JP2000207831A - Digital signal processing device and method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable of electronic watermark information to be embedded into a digital signal almost without degrading the quality of sound and picture. SOLUTION: This digital signal processor is provided with a digital input I/F circuit 11 for inputting at least one digital signal out of the digital audio signal and digital image signal, an FFT(fast Fourier transform) circuit 15 for producing plural transformation coefficients to respectively show a level of the corresponded frequency component among the levels of plural frequency components obtained by the transformation of the inputted digital signal made from the time area to the frequency area, and an MPU(microprocessor unit) 19 for embedding the electronic watermark information by changing a value of at least one transformation coefficient among plural transformation coefficients.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a digital signal processing apparatus and a method for performing processing relating to digital watermark information.

[0002]

2. Description of the Related Art A digital audio (Digital Audio) signal and a digital image (Digital Image) signal are hardly deteriorated in quality compared to an original (MD).
Disk), CD (Cmpact Disk)-R (Recordable) and D
It can be recorded on a recording medium such as a VD (Digital Versatile Disk). Therefore, if the recording of the digital audio signal and the digital image signal on the recording medium is allowed without any restriction, the interest of the copyright holder is significantly impaired. From such a background, in recent years, information for restricting recording of the signal on the recording medium is embedded as digital watermark information (Watermark) in the original digital audio signal or the original digital image signal, and the recording apparatus records the information in the digital watermark information. A method of controlling recording on a recording medium based on the method has been proposed. Here, the digital watermark information is information that is embedded (embedded) in a signal in such a manner that it is hard to be perceived by human eyes and ears, and is difficult to remove or rewrite.

As a technique for embedding such digital watermark information in an original digital audio signal or an original digital image signal, for example, pseudorandom noise such as a PN (Pseudorandom Noise) sequence signal is superimposed on the original digital audio signal or the original digital image signal. There is something to do. According to this method, when recording a digital audio signal or a digital image signal on a recording medium, the recording apparatus detects pseudorandom noise superimposed on the digital audio signal or the digital image signal, and performs either copy prohibition or copy permission. It determines whether or not there is, and controls recording on the recording medium based on the determination result.

[0004]

However, when pseudo-random noise is superimposed as described above, a noise component is directly added to an original digital audio signal or an original digital image signal even though it is a small signal. In some cases, depending on the brightness, volume, distribution of the frequency spectrum, and the like of the digital image signal, deterioration of the sound and image quality due to the embedding of the digital watermark information may be perceived by human eyes and ears. In addition, the digital watermark information detection device cannot detect the pseudo-random noise superimposed on the digital audio signal or digital image signal only after obtaining a sample of a section that is continuous to some extent in time. In such a case, it is necessary to superimpose pseudorandom noise on the original digital audio signal or the original digital image signal for a relatively long time, and there is a problem that sound and sound quality degradation due to embedding digital watermark information continues for a long time.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems of the prior art, and has been made in consideration of the above-described problems. It is an object to provide a processing device and a method thereof. It is another object of the present invention to provide a digital signal processing device and a digital signal processing method capable of detecting embedded digital watermark information by changing a conversion coefficient indicating a level of a predetermined frequency component. Also, the present invention detects digital watermark information embedded by changing a conversion coefficient indicating a level of a predetermined frequency component, and can control output of a digital signal and recording on a recording medium based on the detection result. It is an object to provide a signal processing device.

[0006]

In order to solve the above-mentioned problems of the prior art and achieve the above-mentioned object, a digital signal processing apparatus according to a first aspect of the present invention analyzes a digital signal by frequency analysis. Orthogonal transform means for converting from the time domain to the frequency domain and generating a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal; and Information embedding means for embedding digital watermark information in the digital signal by changing a value of at least one of the conversion coefficients.

In the digital signal processing apparatus according to the first aspect of the present invention, first, the orthogonal signal is frequency-analyzed and converted from the time domain to the frequency domain by the orthogonal transform means.
A plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. Next, in the information embedding means, digital watermark information is embedded by changing the value of at least one of the plurality of transform coefficients. As described above, by embedding the digital watermark information by changing the value of one transform coefficient obtained by the frequency conversion, it is not necessary to superimpose the pseudo random noise on the digital signal as in the related art. It is possible to reduce the time for embedding the digital watermark information in the inside. That is, in the digital watermark information detecting device, if the value of the conversion coefficient in which the digital watermark information is embedded is known, the digital watermark information can be embedded in the digital signal in a detectable manner by temporarily changing the value of the conversion coefficient for a short time. Can be.

Further, the digital signal processing apparatus according to the first aspect of the present invention preferably includes a transform coefficient having the changed value,
The apparatus further includes orthogonal inverse transform means for orthogonally inversely transforming a digital signal represented in a frequency domain by a transform coefficient whose value is not changed among the plurality of transform coefficients into a time domain.

Further, a digital signal processing apparatus according to a second aspect of the present invention includes a trigger signal detecting means for detecting a trigger signal included in the input digital signal based on an amplitude level of the input digital signal; An orthogonal transform for converting an input digital signal from a time domain to a frequency domain by frequency analysis and generating a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency components included in the digital signal Means for embedding digital watermark information by changing a value of at least one of the plurality of transform coefficients after the trigger signal is detected.

In the digital signal processing device according to the second aspect of the present invention, first, the trigger signal detecting means detects a trigger signal included in the input digital signal based on the amplitude level of the input digital signal. . Then, in the orthogonal transformation means, the input digital signal is frequency-analyzed and converted from the time domain to the frequency domain, and a plurality of levels respectively indicating the level of the corresponding frequency component among the levels of the plurality of frequency components included in the digital signal Are generated. Next, after the trigger signal is detected by the information embedding means, digital watermark information is embedded by changing the value of at least one of the plurality of transform coefficients.

The digital signal processing device according to a second aspect of the present invention preferably includes a transform coefficient having the changed value,
The apparatus further includes orthogonal inverse transform means for orthogonally inversely transforming a digital signal represented in a frequency domain by a transform coefficient whose value is not changed among the plurality of transform coefficients into a time domain.

A digital signal processing apparatus according to a third aspect of the present invention analyzes the frequency of a digital signal and converts it from a time domain to a frequency domain, and converts a level of a plurality of frequency components contained in the digital signal. Orthogonal transform means for generating a plurality of transform coefficients each indicating a level of a frequency component to be converted, and information detecting means for detecting whether or not digital watermark information is embedded based on the generated change pattern of the plurality of transform coefficients And

In a digital signal processing apparatus according to a third aspect of the present invention, first, a digital signal is frequency-analyzed and converted from a time domain to a frequency domain by orthogonal transform means.
A plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. Next, the information detecting means detects whether or not digital watermark information is embedded based on the generated change pattern of the plurality of transform coefficients.

A digital signal processing apparatus according to a fourth aspect of the present invention is characterized in that a digital signal is frequency-analyzed and converted from a time domain to a frequency domain in units of a conversion processing block composed of a predetermined amount of digital signals. Orthogonal transform means for generating a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the signal; and a plurality of the transform processing blocks successively processed by the orthogonal transform means. Information detecting means for detecting a difference between predetermined transform coefficients among the generated transform coefficients, and detecting whether or not the digital watermark information is embedded based on a result of the detection. Have.

