JP2000224045A - Vector quantizing device and vector quantizing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a device capable of accurately grasping the illegal use of information by selecting any of plural pieces of index data, whose error to original waveform segmenting data comes within a prescribed error range when decoded when the index data of waveform segmenting data being non- embedding object of watermark information is selected among the waveform segmenting data. SOLUTION: A waveform segmenting device 11 performs waveform segmenting on input signal data DIN and outputs it as waveform segmented data DP to the selector 27 of an index generating part 12. Meanwhile, the embedding position information storing part 21 of the part 12 outputs stored watermark embedding position information data DFP to the selector 27 and a header information generating part 28. Thus, the selector 27 outputs the data DP that does not correspond to an embedding position based on the data DFP to a 1st encoder 25, and a watermark information storing part 23 successively outputs stored watermark information data DWM to a 2nd encoder 26.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vector quantization apparatus and a vector quantization method, and more particularly to embedding watermark information in a coded bit sequence when coding a digital signal such as voice or image. The present invention relates to a vector quantization device and a vector quantization method.

2. Description of the Related Art As a conventional vector quantization apparatus, watermark information is embedded in designated bits in a 5-bit index by embedding position designation information in a digital code string composed of vector-quantized indexes, and a fixed index is set. , A quantization vector which minimizes distortion with respect to an input vector is selected and an index value thereof is output (Japanese Patent Laid-Open No. 10-224).
342).

[0002]

In this case, since the embedding position is fixed, the data and watermark information obtained by encoding the same finite length input signal sequence by embedding the watermark information are encoded. By comparing the data obtained by encoding without embedding the data, the location where the watermark information is embedded is identified from the difference data. Therefore, an object of the present invention is to provide a vector quantization apparatus and a vector quantization apparatus capable of preventing the embedding position of watermark data from being easily specified and accurately grasping the unauthorized use of information after vector quantization. It is to provide a method.

[0003]

According to the first aspect of the present invention,
In a vector quantization apparatus that vector-quantizes input signal data, of the waveform cutout data obtained by performing the waveform cutout of the input signal data, an index for the waveform cutout data of the non-embedding target of the watermark information is provided. When selecting data, any one of a plurality of index data in which an error with the original waveform cut data when decoding is within a predetermined error range is used as index data corresponding to the waveform cut data. It is characterized by having random selection means for selecting at random.

According to a second aspect of the present invention, there is provided a vector quantization apparatus for quantifying a conjugate vector of input signal data.
When selecting a combination of index data with respect to the waveform cutout data for which the watermark information is not embedded, out of the waveform cutout data obtained by performing the waveform cutout of the input signal data, Any one of a plurality of combinations of index data whose error with the waveform cut data is within a predetermined error range is randomly selected as a combination of index data corresponding to the waveform cut data. A combination random selection means is provided.

According to a third aspect of the present invention, in the configuration of the first or second aspect, when generating the waveform cut data, dummy data is added to a part of the waveform cut data. It is characterized by having means.

According to a fourth aspect of the present invention, in the configuration of the first aspect, first to n-th vector quantization means (n: an integer of 2 or more) for performing multi-stage vector quantization are provided. It is characterized by that.

According to a fifth aspect of the present invention, in the vector quantization device according to the fourth aspect, at least one of the first vector quantization means to the n-th vector quantization means includes the vector quantization means. When generating the waveform cutout data, a dummy data adding means for adding dummy data to a part of the waveform cutout data is provided.

According to a sixth aspect of the present invention, in the configuration of the third or fifth aspect, the waveform cutout data is constituted by a predetermined finite number of sample data, and the dummy data adding means includes The input signal data is divided into divided data composed of a predetermined number of sample data smaller than the finite number, and the dummy data is added to each divided data to constitute the finite number of sample data. It is characterized in that waveform cutout data is generated.

According to a seventh aspect of the present invention, in the configuration of the sixth aspect, the number of the sample data in the dummy data is equal to the number of the divided data for embedding watermark information and the number of the non-embedded watermark information. It is characterized in that it differs from the divided data. According to a configuration of claim 8, in the configuration of claim 6, as the dummy data,
It is characterized in that sample data having a predetermined constant value is repeatedly added to the divided data until the number of sample data constituting the divided data reaches the finite number.

[0010] According to a ninth aspect of the present invention, in the configuration of the sixth aspect, sample data constituting the waveform cut-out data obtained by performing a predetermined operation based on the divided data as the dummy data. And operation data composed of a number of sample data corresponding to the difference between the number of sample data and the number of sample data constituting the divided data.

A tenth aspect of the present invention is characterized in that, in the configuration of the ninth aspect, the operation data is calculated by an extrapolation operation.

[0012] According to an eleventh aspect of the present invention, in the configuration of the ninth aspect, the operation data is calculated by an interpolation operation.

According to a twelfth aspect of the present invention, in the vector quantization method for vector-quantizing input signal data, of the waveform cut-out data obtained by performing the waveform cut-out of the input signal data, the non-watermark information is not included. When selecting the index data for the waveform cutout data to be embedded, when decoding, any one of a plurality of index data whose error from the original waveform cutout data is within a predetermined error range. The method further comprises a random selection step of randomly selecting index data corresponding to the waveform cutout data.

