding device 90 reads the header of this data (142) and performs the following processing. The switch 84 selects the terminal 84a so as to provide the output of the Huffman decoding device 90 to the digital watermark reading device 78.
If the header contains additional information regarding the digital watermark, the Huffman decoding device 90 performs Huffman decoding processing on the image data for each block (144), and the Huffman-decoded data is output to the digital watermark reading device. 78
Given to.

The digital watermark reading device 78 receives the given D
The voice data added is extracted from the data of the CT block (148). The audio data is transferred every n bits (156), applied to the D / A device 80, and converted into an analog signal (158).
(160).

On the other hand, the digital watermark reading device 78 supplies the image data to the inverse quantization device 92 via the switch 96. The inverse quantization device 92 performs an inverse quantization process on the image data (150). The inversely quantized image data is provided to the IDCT operation device 94, where an IDCT process is performed (152). The image data thus obtained is provided to the monitor device 76 via the image display device 74 and displayed.

It is determined whether the DCT block thus processed is the last block (154). If the DCT block is the last block, the process is terminated. If it is not the last block, control returns to step 148, and restoration of image data and audio data for each DCT block is repeated until all DCT blocks have been processed.

In this manner, audio data can be embedded in image data as a digital watermark. The audio data is
It can be embedded in the entire image or in only a part of the image. For example, consider a case where data is embedded in image data having a size of 2 megabits. It is assumed that additional data is embedded in each DCT block in 8-bit data units. Assuming that one DCT block is 8 × 8 64 bits, the number of DCT blocks is 2,000,000 / 64 = 31.
250. Therefore, 3125 is added as additional data.
Data of 0 × 8 = 25000 (bits) = 31.25 (K bytes) can be embedded.

When the additional data is audio, if the sampling rate is 8 kHz, approximately 4 seconds of audio can be embedded. If the additional data is character data, about 3
0000 characters can be embedded.
Bytes).

The data embedding device 30 shown in FIG.
Also, as shown in FIG. 7, additional data is embedded as electronic watermarks 172, 174, and 176 only in a part of the image 170, and can be used when reproducing the additional data by pointing with a pointer 178 as shown in FIG. . The operation in that case will be described with reference to FIG.

First, the image data is read, and the information of the header is read (190). Then, additional information relating to the digital watermark included in the header is read, and the image is restored by the same operation as the operation of FIG. 5 described above. However, in the present embodiment, the restoration of the image does not restore the additional data, but only reads the additional information. JPEG decompression unit 7
2 to decompress the compressed image (192). The expanded image is displayed by the image display device 74 and the monitor device 76 (194).

In the image displayed in this way, the area in which the digital watermark information is embedded is designated by the pointer 178 as shown in FIG. 7 (196). Hereinafter, the restoration processing of the digital watermark information is started. First, the DCT of the specified area
The block is searched (198). DCT processing is performed on the retrieved DCT block (200), and quantization processing is further performed (202) to decode the watermark information (2).
04), the audio data is decoded from the decoded watermark information (206). Further, it is determined whether or not there are any more blocks in the designated area (208). If there are, the control returns to step 198 to repeat the decoding of the audio information. When all DCT blocks included in the designated area have been processed, the decoded sound is reproduced (2).
10) End the process.

By using this function, for example, the following items can be realized. Referring to FIG. 9, different voices are embedded as digital watermarks 172, 174, and 176 in some areas of image 170. For example, in the range of the digital watermark 172, the voice “A”, in the range of the digital watermark 174, the voice of “Yes”, in the range of the digital watermark 176, the voice of “U”,
Each is embedded as a digital watermark. When this image is displayed on the screen of the personal computer and an arbitrary range is designated by the pointer 178, the watermark information embedded in the range can be decoded, and the embedded audio can be output. This makes it possible to embed sound corresponding to the image portion in each range of the image and output various sounds in accordance with the designation of the user using one image data.

As described above, according to the present invention, additional data such as audio data added to main data such as image data is embedded in the image data. Therefore, the number of files can be reduced as compared with the case where these data files are separated. Further, since the digital watermarking technique is used, even if the additional data is embedded, the data amount of the file hardly increases. Further, as a characteristic of the digital watermark applied to the image, there is a characteristic that even if the image is once printed out, the watermark information can be taken out again by taking it in with a scanner or the like. Due to this characteristic, data such as sound embedded in the printed image can be restored only by sending the printed image. Further, by embedding watermark information in an arbitrary part of an image, it is possible to embed a plurality of types of additional data in one image data.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing a method of embedding audio data in image data for explaining the principle of the present invention.

FIG. 2 is a schematic diagram showing a method for restoring audio data from image data for explaining the principle of the present invention.

FIG. 3 is a block diagram of a data embedding device according to one embodiment of the present invention;

FIG. 4 is a flowchart of audio data embedding processing.

FIG. 5 is a flowchart of audio data extracting processing.

FIG. 6 is a diagram schematically illustrating audio data embedded in a part of an image;

FIG. 7 is a diagram schematically illustrating a process of extracting audio data embedded in a part of an image;

FIG. 8 is a flowchart of a process of extracting audio data embedded in a part of an image.

FIG. 9 is a schematic diagram illustrating a specific example of a process of extracting audio data embedded in a part of an image;

[Explanation of symbols]

 Reference Signs List 30 data embedding device 36 memory device 40 image capturing device 44 JPEG compression unit 48 audio capturing device 52 digital watermark writing device 72 JPEG decompression unit 74 image display device 76 monitoring device 78 digital watermark reading device 82 audio output device

Claims (8)

[Claims] A main data fetching means for fetching main data; a sub data fetching means for fetching sub data; dividing the fetched main data into a plurality of predetermined data blocks; A digital watermark embedding unit for dividing the sub-data into predetermined bits and performing embedding as digital watermark information in the data block while performing predetermined signal processing. 2. The digital watermark embedding means includes means for dividing the sub-data for each predetermined bit and embedding each of the plurality of data blocks as digital watermark information only in a part of the plurality of data blocks. 2. The data embedding device according to 1. 3. The data embedding device according to claim 1, wherein the sub data includes audio data. 4. The data embedding device according to claim 1, wherein the sub data includes character data. 5. Main data with a digital watermark including a plurality of data blocks subjected to a predetermined process, and in each data block, sub data divided for each predetermined bit is embedded as digital watermark information. Main data preparing means for preparing the data, digital watermark decoding means for decoding sub data from the main data with digital watermark for each of the predetermined bits, and for each of the data blocks subjected to the predetermined processing A data processing unit for restoring the main data by performing a process determined corresponding to the predetermined process. 6. The digital watermark-attached main data includes the sub-data only in a part of the plurality of data blocks, and the digital watermark decoding unit includes: The data generation device according to claim 5, further comprising: means for decoding the sub-data from some of the data blocks for each of the predetermined bits. 7. The data generation device according to claim 5, wherein the sub data includes audio data. 8. The data generation device according to claim 5, wherein the sub data includes character data.