In the digital signal processing apparatus according to a fourth aspect of the present invention, first, the orthogonal signal is frequency-analyzed by the orthogonal transform means in units of a conversion processing block composed of a predetermined amount of digital signals, and is converted from the time domain to the frequency domain. A plurality of transform coefficients are generated that indicate the level of the corresponding frequency component among the plurality of frequency component levels included in the digital signal. Next, in the information detecting means, a difference of a predetermined transform coefficient among the generated transform coefficients is detected among the plurality of transform processing blocks successively processed by the orthogonal transform means, and the detected , It is detected whether or not the digital watermark information is embedded.

A digital signal processing apparatus according to a fifth aspect of the present invention analyzes a frequency of a digital signal by converting the frequency from a time domain to a frequency domain, and converts a level of a plurality of frequency components included in the digital signal. Orthogonal transform means for generating a plurality of transform coefficients each indicating a level of a frequency component to be converted, and a plurality of transforms each indicating a level of a plurality of frequency components continuously located on a frequency among the generated plurality of transform coefficients. Information detecting means for detecting a difference between coefficient values and detecting whether or not the digital watermark information is embedded based on the difference.

In a digital signal processing apparatus according to a fifth aspect of the present invention, first, a digital signal is frequency-analyzed and converted from a time domain to a frequency domain by orthogonal transform means.
A plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. Next, in the information detecting means, a difference between values of a plurality of transform coefficients, each of which indicates a level of a plurality of frequency components continuously located on a frequency among the plurality of generated transform coefficients, is detected. Whether the digital watermark information is embedded is detected based on the information.

A digital signal processing apparatus according to a sixth aspect of the present invention analyzes a frequency of a digital signal by converting the frequency from a time domain to a frequency domain, and converts a level of a plurality of frequency components included in the digital signal into a corresponding signal. Orthogonal transform means for generating a plurality of transform coefficients each indicating a level of a frequency component to be converted, and information detecting means for detecting whether or not digital watermark information is embedded based on the generated change pattern of the plurality of transform coefficients And output means for performing output in accordance with the digital signal, and control means for controlling whether the output means can output the digital signal based on a result of the detection.

In a digital signal processing apparatus according to a sixth aspect of the present invention, first, a digital signal is frequency-analyzed and converted from a time domain to a frequency domain by orthogonal transform means.
A plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. Next, the information detecting means detects whether or not digital watermark information is embedded based on the generated change pattern of the plurality of transform coefficients. Next, the control means controls whether the output means can output the digital signal based on the detection result of the digital watermark information by the information detection means.

A digital signal processing apparatus according to a seventh aspect of the present invention analyzes the frequency of a digital signal by converting the frequency from a time domain to a frequency domain, and outputs a corresponding frequency among a plurality of frequency components included in the digital signal. Orthogonal transform means for generating a plurality of transform coefficients each indicating a component level, and information detecting means for detecting whether or not digital watermark information is embedded based on a change pattern of the generated plurality of transform coefficients, Recording means for recording the digital signal on a recording medium; and control means for controlling whether or not the digital signal can be recorded on the recording medium by the recording means based on a result of the detection.

In a digital signal processing apparatus according to a seventh aspect of the present invention, first, a digital signal is frequency-analyzed and converted from a time domain to a frequency domain by orthogonal transform means.
A plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. Next, the information detecting means detects whether or not digital watermark information is embedded based on the generated change pattern of the plurality of transform coefficients. Next, the control unit controls whether or not the recording unit can record the digital signal on the recording medium based on the detection result of the digital watermark information by the information detection unit.

In a digital signal processing method according to a first aspect of the present invention, a digital signal is frequency-analyzed, converted from a time domain to a frequency domain, and a corresponding frequency among a plurality of frequency components included in the digital signal is output. A plurality of transform coefficients each indicating a component level are generated, and digital watermark information is embedded by changing the value of at least one of the plurality of transform coefficients.

Further, in the digital signal processing method according to a second aspect of the present invention, a trigger signal included in the input digital signal is detected based on an amplitude level of the input digital signal, and the input digital signal is Frequency analysis is converted from the time domain to the frequency domain, and a plurality of transform coefficients each indicating the level of the corresponding frequency component among the levels of the plurality of frequency components included in the digital signal are generated,
After the trigger signal is detected, digital watermark information is embedded by changing a value of at least one of the plurality of transform coefficients.

In a digital signal processing method according to a third aspect of the present invention, a digital signal is frequency-analyzed and converted from a time domain to a frequency domain, and a corresponding one of levels of a plurality of frequency components included in the digital signal is analyzed. A plurality of transform coefficients each indicating the level of the frequency component to be generated are generated, and whether or not digital watermark information is embedded is detected based on the generated change pattern of the plurality of transform coefficients.

In a digital signal processing method according to a fourth aspect of the present invention, the digital signal is frequency-analyzed in units of a conversion processing block composed of a predetermined amount of digital signals, and is converted from a time domain to a frequency domain. Generate a plurality of transform coefficients each indicating the level of the corresponding frequency component among the levels of the plurality of frequency components included in the signal, and between the plurality of transform processing blocks in which the transform has been continuously performed, A difference between predetermined transform coefficients among the generated transform coefficients is detected, and whether or not the digital watermark information is embedded is detected based on a result of the detection.

In a digital signal processing method according to a fifth aspect of the present invention, a digital signal is frequency-analyzed, converted from a time domain to a frequency domain, and a corresponding one of levels of a plurality of frequency components included in the digital signal is analyzed. A plurality of transform coefficients each indicating the level of the frequency component to be generated are generated, and the values of the plurality of transform coefficients indicating the levels of the plurality of frequency components continuously located on the frequency among the generated plurality of transform coefficients are calculated. A difference is detected, and whether the digital watermark information is embedded is detected based on the difference.

[0027]

Embodiments of the present invention will be described below. In the present embodiment, as a digital signal processing device of the present invention, a digital watermark information embedding device that embeds digital watermark information in a digital audio signal, a digital watermark information detecting device that detects digital watermark information from the digital audio signal, and a digital audio signal Although a recording device and a reproducing device are exemplified, the present invention detects digital watermark information from a digital image signal, and a digital watermark information embedding device that embeds digital watermark information in a digital image (image) signal, in addition to a digital audio signal. The present invention can be similarly applied to a digital watermark information detecting device, a digital image signal recording device, and a reproducing device.

First, a digital watermark information embedding device according to an embodiment of the present invention will be described. First Embodiment FIG. 1 is a configuration diagram of a digital watermark information embedding device 100 of the present embodiment. As shown in FIG. 1, the digital watermark information embedding device 100 includes a digital input I / F circuit 11, an analog input I / F circuit 12, an A / D conversion circuit 13, and a memory 14.
(Storage means), FFT (Fast Fourier Transform) circuit 1
5 (orthogonal transformation means), IFFT (Inverse Fast Fourier)
Transform) circuit 16 (orthogonal inverse transformation means), signal compression circuit 17, digital output I / F circuit 18 (output means), MP
A U (Micro Processor Unit) 19 (information embedding means), an internal bus 20 and an operation unit 21 are provided.