According to a thirteenth aspect of the present invention, in the vector quantization method for conjugate vector quantization of input signal data, of the waveform cutout data obtained by performing the waveform cutout of the input signal data, When selecting a combination of index data for non-embedded waveform cutout data, from among a plurality of index data combinations in which the error with the original waveform cutout data when decoding is within a predetermined error range A combination random selection step of randomly selecting any one as a combination of index data corresponding to the waveform cutout data is provided.

According to a fourteenth aspect of the present invention, in the configuration of the twelfth or thirteenth aspect, when generating the waveform cut data, dummy data is added to a part of the waveform cut data. It is characterized by having a process.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, a preferred embodiment of the present invention will be described with reference to the drawings. [1] First Embodiment [1.1] Configuration of Vector Quantization Device FIG. 1 shows a schematic configuration block diagram of a vector quantization device. The vector quantization device 10 performs waveform extraction of the input signal data DIN and outputs it as extracted waveform data DP, generates an index based on the extracted waveform data DP, and generates bit stream data DBS. And an index generating unit 12 for outputting.

FIG. 2 shows a detailed configuration diagram of the index generation unit 12. The index generation unit 12 includes an embedding position information storage unit 21 that stores the embedding position data DFP of the watermark information.
And a random number generator 22 for generating and outputting random number data DRN
A watermark information storage unit 23 for storing watermark information data DWM in advance, a code book 24 for storing index data, and a cutout waveform based on the input cutout waveform data DP, random number data DRD and index data DCB. A first encoder 25 that selects the index data of the n-th candidate (n is a natural number) corresponding to the random number data from the plurality of candidate index data corresponding to the data DP and outputs the selected index data as the first index data DIDX1; The second index data D is selected by selecting index data corresponding to the cut-out waveform data and the watermark information data DWM.
Based on the second encoder 26 output as IDX2 and the embedding position data DFP, the cut-out waveform data DP not corresponding to the embedding position is output to the first encoder 25, and the cut-out waveform corresponding to the embedding position is output. A selector 27 that outputs data DP to a second encoder 26; a header information generator 28 that generates and outputs header information data DHD based on embedding position data DFP; a header information data DHD; And a bit stream generation unit 29 that rearranges the first index data DIDX1 output from the second encoder 26 and the second index data DIDX2 output from the second encoder 26 into serial and outputs the same as bit stream data DBS. It is configured.

[1.2] Operation of Vector Quantizing Apparatus The waveform extracting apparatus 11 extracts the waveform of the input signal data DIN and outputs the index generating section 12 as extracted waveform data DP.
Is output to the selector 27. On the other hand, the embedding position information storage unit 21 of the index generation unit 12 outputs the stored embedding position data DFP of the watermark information to the selector 27 and the header information generation unit 28. This allows the selector 27
Outputs the cut-out waveform data DP not corresponding to the embedding position to the first encoder 25 and outputs the cut-out waveform data DP corresponding to the embedding position to the second encoder 26 based on the embedding position data DFP.
Output to Further, the random number generation unit 22 outputs the random number data DRN
Is generated and output to the first encoder 25, and the watermark information storage unit 23 sequentially outputs the stored watermark information data DWM to the second encoder 26. Thus, the first encoder 25 receives the cut-out waveform data DP, the random number data DRD, and the index data DCB read from the codebook 24.
, The n-th candidate index data (n is a natural number) corresponding to the random number data is selected from the plurality of candidate index data corresponding to the cut-out waveform data DP, and
The data is output to the bit stream generation unit 29 as index data DIDX1.

More specifically, among the index data corresponding to the cut-out waveform data DP, the index data having the least error is set as the first candidate index data,
If the index data in which the error increases sequentially (including the same error) is defined as second candidate index data → third candidate index data →..., For example, a predetermined error range predetermined based on random number data DRD , Any one of the first to third candidate index data items belonging to the selected candidate index data item is selected and output to the bit stream generation unit 29 as the first index data item DIDX1. Therefore, the candidate index data having the smallest error is not necessarily selected, and it is difficult to find watermark information actually embedded as decoy index data. Further, the second encoder 26 reads out the index data corresponding to the inputted cut-out waveform data DP and the watermark information data DWM from the codebook 24, selects and outputs the selected index data to the bit stream generation unit 29 as the second index data DIDX2. I do. More specifically,
For example, if the watermark information data DWM has the fourth bit of the index data set to “1”, the fourth bit
Of the index data whose bit is “1”, the index data with the least error is output to the bit stream generation unit 29 as the second index data DIDX2.

Here, the operation of the header information generator 28 will be described with reference to FIG. FIG. 3 is an explanatory diagram of the configuration of the bit stream data DBS. The bit stream data DB is roughly composed of header information data DHD and vector index information data DVI. The header information data DHD corresponds to, for example, ID data DID having a non-overlapping value that can identify the vector quantization device 10 in the world, coding system data DEM indicating a coding system, and the bit stream data DBS. And title data DTI representing the title to be reproduced, and m pieces of watermark position information data DWMP1 to DWMPm representing the positions at which the watermark information is included. In this case, the ID data DID is “GUID (Glov
al Unique ID) ”. The header information data DHD other than the ID data DID may be encrypted by a predetermined encryption method. The data included in the header information data DHD is not limited to these, and various data can be included as needed.