The digital input I / F circuit 11 outputs the input original digital audio signal S 9 to the memory 14 via the internal bus 20 based on the transfer control of the MPU 19. The analog input I / F circuit 12 outputs the input original analog audio signal S10 to the A / D conversion circuit 13. The A / D conversion circuit 13 converts the original analog audio signal S10 input from the analog input I / F circuit 12 into a digital audio signal, and outputs the digital audio signal to the memory 14 via the internal bus 20.

The memory 14 has a digital input I / F circuit 1
1 and a digital audio signal input from the A / D conversion circuit 13, and based on control from the MPU 19, an FFT processing block (corresponding to a conversion processing block of the present invention) serving as a unit of frequency analysis processing in the FFT circuit 15 ), The stored 512-sample digital audio signal is read out and output to the FFT circuit 15 via the internal bus 20.

The FFT circuit 15 performs a frequency analysis on the digital audio signal input from the memory 14 by FFT processing, converts the digital audio signal from a time domain to a frequency domain, and outputs a conversion coefficient obtained by the conversion from the MPU 19. Internal bus 2 based on the control of
0 to the MPU 19. Here, FFT processing is often used when performing waveform analysis of digital signals such as digital audio signals and digital image signals because the arithmetic processing is fast and relatively easy. F
The FT circuit 15 basically performs FFT on samples of a power of 2 and, for example, as shown in FIG. 2A, cuts out a digital audio signal by 512 (2 9) samples at a time and performs FFT processing. For example, a frequency spectrum as shown in FIG. At this time, the frequency resolution is a value obtained by dividing the sampling frequency by the number of cut out samples. For example, as shown in FIG. 3, when the sampling frequency fs of the digital audio signal is 44.1 kHz and the number of samples N is 512, the frequency resolution T is about 86.1.
(= 44100/512) Hz.

The IFFT circuit 16 restores an audio signal by performing IFFT processing on the conversion coefficient input from the MPU 19 via the internal bus 20 based on the control from the MPU 19, and restores the restored audio signal via the internal bus 20. And outputs it to the digital output I / F circuit 18. At this time, as described later, among the conversion coefficients input from the MPU 19, the conversion coefficient of a predetermined frequency component is set to 0 as necessary based on the digital watermark information. That is, I
Digital watermark information is already embedded in the audio signal restored by the FFT circuit 16 as necessary.

The digital output I / F circuit 18 is connected to the IFFT circuit 16 or the signal compression circuit 1 via the internal bus 20.
The digital audio signal S18 input from 7 is recorded on a recording medium 30 such as an MD or a CD-R.

The signal compression circuit 17 converts, for example, a digital audio signal embedded with digital watermark information output from the IFFT circuit 16 into an MPEG (Movin
g Picture Experts Group) and outputs the compressed digital audio signal to a digital output I / F circuit 1
8 is output.

The MPU 19 is a digital watermark information embedding device 1
Control of each component in 00 is performed in an integrated manner. MPU19
Has a built-in DMA (Direct Memory Access) controller and controls data transfer between components via the internal bus 20.

The MPU 19 monitors the storage state of the memory 14, and as described above, converts the digital audio signal of the data amount necessary for the FFT processing in the FFT circuit 15 into the digital input I / F circuit 11 or the A / D converter. When it is confirmed that the digital audio signal is stored in the memory 14 from the circuit 13, the digital audio signal stored in the memory 14 is controlled to be output to the FFT circuit 15.

The MPU 19 has a RAM for storing information for specifying a frequency component to be varied based on digital watermark information to be embedded.
Of the transform coefficients input from the FFT circuit 15 via “0”, the transform coefficient corresponding to the frequency component specified from the information stored in the RAM is set to “0”. And
The MPU 19 stores the variation-added conversion coefficient and the variation-free conversion coefficient input from the memory 14 into the internal bus 2.
0 to the IFFT circuit 16. In the present embodiment, the MPU 19 embeds digital watermark information indicating “copy prohibited” by setting the conversion coefficient C _f1 of the frequency component f ₁ to “0”, and inserts digital watermark information indicating “only one generation can be copied”. Embedding is performed by setting the conversion coefficient C _f2 of the frequency component f ₂ to “0”. At this time, as the frequency components f ₁ and f _{2 in} which the digital watermark information is embedded, for example, frequency components that are hardly perceived by human hearing are selected.

FIGS. 4 and 5 are flowcharts for explaining the process of embedding digital watermark information in the MPU 19. Step S1: For example, when the MPU 19 determines that the processing start instruction is issued by the operation of the operation unit 21 by the user, the MPU 19 includes an internal variable count indicating the number of samples of the digital audio signal stored in the memory 14, and FF
FFT which is a unit for performing the FFT processing in the T circuit 15
Internal variable FE indicating the number of samples constituting the processing block
Initialize T_sample. Specifically, “0” is substituted for the internal variable count, and the internal variable FET_sa
Substitute “512” for mple.

Step S2: The MPU 19 receives an instruction to stop processing by the operation of the operation unit 21 by the user, or inputs an audio signal from both the digital input I / F circuit 11 and the analog input I / F circuit 12. If at least one of the non-operations is detected, the process is terminated, and otherwise, the process of step S3 is executed.

Step S3: The MPU 19 outputs the digital audio signal S9 for one sample to the digital input I / F.
Each time the digital audio signal is input to the circuit 11, the digital audio signal for one sample is stored in the memory 14 via the internal bus 20, and the internal variable count is increased by one.

Step S4: The MPU 19 sets the internal variable c
It is determined whether or not “out” matches the internal variable FET_sample. If it is determined that they match, the process of step S5 is performed. If it is determined that they do not match, the process returns to step S2. That is, the MPU 19
Digital audio signals for 512 samples are stored in memory 1.
4, the processing of step S5 is performed.

Step S5: The MPU 19 sets the internal variable c
“0” is substituted for “out”.

Step S6: The MPU 19 stores in the memory 14
Is transferred to the FFT circuit 15 via the internal bus 20. As a result, the FFT circuit 15 performs the FFT processing on the digital audio signal for 512 samples, and outputs the transform coefficients obtained by the FFT processing to the MPU 19 via the internal bus 20.

Step S7: The MPU 19 determines whether or not the digital watermark information to be embedded indicates "copy prohibition", and if it indicates "copy prohibition", the process of step S8. Is carried out, and if it does not indicate "copy prohibited", the process of step S9 is executed.

Step S8: This is executed when the digital watermark information to be embedded indicates “copy prohibited”, and the MPU 19 sets the conversion coefficient C _f1 corresponding to the frequency component f ₁ to “0”.

Step S 9: The MPU 19 determines that the digital watermark information to be embedded is “only one generation can be copied”
Is determined, and if it indicates "only one generation can be duplicated", the process of step S10 is executed. If it does not indicate "only one generation can be duplicated", the process proceeds to step S10. The process of step S11 is performed.

Step S10: This step is executed when the digital watermark information to be embedded indicates "only one generation can be copied", and the MPU 19 sets the conversion coefficient C _f2 corresponding to the frequency component f ₂ to “0”. To That is, if the digital watermark information to be embedded is neither “copy prohibited” nor “copy only one generation”,
No variation is given to any frequency components.

Step S11: If step S8 or S10 is performed, the MPU 19 executes step S8.
Alternatively, the conversion coefficient set to “0” in S10 and the other conversion coefficients input from the memory 14 are stored in the internal bus 2
0 to the IFFT circuit 16. Also, MPU
19 outputs the transform coefficients input from the memory 14 to the IFFT circuit 16 via the internal bus 20 when neither step S8 nor S10 is performed. As a result, the IFFT processing is performed in the IFFT circuit 16 using the conversion coefficient input from the MPU 19, and the digital audio signal is restored.

Step S12: The MPU 19 executes the IFFT
The digital audio signal restored in the circuit 16 is output to the digital output I / F circuit 18 via the internal bus 20. Thereby, the digital audio signal S18 output from the digital output I / F circuit 18 is recorded on the recording medium 30.

Step S13: The MPU 19 stores the memory 1
4 after erasing (clearing) the stored contents of step S2.
Return to the processing of.

In the digital watermark information embedding device 100, when “0” is substituted for the transform coefficient C _f1 of the frequency component f ₁ of the frequency spectrum of one FFT processing block (when digital watermark information is embedded) ), The FFT
Transform coefficient C of FFT processing block before and after processing block
It is assumed that electronic watermark information is not embedded in _f1 . Similarly, in the digital watermark information embedding apparatus 100, the frequency component f of the frequency spectrum of one FFT processing block
_When “0” is substituted for the _second transform coefficient C _f2 (when digital watermark information is embedded), digital watermark information is embedded in the transform coefficients C _f2 of the FFT processing blocks before and after the FFT processing block. Make it not exist. This is to enable the digital watermark information detection device to determine whether to embed digital watermark information based on a temporal change in the conversion coefficient, as described later.

As described above, according to the digital watermark information embedding apparatus 100, among the transform coefficients of the frequency spectrum obtained by performing the FFT processing on the original digital audio signal S9, the predetermined frequency component Digital watermark information is embedded depending on whether the conversion coefficient is set to “0”. Therefore, according to the digital watermark information embedding device 100, compared to the conventional method of superimposing pseudorandom noise on the original digital audio signal, deterioration of sound quality due to embedding of digital watermark information is perceived by humans. Can be suppressed. That is, according to the digital watermark information embedding device 100, one FFT processing block (512 samples)
By setting a specific frequency component to “0”, digital watermark information can be embedded, and the time during which digital watermark information is embedded is significantly shorter than when conventional pseudo-random noise is superimposed. Not perceived.

Since the time for embedding the digital watermark information is short as described above, the digital watermark information is embedded at the timing when the amplitude of the original digital audio signal sharply increases. When outputting a sound corresponding to the digital audio signal, the masking effect makes it possible to make the influence of the embedded digital watermark information on the sound hardly perceived. At this time, in order to further enhance the masking effect, the masking effect is further enhanced by embedding the digital watermark information only in a peak portion having a certain degree of sharpness by observing how the amplitude of the original digital audio signal rises.

Second Embodiment FIG. 6 is a block diagram of a digital watermark information embedding device 200 according to the second embodiment . As shown in FIG. 6, the digital watermark information embedding device 200 includes a digital input I / F circuit 11, an analog input I / F circuit 12, an A / D conversion circuit 13, and a memory 14.
(Storage means), FFT circuit 15 (orthogonal transformation means), IF
FT circuit 16 (orthogonal inverse transform means), signal compression circuit 17,
Digital output I / F circuit 18 (output means), MPU 21
9 (information embedding means), an internal bus 20 and an operation unit 21.

Here, digital input I / F circuit 11, analog input I / F circuit 12, A / D conversion circuit 13, memory 14, FFT circuit 15, IFFT circuit 16, signal compression circuit 17, digital output I / F Circuit 18, internal bus 20
The operation unit 21 is the same as that described in the first embodiment. That is, the digital watermark information embedding device 200
Means that the processing in the MPU 219 is the MPU of the first embodiment.
19 is different. Hereinafter, the processing in the MPU 219 will be mainly described.

FIGS. 7 and 8 are flowcharts for explaining the process of embedding digital watermark information in the MPU 219. Steps S1 to S6 shown in FIG.
Is performed in step S of the MPU 19 shown in FIG.
This is the same as the processing of 1 to S6. Steps S11 to S13 shown in FIG. 8 are the same as the processes of steps S11 to S13 of the MPU 19 shown in FIG. 5 described above.

Hereinafter, steps S21 to S28 shown in FIG.
Will be described. Step S21: The MPU 219 determines whether or not the digital watermark information to be embedded indicates "copy prohibition", and if it indicates "copy prohibition", executes the process of step S22. If it does not indicate "copy prohibited", the process of step S25 is executed.

Step S22: This is executed when the digital watermark information to be embedded indicates "copy prohibited", and the MPU 219 adds all the conversion coefficients of 512 samples input from the memory 14, for example. To calculate the ratio of the conversion coefficient C _f1 of the predetermined frequency component f ₁ to the calculated total energy Energy. If the ratio is smaller than the predetermined value A, the process of step S23 is performed. Run
If the value is equal to or larger than the predetermined value A, the process of step S24 is executed. Here, the frequency component f ₁ is specified in advance by information stored in a RAM built in the MPU 219. Further, the predetermined value A is, for example, 1/100, and is determined in consideration of the effect on the sound quality by changing the conversion coefficient.

Step S23: When the ratio of the conversion coefficient C _f1 of the frequency component f ₁ to the total energy Energy is smaller than the predetermined value A, the MPU 219
_{Set f1} to “0”.

Step S24: The MPU 219 sets the conversion coefficient C _f1 to C _f1 / 2 when the ratio of the conversion coefficient C _f1 of the frequency component f ₁ to the total energy Energy is equal to or larger than the predetermined value A. This is because when the ratio of the conversion coefficient C _f1 of the frequency component f ₁ to the total energy Energy is equal to or more than the predetermined value A, the conversion coefficient C _{f1 is set} to “0”, and a variation is added to the conversion coefficient. This is because the influence on the sound quality is relatively large, and the effect is reduced by setting the conversion coefficient C _f1 to C _f1 / 2.

Step S25: The MPU 219 determines whether the digital watermark information to be embedded indicates "only one generation can be duplicated", and indicates that "only one generation can be duplicated". Executes the processing of step S26, and if it does not indicate "only one generation can be copied", executes the processing of step S11.

Step S26: This is executed when the digital watermark information to be embedded indicates "copy prohibited", and the MPU 219 adds all the conversion coefficients of 512 samples input from the memory 14 for example. To calculate the ratio of the conversion coefficient C _f2 of the frequency component f ₂ to the calculated total energy Energy. If the ratio is smaller than the predetermined value A, the process of step S27 is executed. If the value is equal to or larger than the predetermined value A, the process of step S28 is executed. Here, the frequency component f ₂ is determined by the R
It is specified in advance by information stored in the AM.

Step S27: The MPU 219 sets the conversion coefficient C _f2 to “0” when the ratio of the conversion coefficient C _f2 of the predetermined frequency component f ₂ to the total energy Energy is smaller than the predetermined value A. That is, if the digital watermark information to be embedded is neither “copy prohibited” nor “copy only one generation”, no variation is given to any frequency component.