[1.3] Vector Decoding Apparatus Next, the vector decoding apparatus will be described. [1.3.1] Configuration of Vector Decoding Apparatus FIG. 4 shows a schematic configuration block diagram of the vector decoding apparatus. The vector decoding device 15 converts the input bit stream data DBS (see FIG. 3) into the encoding system data DEM.
Is decoded by a decoding method corresponding to the coding method corresponding to or a predetermined decoding method, and output signal data D
The decoder 16 outputting as OUT and the watermark position information data DWMP (= any one of the watermark position information data DWPM1 to DWPMn in the bit stream data DBS, or the watermark position information input from an external watermark position information database not shown) And a watermark information decoder 17 for outputting output watermark information data DWMOUT based on the data. Note that the watermark information decoder 17 does not necessarily need to be included in a device that does not always need watermark information, such as a decoding device used by a general user. Here, prior to the description of the vector decoding apparatus, bit stream data DBS ′ when watermark position information data is input from an external watermark position information database
Will be described. Bit stream data DB
S 'is roughly composed of header information data DHD and vector index information data DVI. The header information data DHD includes, for example, ID data DID having a non-overlapping value that can identify the vector quantization device 10 and watermark position storage information data DWMST indicating a storage position of the watermark information on the watermark position information database. And is provided.

Also in this case, "GUID (Global Unique ID)" or the like is used as the ID data DID.
Further, the watermark position storage information data DWMST may be encrypted by a predetermined encryption method. The data included in the header information data DHD is not limited to these,
Various data can be included as needed. When the watermark position storage information data DWMST is passed to the watermark position information database, the watermark position information database outputs the watermark position information data corresponding to the watermark position storage information data DWMST to the watermark information decoder 17 as the watermark position information data DWMP. Will be done. Therefore, since the bit stream data DBS does not directly include information on the position where the watermark information is embedded, data security can be further secured.

[1.3.2] Operation of Vector Decoding Apparatus The decoder 16 of the vector decoding apparatus 15 decodes the input bit stream data DBS into an encoding scheme corresponding to the encoding scheme data DEM. The output signal data DOUT is decoded by a decoding method or a predetermined decoding method.
Output as Thereby, the user can reproduce an image, a sound, and the like corresponding to the output signal data DOUT. On the other hand, the watermark information decoder 17 outputs the watermark position information data DWMP (= one of the watermark position information data DWPM1 to DWPMn in the bit stream data DBS, or the watermark position information input from an external watermark position information database (not shown). Output watermark information data DWMOUT based on the data (data). Thus, if the copyright information or the like is included in the output watermark information data DWMOUT, the source of the output signal data DOUT can be easily specified.

[1.4] Effects of the First Embodiment As described above, according to the first embodiment, in the vector quantization, the index data corresponding to the non-embedded position of the watermark information is selected. In addition to the first candidate index data having the least error, other index data belonging to a predetermined error range are selected at random by random numbers, so that bit stream data is generated without including watermark information. If bit data is generated by including bit information and watermark information, and difference data is obtained, difference data including information other than the watermark information can be obtained, making it difficult to extract the watermark information. Bits that contain more robust watermark information against malicious attacks such as falsification or deletion of watermark information Stream data can be obtained.

[1.5] Modification of First Embodiment [1.5.1] First Modification In the above description, only the case where watermark information is embedded has been described. Similarly, not only the first candidate index data having the least error at the time of selecting the index information on the first encoder side, but also other index data belonging to a predetermined error range are randomly selected by random numbers. With this configuration, the malicious attacker can compare the difference between the bitstream data generated without watermark information and the bitstream data generated with watermark information to determine the difference data. Even if it is obtained, the difference data including information other than the watermark information is obtained in the same manner, which makes it difficult to extract the watermark information. In this case, bit stream data that is resistant to falsification or erasure of watermark information can be obtained. [1.5.2] Second Modification In the above description, the watermark information is embedded in a part of the index data. However, the watermark information may be embedded in all index data. It is possible.

[2] Second Embodiment In the above-described first embodiment, no processing is performed on the input signal data DIN input to the waveform extracting device 11, but this second embodiment is different from the first embodiment. , Input signal data D
IN is one or a plurality of finite-length sample sequences (for example, in the case of audio data, 16 bits per sample sequence), at the beginning of each finite-length sample sequence when watermark information is embedded and when watermark information is not embedded. This is an embodiment in a case where waveform extraction is performed after adding dummy data having a different number of sample columns. In the following description, a case where audio data is used as input signal data will be described for easy understanding, but the present invention can be applied to other data such as image data. [2.1] Explanation of Principle of Second Embodiment Here, prior to specific description of the second embodiment, the principle of the second embodiment will be described. FIG. 5 is a diagram illustrating the principle of the second embodiment. Assume that the input signal waveform corresponding to the input signal data DIN is as shown by a solid line in FIG.
Input signal data DIN input from time t0 to time t1
Is a finite number of sample sequences, index data obtained when vector quantization is performed from the beginning of the finite length sample sequence, such that the index data ~ with the corresponding input signal waveform Become.
On the other hand, when dummy data of a predetermined number of samples (in FIG. 5A, a signal waveform corresponding to the dummy data is indicated by an alternate long and short dash line) is added to the head of the finite length sample sequence. From time t2 to time t3 (t3−t2 =
The index data obtained when vector quantization is performed from the beginning of a finite number of sample strings corresponding to the input signal data DIN input up to t1-t0) is the index data '~ with the corresponding input signal waveform. ', Which is completely different from the index data ~. Therefore, in the second embodiment, using this, the number of samples of dummy data to be added to the head of the finite-length bit string when watermark information is embedded and the dummy data to be added to the head of the finite-length sample string when watermark information is not embedded By making the number of samples different from each other, a malicious attacker can compare bitstream data generated without watermark information with bitstream data generated with watermark information. Even when the difference data is obtained, the difference data including information other than the watermark information is obtained, which makes it difficult to extract the watermark information.