Step S28: The MPU 219 checks all the energy
Frequency component f for Luggy Energy_TwoTranslator
Number C_f2Is greater than a predetermined value A, the conversion coefficient C_f2To
"C _f2/ 2 ".

In the digital watermark information embedding apparatus 200, similarly to the digital watermark information embedding apparatus 100, the conversion coefficient C _f1 of the frequency component f ₁ of the frequency spectrum of one FFT processing block is “0” or "C _f1 /
When “2” is substituted (when digital watermark information is embedded), digital watermark information is not embedded in the transform coefficient C _f1 of the FFT processing block before and after the FFT processing block. Similarly, in the digital watermark information embedding device 200, when “0” or “C _f2 / 2” is substituted into the transform coefficient C _f2 of the frequency component f ₂ of the frequency spectrum of one FFT processing block (electronic When the watermark information is embedded), the digital watermark information is not embedded in the transform coefficient C _f2 of the FFT processing block before and after the FFT processing block. This is because, as will be described later, the digital watermark information detection device determines whether or not digital watermark information is embedded based on the temporal change of the conversion coefficient.

As described above, according to the digital watermark information embedding device 200, unlike the first embodiment, the conversion coefficients C _f1 and C _f2 of the frequency components f ₁ and f ₂ with respect to the total energy Energy are different. When the ratio is equal to or greater than the predetermined value A, if the conversion coefficients C _f1 and C _f2 are set to “0” when embedding the digital watermark information, the influence on the sound quality due to the variation added to the conversion coefficients becomes relatively large. _Therefore , the conversion coefficients C _f1 and C _f2
Are set to C _f1 / 2 and C _f2 / 2, respectively, to reduce the effect. Therefore, the digital watermark information embedding device 20
0, compared to the digital watermark information embedding device 100,
Deterioration of sound quality due to embedding of digital watermark information in the original digital audio signal can be further suppressed.

Third Embodiment FIG. 9 is a block diagram of a digital watermark information embedding device 300 according to the third embodiment . As shown in FIG. 9, the digital watermark information embedding device 300 includes a digital input I / F circuit 11, an analog input I / F circuit 12, an A / D conversion circuit 13, and a memory 14.
(Storage means), FFT circuit 15 (orthogonal transformation means), IF
FT circuit 16 (orthogonal inverse transform means), signal compression circuit 17,
Digital output I / F circuit 18 (output means), MPU 31
9 (trigger signal detection means and information embedding means), an internal bus 20 and an operation unit 21.

Here, digital input I / F circuit 11, analog input I / F circuit 12, A / D conversion circuit 13, memory 14, FFT circuit 15, IFFT circuit 16, signal compression circuit 17, digital output I / F Circuit 18, internal bus 20
The operation unit 21 is the same as that described in the first embodiment. That is, the electronic watermark information embedding device 300
Means that the processing in the MPU 319 is the MPU of the first embodiment.
19 is different. Hereinafter, the processing in the MPU 319 will be mainly described.

FIGS. 10 and 11 are flowcharts for explaining the process of embedding digital watermark information in MPU 319. Step S31: When the MPU 319 determines that the processing start instruction has been issued by the operation of the operation unit 21 by the user, for example, the internal variable count indicating the number of audio signal samples stored in the memory 14 and the FFT in the FFT circuit 15 Internal variable FET_s indicating the number of samples constituting a unit (FFT processing block) for performing processing
sample, an internal variable flag_peak indicating whether or not the input sample is a trigger signal, and an internal variable flag_coef indicating whether the conversion coefficient of a predetermined frequency component after frequency analysis is equal to or smaller than a predetermined value B. initialize. Specifically, the internal variables count, flag_p
“0” is substituted for eak and flag_coef, and “512” is substituted for the internal variable FET_sample.

Step S32: The MPU 319 sends an instruction to stop the processing by operating the operation unit 21 by the user, or inputs an audio signal from both the digital input I / F circuit 11 and the analog input I / F circuit 12. If at least one of the non-operations has been detected, the process is terminated. Otherwise, the process of step S33 is executed.

Step S33: The MPU 319 monitors the amplitude level of the sample of the digital audio signal input to the digital input I / F circuit 11, and when it is determined that the amplitude level is the peak level, the sample is triggered. If it is determined that the signal is a signal, the process of step S34 is executed, and if not, the process of step S35 is executed.

Step S34: The MPU 319 substitutes "1" for the internal variable flag_peak.

Step S35: The MPU 319 determines whether or not the internal variable flag_peak is "1".
If it is determined to be "1", the process of step S36 is executed, and if it is not "1", the process of step S49 shown in FIG. 11 is executed.

Step S36: The MPU 319 stores the audio signal S9 of one sample input to the digital input I / F circuit 11 in the memory 14 via the internal bus 20, and increases the internal variable count by "1". I do.

Step S37: The MPU 319 determines whether or not the internal variable count and the internal variable FET_sample match, and if so, executes the processing of step S39 and determines that they do not match. In this case, the process returns to step S32. That is, MPU
In step 319, when the audio signal for 512 samples is stored in the memory 14, the process in step S39 is performed.

Step S39: The MPU 319 substitutes "0" for the internal variable "count".

Step S40: The MPU 319 transfers the audio signal of 512 samples stored in the memory 14 to the FFT circuit 15 via the internal bus 20.
As a result, the FFT circuit 15 performs the FFT processing on the audio signal for 512 samples, and converts the transform coefficients obtained by the FFT processing into the internal bus 20.
Is output to the MPU 319 via the.

Step S41: The MPU 319 determines whether or not the conversion coefficient C _f1 of the predetermined frequency component f ₁ is smaller than the predetermined value B. If the conversion coefficient C _f1 is smaller than the predetermined value B, the MPU 319 executes the processing of step S42. The process is executed, and if it is determined that the value is equal to or larger than the predetermined value B, the process of step S43 is executed.
Here, the frequency component f ₁ is specified in advance by information stored in a RAM built in the MPU 319.

Step S42: The MPU 319 substitutes "1" for the internal variable flag_coef.

Step S43: The MPU 319 substitutes “0” for the conversion coefficient C _f1 .

Step S44: The MPU 319 outputs to the IFFT circuit 16 via the internal bus 20 the conversion coefficient C _f1 to which “0” has been substituted in step S43 and the other conversion coefficients input from the memory 14. As a result, the IFFT circuit 16 performs IFFT processing using the conversion coefficient input from the memory 14 to restore the digital audio signal.

Step S45: The MPU 319 sets the amplitude level (peak level) of the sample determined to be the trigger signal in step S33 in the sample of the digital audio signal restored in step S44.
It is determined whether or not a sample exceeding the number exists, and if it is determined that the sample exists, the process of step S46 is executed. If it is determined that the sample does not exist, the process of step S47 is executed. Normally, in step S41, when it is determined that the conversion coefficient C _f1 is equal to or larger than the predetermined value B (when the internal variable flag_coef is “0”), the MPU
319 executes the process of step S46.

Step S46: In step S45, the MPU 319 lowers the amplitude level of the sample determined to be a sample having an amplitude level exceeding the peak level below the peak level. This ensures proper detection of the trigger signal when detecting digital watermark information.

Step S47: The MPU 319 determines whether or not the internal variable flag_coef is “1”.
If it is determined to be “1”, the process of step S48 is executed, and if it is not “1”, the process of step S40 is executed. That is, when it is determined in step S41 that the transform coefficient C _f1 of the frequency component f ₁ is equal to or _larger than the predetermined value B, the process returns to step S40 and the FFT processing is performed again. Thereby, step S
When the processing in step S46 is performed, the processing in steps S40 to S47 is repeated until the conversion coefficient C _f1 after the frequency analysis in step S40 becomes smaller than the predetermined value B, and the conversion coefficient C _f1 when detecting the digital watermark information. Can be detected more accurately.

Step S48: The MPU 319 substitutes “0” for the internal variables flag_peak and flag_coef.

Step S49: The MPU 319 sets the IFF
The digital audio signal restored in the T circuit 16 is output to the digital output I / F circuit 18 via the internal bus 20. Thereby, the digital output I / F circuit 18
Digital audio signal output from the recording medium 3
Recorded as 0.

Step S50: The MPU 19 stores the memory 1
4 after erasing (clearing) the stored contents of step S3
It returns to the process of 2.

In the digital watermark information embedding apparatus 300, similarly to the digital watermark information embedding apparatus 100 described above, “0” is set to the transform coefficient C _f1 of the frequency component f ₁ of the frequency spectrum of one FFT processing block. When it is substituted (when digital watermark information is embedded), the FFT
Transform coefficient C of FFT processing block before and after processing block
It is assumed that electronic watermark information is not embedded in _f1 .

As described above, according to the digital watermark information embedding device 300, the digital watermark information is embedded only immediately after the trigger signal.
The operation of detecting the digital watermark information only needs to be performed immediately after the detection of the trigger signal, and the load is reduced. Further, according to the digital watermark information embedding device 300, after the IFFT processing,
If there is a sample having an amplitude level exceeding the peak level, the digital watermark information detection device can accurately detect a trigger signal because the amplitude level of the sample is lowered to a level lower than the peak level. Further, according to the digital watermark information embedding device 300, the processing of steps S40 to S47 is repeated until the conversion coefficient C _f1 after the frequency analysis in step S40 becomes smaller than the predetermined value B. , The variation of the conversion coefficient C _f1 can be detected more accurately.

Other Embodiments of Digital Watermark Information Embedding Apparatus
In the first to third embodiments described above, the MPUs 19, 219,
319, the frequency components f ₁ and f _{2 in} which the digital watermark information is embedded are specified in advance by the information stored in the built-in RAM.
9, 219, and 319 may dynamically determine the frequency component in which the digital watermark information is embedded based on the frequency spectrum obtained by the FFT circuit 15. In particular,
For example, even if one frequency component adjacent to a frequency component having a peak level is set to “0” or a frequency component having an average level is set to “0” based on a frequency spectrum, digital watermark information is embedded. Good.

In the first to third embodiments described above, M
When embedding digital watermark information in the PU 19, 219, 319, the conversion coefficient of a predetermined frequency component is set to “0”.
However, the value may be other than “0”. Instead of assigning a predetermined value to the conversion coefficient, a predetermined amount of value may be added or subtracted from the value of the conversion coefficient.

In the above-described first to third embodiments, embedding of one piece of digital watermark information is performed by adding variation to one frequency component.
In order to determine whether or not digital watermark information is embedded based on a temporal change in the transform coefficient, a variation is added to the transform coefficient of a predetermined frequency component of the frequency spectrum of one FFT processing block (digital watermark information Is embedded), the digital watermark information is not embedded in the transform coefficients of the FFT processing blocks before and after the FFT processing block. However, in the present invention, the embedding of one piece of digital watermark information may be performed, for example, by giving a plurality of adjacent frequency components a stepwise level. Specifically, for example, the conversion coefficient C of the continuous frequency components f ₁ , f ₂ , f _3
f1 , _Cf2 , and _Cf3 are "0", "10", and "2", respectively.
A level that changes stepwise with an equal difference amount such as “0” or a level that changes stepwise with multiples such as “10”, “20”, and “40” may be provided. Good.

The orthogonal transform means is not limited to FFT, but may be DCT (Discrete Cosine Transform) or MDCT (M
Digital watermark information may be embedded by adding variation to the coefficients of each frequency component obtained by, for example, odified discrete cosine transform.

Hereinafter, a digital watermark information detecting device according to an embodiment of the present invention will be described. Fourth Embodiment In the present embodiment, a digital watermark information detecting device that detects digital watermark information from a digital audio signal in which digital watermark information is embedded using the digital watermark information embedding device of the first embodiment will be described. I do.

FIG. 12 is a configuration diagram of the digital watermark information detecting device 400 of the present embodiment. As shown in FIG. 12, the digital watermark information detection device 400 includes a digital input I / F circuit 411, a memory 412 (storage means), and an FFT circuit 41.
3 (orthogonal conversion means), signal decompression circuit 414, digital output I / F circuit 415 (output means), D / A conversion circuit 41
6,417, analog output I / F circuits 418,419
(Output means), internal bus 420, display 421,
An MPU 422 (information detecting means) and an operation unit 423 are provided.

The digital input I / F circuit 411 converts an audio signal input from a recording medium (package medium) 30 such as an MD, a CD-R or a DVD, the Internet, or the like into an internal bus 420 based on transfer control of the MPU 422. Is output to the memory 412 via.

The memory 412 stores a digital audio signal input from the digital input I / F circuit 411 or the signal decompression circuit 414, and performs FFT processing as a unit of frequency analysis processing in the FFT circuit 413 based on control from the MPU 422. When an audio signal (512 samples in this embodiment) constituting a block is stored,
The stored digital audio signal is read and output to the FFT circuit 413 via the internal bus 420.

[0098] The FFT circuit 413 performs FFT processing on the digital audio signal input from the memory 412 in units of FFT processing blocks, performs frequency analysis to convert the time domain to the frequency domain, and performs conversion based on control from the MPU 422. The coefficient is transferred to the MPU 42 via the internal bus 420.
Output to 2.

The signal expansion circuit 414 has a digital input I /
If the digital audio signal input from the F circuit 411 is compressed, the input digital audio signal is expanded and stored in the memory 412.

When the digital audio signal S30 can be copied, the digital output I / F circuit 415
Based on the control from U422, the digital audio signal read from the memory 412 is input and output as a digital audio signal S415. The A / D conversion circuits 416 and 417 convert the digital audio signals input from the memory 412 into analog audio signals based on the control from the MPU 422, and convert the analog audio signals into analog output I / F circuits 418 and 41.
9 respectively.

The MPU 422 performs overall control of each component in the digital watermark information detection device 400. Also, M
The PU 422 has a built-in DMA (Direct Memory Access) controller and controls data transfer between components via an internal bus 420.

Further, the MPU 422 monitors the storage state of the memory 412, and as described above, the digital audio signal constituting the FFT processing block which is the unit of the FFT processing in the FFT circuit 413 is transmitted from the digital input I / F circuit 411. When it is confirmed that the digital audio signal is stored in the memory 412, control is performed so that the digital audio signal stored in the memory 412 is output to the FFT circuit 413.