[2.2] Operation of Vector Quantizer The operation of the vector quantizer according to the second embodiment will now be described. The operation of the vector quantization apparatus according to the second embodiment is the same as that of the first embodiment with respect to the operation after waveform extraction, and thus detailed description thereof will be omitted, and processing up to waveform extraction will be omitted. explain. FIG. 6 shows a processing flowchart up to waveform extraction according to the second embodiment. When the input signal data DIN is input (step S1), it is determined whether or not the input data is the head of a finite number of sample sequences (step S2). Step S2; No), the process proceeds to step S
Move to 6. In this case, the input signal data DIN
All may be set to be treated as one finite sample sequence, or the input signal data DIN may be divided into a plurality of sets and each obtained divided data set to be treated as a finite sample sequence. good. For example, if the number L of sample data of a finite number of sample rows is set as L = the number of sample data of the cut-out waveform data DP−the number of sample data of the dummy data, the sample data of the cut-out waveform data after adding the dummy data is obtained. The number can be the same as the number of sample data of the cut-out waveform data when no dummy data is added. If it is determined in step S2 that the input data is the head of a finite number of sample strings (step S2; Yes), it is determined whether or not encoding for embedding watermark information is performed (step S3). Step S3
In the discrimination, when watermark information is embedded (watermark) encoding (step S3; Yes), X (X
Is a natural number, for example, "00h" (h represents a hexadecimal number; the same applies hereinafter) is added as dummy data to the head of the data. On the other hand, in the determination in step S3, when normal encoding without embedding watermark information is performed (step S3; No), Y constant values (Y is a natural number and Y ≠ X), for example, “00h” "(= 8 bits) is added to the head of the data as dummy data.

Next, the waveform of the data to which the dummy data is added or not is extracted (step S6).
The processing shifts to the encoding processing. Here, a more specific description will be given. In the following description, it is assumed that the data length of one sample data is 8 bits, but the present invention is not limited to this. For example, any data length such as 16 bits, 32 bits, and 64 bits is possible. (1) When dummy data is embedded at the beginning of input data When watermark information is embedded (watermark) encoding, a finite number of input sample strings are set to “08h” as shown in FIG. , “F7h”, “04h”, “06”
h "," FAh "," 01h "," 02h "," FCh "
“04h”, “06h”, “FAh”,.
= 2, the number of samples of the cut-out waveform data is 6 (corresponding to a data length of 48 bits), and the dummy data is arbitrary sample data such as "## h" (actually, "00h"). , It is described as “## h” for the sake of explanation. The same applies hereinafter.), The obtained first cut-out waveform data is “## h” as shown in FIG. ”,“ ## h ”,“ 08h ”,“ F7h ”,“ 0
4h ”and“ 06h. ”Accordingly, the second cut-out waveform data is shown in FIG.
As shown in (c), “FAh”, “01h”, “02h”, “FCh”, “0
4h "and" 06h ".

On the other hand, when no dummy data is added (conventional example), the obtained first cut-out waveform data is "08h""F7h" as shown in FIG. , “04h”, “06h”, “F
Ah "and" 01h ", and the second cut-out waveform data includes" 02h "," FCh "," 04h "," 06h ", and" FA "as shown in FIG.
h ″,..., which is completely different from the cut-out waveform data to which the dummy data is added and the cut-out waveform data obtained when the dummy data is not added (conventional example). When encoding is performed, as shown in FIG. 11A, the input finite number of sample sequences
“08h”, “F7h”, “04h”, “06h”, “FA
h "," 01h "," 02h "," FCh "," 04h ",
“06h”, “FAh”,..., Y = 3,
When the number of samples of the cut-out waveform data is 6 and the dummy data is “## h”, the first cut-out waveform data obtained is similarly “## h”, “ ## h, "## h", "08h", "F"
7h ”and“ 04h. ”Accordingly, the second cut-out waveform data includes“ 06h ”,“ FAh ”,“ 01h ”,“ 02h ”, and“ FC ”.
h "and" 04h ".