[0103] The MPU 422 inputs a transform coefficient as a result of the FFT processing from the FFT circuit 413 via the internal bus 420 and stores it in a built-in RAM. And M
The PU 422, based on the information that specifies the transform coefficient of the frequency component in which the digital watermark information is stored in advance in the RAM and which is stored in the RAM from the FFT circuit 413, converts the transform coefficient specified by the information. Is calculated, and it is determined that the digital watermark information is embedded in the conversion coefficient whose temporal change rate is a certain value or more. Then, the MPU 422 determines that the frequency component determined to have the digital watermark information embedded therein is “copy prohibited”.
Indicates that "copy prohibited" is displayed on the display 421 and, for example, the digital output I
/ F circuit 415 is controlled so that a digital audio signal cannot be output. However, the MPU 422 outputs the analog audio signal to the analog output I / F circuits 418, 41.
9 is controlled so that it can be output. In this case, when a recording device capable of analog input is connected to the analog output I / F circuits 418 and 419, the audio signal can be copied (copied). However, the duplication is performed based on the digital watermark information. It is also possible to prohibit the duplication by providing the prohibition function in the recording device.

On the other hand, if the frequency component determined that the digital watermark information is embedded indicates “permit one generation copy,” the MPU 422 displays the display 4.
In addition to displaying "permission to copy only one generation" at 21, the digital output I / F circuit 415 enables the output of a digital audio signal. Also, the generation restriction information is rewritten to “copy prohibited”.

When the MPU 422 determines that the digital watermark information is not embedded, it displays "copy permitted" on the display 421. In addition, MPU422
Is "copy permitted", for example, the memory 41
2 is converted to a digital output I /
F circuit 415 and analog output I / F circuits 418 and 4
Control is performed so that output can be performed from step 19. In addition, the MPU 422
It is determined that the digital watermark information indicates "copy permitted", and thereafter,
If it is determined that the decoding of the digital watermark information is unnecessary, the audio signal input from the digital input I / F circuit 411 is converted to the digital output I / F via the internal bus 420.
It is also possible to control so as to directly output to the circuit 415 and the D / A conversion circuits 416 and 417.

As described above, the display 421 displays "copy prohibited", "copy permitted only for one generation" and "copy permitted".
In addition to displaying "copy permitted", for example, as shown in Fig. 13, a track number or a title of a digital audio signal read from the recording medium 30 may be displayed. , A liquid crystal, a cathode ray tube or a device using a light emitting diode is used, and a GUI (Graph
ical User Interface). Instead of the display 421, “copy prohibited”, “
A speaker that outputs audio such as "copy permitted only for one generation" and "copy permitted" may be used.

FIGS. 14 and 15 are flowcharts for explaining the process of decrypting digital watermark information in MPU 422. In the decryption process described below, digital watermark information indicating “copy prohibited” and “copy only one generation” is converted into a digital audio signal by setting the corresponding conversion coefficient of a predetermined frequency component to “0”. FFT circuit 41 for three embedded and temporally continuous FFT processing blocks (512 × 3 samples)
3 illustrates a case where the digital watermark information is decrypted based on the temporal change of the conversion coefficient obtained in step 3. Here, one F
When digital watermark information is embedded in one frequency component of the FT processing block, it is assumed that digital watermark information is not embedded in the frequency components of the FFT processing blocks before and after the digital watermark information.

Step S61: If the MPU 422 determines that a reproduction instruction has been issued by a user's operation of the operation unit 423, for example, the internal variables count and FET
_Sample, flag_copy_en are initialized. Specifically, “0” is substituted for the internal variable “count”, “512” is substituted for the internal variable FET_sample, and “0” is substituted for the internal variable “flag_copy_en”.

Step S62: The MPU 422 detects at least one of a processing stop instruction issued by the user operating the operation unit 423 and a digital audio signal not being input from the digital input I / F circuit 411. If so, the process ends; otherwise, the process of step S63 is executed.

Step S63: The MPU 422 determines whether or not the internal variable flag_copy_en is “1”, and if it determines that it is “1”, the step S65.
Is performed, and if it is determined that the value is not "1", the process of step S64 is performed.

Step S64: The MPU 422 performs control to stop outputting audio signals from the digital output I / F circuit 415 and the analog output I / F circuits 418 and 419.

Step S65: Every time the digital audio signal S30 for one sample is input to the digital input I / F circuit 411, the MPU 422 transfers the digital audio signal for one sample to the memory 412 via the internal bus 420. Then, the internal variable count is increased by “1”.

Step S66: The MPU 422 determines whether or not the internal variable count and the internal variable FET_sample match, and if so, executes the processing of step S67 and determines that they do not match. In this case, the process returns to step S62. That is, MPU
In step 422, when the audio signal for 512 samples is stored in the memory 412, the process in step S62 is performed.

Step S67: The MPU 422 substitutes “0” for the internal variable “count”.

Step S68: The MPU 422 converts the 512-sample digital audio signal stored in the memory 412 into the FFT circuit 41 via the internal bus 420.
Transfer to 3. Thus, the FFT circuit 413 performs the FFT processing on the digital audio signal for 512 samples, and outputs the transform coefficients obtained by the FFT processing to the MPU 422 via the internal bus 420. The conversion coefficient is stored in a RAM built in the MPU 422.

[0116] Step S69: MPU422 determines whether the digital watermark information into a frequency component f ₁ is embedded. Specifically, MPU422 is a conversion coefficient C _f1b frequency components f ₁ stored in the RAM in the previous step S68, step conversion factor C _f1a and second last frequency components f ₁ stored in the RAM in this step S68 S6
8, the conversion coefficient C _f1c of the frequency component f ₁ stored in the RAM
Preparative was divided by the value obtained by adding the division results "C _f1b / (C _f1a
+ C _f1c ) ”is smaller than a predetermined value C such as 1/10 (when it is determined that the digital watermark information is embedded in the frequency component f ₁ ), step S70 is executed, and If (determined watermark information into a frequency component f ₁ is not embedded) executes step S72. This is a process to determine whether the electronic watermark information FFT processing on the frequency components f ₁ of 512 samples previously performed is embedded, when the electronic watermark information is not embedded, transform coefficients C _f1a , C _f1b, C _f1c becomes substantially the same value, the division result _{"C f1b / (C f1a + C} f1c) " is about 1/2.
On the other hand, when the digital watermark information is embedded in the frequency component f ₁ of 512 samples on which the FFT processing was performed last time, the transform coefficient C _f1b is “0” and the transform coefficient C _f1b is “0”.
_f1a, since C _f1c is not "0", even in consideration of an error of the FFT processing, the division result "C _f1b / (C _f1a +
C _f1c ) ”is, for example, 1/20 or less.

[0117] Step S70: the digital watermark information into a frequency component f ₁ in the step S69 is executed if it is determined that the embedded, MPU422 assigns "1" to the internal variable Flag_copy_en. Accordingly, after step S63 is performed by the MPU 422, the process of step S64 is subsequently performed, and the digital output I / F circuit 415 and the analog output I / F
No audio signal is output from the circuits 418 and 419.