Therefore, the obtained cut-out waveform data is cut-out waveform data obtained when no dummy data is added, "08h", "F7h", "04h", "06h", "F".
Ah, “01h” and “02h”, “FCh”, “04h”, “06h”, “FA”
h ",..., and cut-out waveform data obtained when performing (watermark) encoding for embedding watermark information,“ ## h ”,“ ## h ”,“ 08h ”,“ F7h ”,
“04h”, “06h” and “FAh”, “01h”, “02h”, “FCh”,
Both "04h" and "06h" are completely different. (2) When dummy data is embedded at the beginning of each cut-out waveform data When watermark information is embedded (watermark) encoding, a finite number of input sample strings are set as shown in FIG. “08h”, “F7h”, “04h”, “06”
h "," FAh "," 01h "," 02h "," FCh "
“04h”, “06h”, “FAh”,.
= 2, the number of samples of the cut-out waveform data is 6, and the dummy data is “## h”, the first cut-out waveform data obtained is as shown in FIG. As shown, “## h”, “## h”, “08h”, “F7h”, “0”
4h ”and“ 06h. ”Accordingly, the second cut-out waveform data is shown in FIG.
As shown in (e), “## h”, “## h”, “FAh”, “01h”,
“02h” “FCh”. Therefore, even by embedding dummy data at the beginning of each cut-out waveform data, data completely different from cut-out waveform data obtained when no dummy data is added can be obtained.

[2.3] Effects of the Second Embodiment Accordingly, the bit stream data generated without watermark information and the bit stream data generated with watermark information included are compared. Even when the difference data is obtained, the difference data including the information other than the watermark information can be obtained, which makes it difficult to extract the watermark information, and prevents malicious attacks such as falsification or deletion of the watermark information. Bit stream data containing stronger watermark information can be obtained. [2.4] Modification of Second Embodiment In the above description, the number of sample data having a predetermined constant value (the number of sample data) is determined as the dummy data. It is also possible to determine the number of sample data.

[3] Third Embodiment In the above-described second embodiment, the dummy data is determined as a predetermined number of constant values. However, in the third embodiment, real data of a finite number of sample strings are used. This is an embodiment in which obtained interpolation data is added as dummy data. [3.1] Operation of Third Embodiment FIG. 7 shows a processing flowchart up to waveform extraction in the third embodiment. When the input signal data DIN is input (step S11), it is determined whether or not the input data is the head of a finite number of sample sequences (step S12). Step S1
2; No), the process proceeds to step S16. In this case, as in the second embodiment, the finite number of sample strings may be set so that all the input signal data DIN are treated as one finite number of sample strings, or a plurality of input signal data DIN may be used. May be set so that each of them is treated as a finite number of sample strings. If it is determined in step S12 that the input data is the head of a finite number of sample strings (step S1).
2; Yes), it is determined whether or not encoding for embedding watermark information is performed (step S13). In the discrimination in step S3, when encoding for embedding watermark information is performed (step S13; Yes), the data of N samples (N is a natural number of 2 or more) from the beginning of a finite number of sample strings is used, X data (X is a natural number) is estimated by extrapolation, and the obtained X sample data is added as dummy data to the head of the data.

On the other hand, in the discrimination in step S13, when normal encoding without embedding watermark information is performed (step S13; No), N samples (N is a natural number of 2 or more) from the head of a finite number of sample strings Using the data of the immediately preceding Y (Y is a natural number of 2 or more and Y ≠ X)
Are extrapolated by extrapolation, and the obtained Y sample data are added as dummy data to the head of the data. Next, the waveform of the data to which the dummy data is added or not is extracted (step S16), and the process proceeds to the encoding process. [3.2] Effects of Third Embodiment Accordingly, in the third embodiment as well, there are a case where bit stream data is generated without including watermark information and a case where bit stream data is generated including watermark information. Even if the difference data is obtained, the difference data including information other than the watermark information can be obtained, which makes it difficult to extract the watermark information, and the malicious information such as falsification or deletion of the watermark information can be obtained. Bit stream data containing stronger watermark information can be obtained for a certain attack.

[3.3] Modification of Third Embodiment [3.3.1] First Modification In the above description, a predetermined fixed number of sample data is determined as extra data by extrapolation. However, it is also possible to determine the number of samples of the dummy data using a pseudo random number. This will be described more specifically with reference to the processing flowchart of FIG. First, a random number is generated (step S22), and the number of sample data of dummy data (the number of dummy data DC) is set based on the obtained random number (step S21). Next, a data counter for counting the generated dummy data Initialize C.
That is, C = 0. Next, it is determined whether or not the count value of the data counter C is equal to the number of generated dummy data (step S).
24). If it is determined in step S24 that the count value of the data counter C is not equal to the number of generated dummy data (step S24; No), the dummy data still needs to be generated. The calculation is performed (step S25). Then, the obtained interpolation data is set in an extraction buffer for extracting a waveform (step S26).

Subsequently, the count value of the counter C is incremented by 1 (step S29), and the process proceeds to step S24 again. Thereafter, the same process is repeated.
The process is repeated until the number of interpolation data equal to the number of dummy data DC is obtained. If it is determined in step S24 that the count value of the data counter C is equal to the number of generated dummy data (step S24; Yes), since the storage of the dummy data in the cutout buffer has been completed, the input signal data DIN Is set in the extraction buffer (step S27). Subsequently, it is determined whether or not the count value of the counter C has become equal to a finite word length equal to the predetermined number of sample data of the cut-out waveform data (step S28). In the determination in step S28,
If the count value of the counter C is not yet equal to the predetermined finite word length corresponding to the number of sample data of the cut-out waveform data (step S28; No), the count value of the counter C is incremented by 1 (step S28). S2
9) The process proceeds to step S27 again, and thereafter, the same process is repeated until the number of samples of the data stored in the extraction buffer becomes equal to the finite word length corresponding to the number of samples of the extraction waveform data. The input signal data is stored in the extraction buffer. If it is determined in step S28 that the count value of the counter C is equal to a predetermined number of samples of the cut-out waveform data (step S28; Ye).
s) The cut-out waveform data stored in the cut-out buffer is output (step S30), and the same encoding process is performed thereafter. With the above-described configuration, according to the first modification, extraction of watermark information becomes more difficult, and bit stream data including watermark information that is more resistant to malicious attacks such as falsification or deletion of watermark information. Can be obtained.

[3.3.2] Second Modification In the above description, the dummy data is calculated by the extrapolation method (extrapolation method). However, the last data of the previous finite sample sequence and the current data are calculated. It is also possible to calculate dummy data by interpolation (interpolation) based on the leading data of the finite sample sequence. As the interpolation method, a general interpolation method such as linear interpolation, Lagrange interpolation, and spline interpolation can be used.

[4] Fourth Embodiment In each of the embodiments described above, a basic vector quantization apparatus has been described as an example. However, in the fourth embodiment, a multi-stage vector quantization This is an embodiment in the case of using a chemical conversion device. [4.1] Configuration of Multistage Vector Quantization Device FIG. 9 is a block diagram illustrating a configuration of the multistage vector quantization device. The multi-stage vector quantizer 40 calculates the input signal data D
A waveform extracting device 41 that extracts an IN waveform and outputs it as extracted waveform data DP, and generates an index based on the extracted waveform data DP to generate bit stream data DB1.
And an index generation unit 42 that outputs The index generation unit 42 includes a first index selection unit 45 that outputs first index data DIDX11 based on the cut-out waveform data DP, and a cut-out waveform data D
A first adder 46 that adds P and the first index data DIDX11 to output the first index data DIDX11 ′, and a second that outputs the second index data DIDX12 based on the first index data DIDX11 ′.
An index selecting section 47 and first index data D
The second adder 48 adds the IDX11 and the second index data DIDX12 to output the added second index data DIDX12 ', and the third index outputs the third index data DIDX13 based on the added second index data DIDX12'. The configuration includes a selection unit 49 and a multiplexer unit 50 that generates and outputs bit stream data DB1 from the first index data DIDX11, the second index data DIDX12, and the third index data DIDX13 by a time division method or the like.

In this case, the first index selecting section 45, the second index selecting section 47, and the third index selecting section 49 include the embedding position information storing section 21, the random number generating section 22, the watermark information storing section in the first embodiment. Part 2
3, codebook 24, first encoder 25, second encoder 2
6 and the selector 27. [4.2] Operation of Vector Quantizing Device The waveform extracting device 41 of the multi-stage vector quantizing device 40 extracts the waveform of the input signal data DIN and extracts the extracted waveform data DP.
To the first index selecting section 45 and the first adding section 46 of the index generating section 42. Thereby, the first index selecting unit 45 of the index generating unit 42
Performs an index generation based on the cut-out waveform data DP, and adds the first index data DIDX11 to the first adder 4.
6, output to the second adder 48 and the multiplexer 50. In this case, the first index selecting unit 45
When the first index data DIDX11 corresponding to the watermark information non-embedded position is selected, not only the first candidate index data having the least error but also other index data belonging to a predetermined error range are determined by random numbers. , The same effect as in the first embodiment can be obtained. The first adder 46 adds the cut-out waveform data DP and the first index data DIDX11 and outputs the added first index data DIDX11 'to the second index selector 47.

The second index selecting section 47 determines whether the first
The second index data DIDX12 is generated based on the index data DIDX11 ', and is output to the second adder 48 and the multiplexer 50. Also in this case, similarly to the first index selecting unit 45, the second index selecting unit 47 selects the first candidate index data with the least error when selecting the second index data DIDX12 corresponding to the watermark information non-embedded position. Not only
Other index data belonging to a predetermined error range are selected at random by random numbers. Second adder 48
Adds the first index data DIDX11 and the second index data DIDX12 and outputs the added second index data DIDX12 ′ to the third index selecting unit 49 and the multiplexer unit 50. The third index selecting section 49 outputs the third index data DIDX13 to the multiplexer section 50 based on the added second index data DIDX12 '. Also in this case, the third index selecting unit 49, like the first index selecting unit 45 or the second index selecting unit 47, outputs the third index data DIDX1 corresponding to the watermark information non-embedded position.
At the time of selecting 3, not only the first candidate index data having the least error but also other index data belonging to a predetermined error range are selected at random by random numbers.

[4.3] Effects of the Fourth Embodiment As a result, the multiplexer unit 50 converts the first index data DIDX11, the second index data DIDX12, and the third index data DIDX13 into bit stream data DB1 by a time division method or the like. Is generated and output, but the first index selection unit 45, the second index selection unit 47, or the third index selection unit 49
In selecting the index data corresponding to the non-embedded position of the watermark information, not only the first candidate index data having the least error but also other index data belonging to a predetermined error range is randomly selected by using a random number. Therefore, the bit stream data generated without including the watermark information is compared with the bit stream data generated including the watermark information, and even if the difference data is obtained, information other than the watermark information is obtained. Can be obtained in a state in which the watermark information is complicatedly inserted, making it difficult to extract the watermark information, and making the bit containing the watermark information stronger against malicious attacks such as falsification or erasure of the watermark information. Stream data can be obtained. [4.4] Modification of Fourth Embodiment In the above description, as in the case of the first embodiment,
Although the case where index data is selected at random by random numbers has been described, it is possible to apply the same aspect as the second and third embodiments also in multistage vector quantization.

[5] Fifth Embodiment In the first to third embodiments described above, a basic vector quantization device has been described as an example. However, in the fifth embodiment, as a vector quantization device, This is an embodiment in which a conjugate vector quantization device is used. [5.1] Configuration of Conjugate Vector Quantizer FIG. 10 is a block diagram showing the configuration of the conjugate vector quantizer. The conjugate vector quantizer 60 calculates the input signal data D
A waveform extracting device 61 that extracts an IN waveform and outputs the extracted waveform data DP, and generates an index based on the extracted waveform data DP to generate bit stream data BS2.
And an index generation unit 62 that outputs The index generation unit 62 stores a first codebook 65 storing first index data DCB1 having a first system, and a second codebook 65 storing second index data DCB2 having a second system different from the first system. Of the combinations of the codebook 66 and the first index data DCB1 and the second index data DCB2 corresponding to the same cut-out waveform data DP, among a plurality of combinations in which the amount of error is included in a predetermined error range. And selecting one of the combinations by random numbers, and outputting the first index data DCB1 and the second index data DCB21 corresponding to the selected combination as the first selected index data DIDX21 and the second selected index data DIDX22. Part 66
And a multiplexer 68 that generates and outputs bit stream data DB2 by a time division method or the like based on the first selected index data DIDX21 and the second selected index data DIDX22.

[5.2] Operation of Conjugate Vector Quantizing Device The waveform extracting device 61 of the conjugate vector quantizing device 60 extracts the waveform of the input signal data DIN and outputs the extracted waveform data DP.
Is output to the index generation unit 62. The combination selecting section 66 of the index generating section 62 outputs the first index data DC corresponding to the input cut-out waveform data DP.
Among the combinations of B1 and the second index data DCB2, one of a plurality of combinations in which the amount of error is included in a predetermined error range is selected by a random number, and corresponds to the selected combination. The first index data DCB1 and the second index data DCB21 to be converted into the first selected index data DIDX21.
And outputs it to the multiplexer unit 68 as the second selection index data DIDX22. As a result, the multiplexer 68 generates and outputs the bit stream data DB2 by a time division method or the like based on the first selected index data DIDX21 and the second selected index data DIDX22.

[5.3] Effect of Fifth Embodiment As described above, among the combinations of the first index data DCB1 and the second index data DCB2 corresponding to the cut-out waveform data DP, the error amount is determined in advance. Since any one of a plurality of combinations included in the predetermined error range is selected by using a random number, the bit stream data generated without including the watermark information and the watermark information are selected. Is included, bit stream data is generated and compared, and even if the difference data is obtained, difference data including information other than the watermark information is obtained, making it difficult to extract the actual watermark information. Bit stream data that contains stronger watermark information against malicious attacks such as falsification or erasure of watermark information. Rukoto can.

[5.4] Effect of Fifth Embodiment In the above description, when selecting a combination of the first index data DCB1 and the second index data DCB2 corresponding to the cut-out waveform data DP, the combination is determined by a random number. However, as in the second and third embodiments, when the cut-out waveform data DP is obtained, the extraction of the watermark information becomes more difficult by including the dummy data. can do.

[0045]

According to the present invention, a difference between bit stream data generated without watermark information and bit stream data generated with watermark information included is obtained. Even in this case, difference data including information other than the watermark information in a complicated state can be obtained, making it difficult to extract the watermark information, and preventing malicious attacks such as falsification or deletion of the watermark information. Thus, bit stream data including stronger watermark information can be obtained, and illegal use of the information after vector quantization can be accurately grasped.

[Brief description of the drawings]

FIG. 1 is a schematic configuration block diagram of a vector quantization device according to a first embodiment.

FIG. 2 is a detailed configuration block diagram of an index generation unit 12 according to the first embodiment.

FIG. 3 is a diagram illustrating a configuration of bit stream data.

FIG. 4 is a schematic configuration block diagram of a vector decoding device.

FIG. 5 is a diagram illustrating dummy data.

FIG. 6 is a processing flowchart of a second embodiment.

FIG. 7 is a processing flowchart of a third embodiment.

FIG. 8 is a processing flowchart of a first modified example of the third embodiment.

FIG. 9 is a schematic block diagram of a multi-stage vector quantization apparatus according to a fourth embodiment.

FIG. 10 is a schematic block diagram of a conjugate vector quantization device according to a fifth embodiment.

FIG. 11 is a specific explanatory diagram of cut-out waveform data.

[Explanation of symbols]

Reference numeral 10: vector quantization device, 11: waveform extraction device, 12
... Index generation unit, 15 ... Vector decoding device, 1
6 decoder, 17 watermark information decoder, 21 embedding position information storage unit, 22 random number generation unit, 23 watermark information storage unit, 24 codebook, 25 first encoder, 26 second Encoder, 27... Selector, 28-point header information generation unit,
29: bit stream generator, 40: multi-stage vector quantizer, 41: waveform extracting device, 42: index generator, 45: first index selector, 46: first adder, 47: second index selector 48, a second adder, 49, a third index selector, 50, a multiplexer, 60, a conjugate vector quantizer, 61, a waveform extractor, 62, an index generator, 65, a first codebook, 66 ... Combination selection section, 67... 2nd codebook, 68... Multiplexer section, DIN... Input signal data, DP... Waveform cutout data, DBS .bit stream data, DFP... Watermark embedding position information data, DWM. , DHD ... header information data, DIDX1 ... first index data, DIDX2 ... second index data,
DRD: random number data.

Claims (14)

[Claims] 1. A vector quantization apparatus for vector-quantizing input signal data, wherein, in the waveform cutout data obtained by performing the waveform cutout of the input signal data, the waveform cutout of the non-embedding target of the watermark information is performed. When selecting index data for output data, any one of a plurality of index data in which an error with the original waveform cut data when decoding is within a predetermined error range is used as the waveform cut data. A vector quantization device comprising random selection means for randomly selecting corresponding index data. 2. A vector quantizer for quantifying conjugate vectors of input signal data, wherein a waveform to be subjected to non-embedding of watermark information is included in waveform cut data obtained by performing a waveform cut of the input signal data. When selecting a combination of index data for cut data, any one of a plurality of index data combinations from a plurality of index data in which an error from the original waveform cut data is within a predetermined error range when decoded. A combination random selection means for randomly selecting a combination of the above as a combination of index data corresponding to the waveform cutout data. 3. The vector quantization device according to claim 1, further comprising: a dummy data adding unit configured to add dummy data to a part of the waveform cut data when generating the waveform cut data. A vector quantization device, characterized in that: 4. The vector quantization apparatus according to claim 1, further comprising: a first vector quantization unit to an n-th vector quantization unit (n: an integer of 2 or more) for performing multi-stage vector quantization. Characteristic vector quantization device. 5. The vector quantization device according to claim 4, wherein at least one of the first vector quantization unit to the n-th vector quantization unit is:
A vector quantization apparatus, comprising: dummy data adding means for adding dummy data to a part of the waveform cut data when generating the waveform cut data. 6. The vector quantization device according to claim 3, wherein the waveform cutout data is constituted by a predetermined finite number of sample data, and wherein the dummy data adding unit is configured to output the input signal data. Is divided into divided data composed of a predetermined number of sample data smaller than the finite number, and the waveform cutout composed of the finite number of sample data by adding the dummy data for each divided data. A vector quantization device for generating data. 7. The vector quantization apparatus according to claim 6, wherein the number of the sample data in the dummy data is determined by the divided data to be embedded with watermark information and the divided data to be not embedded with watermark information. A vector quantization apparatus characterized in that the vector quantization is different. 8. The vector quantization device according to claim 6, wherein the dummy data is a predetermined constant value of sample data which is divided into the divided data until the number of sample data constituting the divided data reaches the finite number. A vector quantization device characterized by repeatedly adding. 9. The vector quantization apparatus according to claim 6, wherein the dummy data is obtained by performing a predetermined operation based on the divided data, and the number of sample data constituting the waveform cutout data. A vector quantization apparatus, wherein operation data composed of a number of sample data corresponding to a difference from the number of sample data constituting the divided data is used. 10. The vector quantization device according to claim 9, wherein the operation data is calculated by an extrapolation operation. 11. The vector quantization device according to claim 9, wherein the operation data is calculated by an interpolation operation. 12. A vector quantization method for vector-quantizing input signal data, wherein, of the waveform cut data obtained by performing the waveform cut of the input signal data, the waveform cut of the non-embedding target of the watermark information is performed. When selecting index data for output data, any one of a plurality of index data in which an error with the original waveform cut data when decoding is within a predetermined error range is used as the waveform cut data. A vector quantization method comprising a random selection step of randomly selecting corresponding index data. 13. A vector quantization method for performing conjugate vector quantization on input signal data, wherein, among the waveform cutout data obtained by performing the waveform cutout of the input signal data, the waveform of the non-embedded watermark information. Upon selecting a combination of index data for the cutout data, when decoding, the error with the original waveform cutout data is one of a plurality of index data combinations that fall within a predetermined error range. A vector quantization method comprising a combination random selection step of randomly selecting a combination of index data corresponding to waveform cutout data. 14. The vector quantization method according to claim 12, further comprising a dummy data adding step of adding dummy data to a part of the waveform cut data when generating the waveform cut data. A vector quantization method.