Step S71: The MPU 422 displays "copy prohibited" on the display 421.

[0119] Step S72: the digital watermark information into a frequency component f ₁ in the step S69 is executed if it is determined that not embedded, MPU422 is whether digital watermark information into frequency components f ₂ is embedded to decide. More specifically, the MPU 422 calculates the conversion coefficient C _f2b of the frequency component f ₂ stored in the RAM in the previous step S68 and the conversion coefficient C _f2a of the frequency component f ₂ stored in the RAM in the current step S68 in the last two steps. S6
8, the conversion coefficient C _f2c of the frequency component f ₂ stored in the RAM
And the result of dividing by the value obtained by adding “C _f2b / (C _f2a
+ C _f2c ) ”is smaller than a predetermined value C such as 1/10 (when it is determined that the digital watermark information is embedded in the frequency component f ₂ ), step S73 is executed, and If (determined watermark information into frequency components f ₂ is not embedded) executes step S75. This is a process for determining whether or not the digital watermark information into frequency components f ₂ of 512 samples previously performed FFT processing is embedded, when the electronic watermark information is not embedded, transform coefficients C _f2a , C _f2b , and C _f2c have substantially the same value, and the division result “C _f2b / (C _f2a + C _f2c )” is about 1/2.
On the other hand, when the digital watermark information into frequency components f ₂ of 512 samples previously performed FFT processing is embedded, transform coefficients C _f2b has become "0", the conversion factor C
_f2a, C for _f2c is not both "0", even in consideration of an error of the FFT processing, the division result "C _f2b / (C _f2a + C
_f2c ) ”is, for example, 1/20 or less.

[0120] Step S73: the digital watermark information into frequency components f ₂ in step S72 is executed if it is determined that the embedded, MPU422 displays the "one generation duplication allowed" on the display 421. At this time, since the internal variable flag_copy_en remains “0”, if the process of step S63 is performed next, the process of step S65 is subsequently performed, and the audio signal stored in the memory 412 is converted to a digital output I / F. Circuit 41
5 and analog output I / F circuits 418 and 419.

Step S74: The MPU 422 rewrites the generation restriction information to “copy prohibited”. That is, the conversion coefficient C _f1a is rewritten to “0”.

[0122] Step S75: the digital watermark information into frequency components f ₂ in step S72 is executed if it is determined that not embedded, MPU422 displays a "copy permission" on the display 421. At this time,
Since the internal variable flag_copy_en remains “0”, the process of step S65 is performed when the process of step S63 is performed next, and the audio signal stored in the memory 412 is converted to the digital output I / F circuit 415 and the analog signal. Output is possible from the output I / F circuits 418 and 419.

Step S76: After erasing the contents stored in the memory 412, the MPU 422 returns to the processing of step S62.

As described above, according to the digital watermark information detecting device 400, the original digital audio signal is subjected to the FFT processing as in the digital watermark information embedding device 100 of the first embodiment. From the digital audio signal generated by using the digital watermark information embedding device that sets the conversion coefficient of the predetermined frequency component among the obtained conversion coefficients of the frequency spectrum to “0” and embeds the digital watermark information, the digital watermark information is obtained. Can be detected properly. Further, according to the digital watermark information detection device 400, by performing FFT processing on three consecutive FFT processing blocks, it is possible to detect whether or not digital watermark information is embedded in a specific frequency component. Can be detected in a short time.

Other Embodiments of Digital Watermark Information Detecting Apparatus
In the above-described fourth embodiment, an example has been described in which digital watermark information is detected from a digital audio signal in which digital watermark information is embedded by setting a predetermined frequency component to “0”. As in the form, the frequency components f ₁ and f _{2 with} respect to the total energy Energy
Is greater than or equal to a predetermined value C, the conversion coefficients C _f1 and C _f2 are set to C _f1 / 2 and C _f2 / 2 to detect digital watermark information embedded in the digital audio signal. The present invention can also be applied. However, in this case, when the digital watermark information is embedded, the division result “C _f1b
/ (C _f1a + C _f1c) "," C _f2b / (C _f2a +
C _f2c ) ”is about 1/4, so that the value of the predetermined value C used in steps S69 and S72 of FIG.
For example, 1/3 which is between / 2 and 1/4 is used.

When detecting digital watermark information from a digital audio signal in which digital watermark information is embedded by using the digital watermark information embedding device of the third embodiment, for example, step S62 shown in FIG. And S63
It is also possible to perform a process of detecting a trigger signal between the steps and to perform the processes after step S63 only when a trigger signal is detected.

Further, in the fourth embodiment, an example has been described in which digital watermark information is detected based on a temporal change of a transform coefficient of a predetermined frequency component. However, as described above, a plurality of adjacent frequency components are detected. In the case where digital watermark information is embedded by giving a predetermined change pattern such as a gradual change to the level of the digital watermark, the digital watermark is detected by detecting the level change pattern of the plurality of adjacent frequency components. Detect information. Also, the change pattern of the conversion coefficient is
A plurality of time domains and / or frequency domains may be detected, and the presence or absence of embedding of digital watermark information may be detected based on a logical sum or a logical product of the plurality of detection results.

Fifth Embodiment FIG. 16 is a block diagram of a digital audio signal reproducing apparatus 500 of the present embodiment. As shown in FIG. 16, the digital audio signal reproducing device 500 is the same as the analog output I / F circuit 4 of the digital watermark information detecting device 400 shown in FIG.
18 and 419 are connected to a speaker 501.

In the digital audio signal reproducing device 500, the digital audio signal S 30 read from the recording medium 30 is D / A converted in the digital watermark information detecting device 400, and the analog audio signal S 41
8, S419 is output to the speaker 501. And
From the speaker 501, the analog audio signal S41
8. Sound corresponding to S419 is output. At this time, if the digital watermark information detecting device 400 determines that the digital watermark information of “reproduction prohibited” is embedded, for example, the output of the analog audio signals S418 and S419 to the speaker 501 is prohibited. Good. The output to the speaker 501 may be the digital audio signal S415 shown in FIG.

The digital audio signal reproducing device 5
In the case where the digital watermark information is embedded by giving a plurality of adjacent frequency components a stepwise level instead of the digital watermark information detecting device 400, the plurality of adjacent frequency components are A digital watermark information detecting device that detects digital watermark information by detecting that the digital watermark information has a specific level may be used.

Sixth Embodiment FIG. 17 is a block diagram of a digital audio signal recording device 600 of the present embodiment. As shown in FIG. 17, the digital audio signal recording device 600 is a digital output I / F circuit 4 of the digital watermark information detecting device 400 shown in FIG.
15 is connected to a recorder 601.

In the digital audio signal recording device 600, the digital audio signal S30 read from the recording medium 30 is input to the digital watermark information detecting device 400. The following processing is performed, and "recordable"
Is determined. Then, when it is determined that “recording is possible”, the digital output I /
F circuit 415 and / or analog output I / F circuit 4
18, 419 from the digital audio signal S415 and / or the analog audio signal S to the recorder 601.
418 and S419 are output and recorded on a predetermined recording medium.

Note that the digital audio signal recording device 6
In the case where the digital watermark information is embedded by giving a plurality of adjacent frequency components a stepwise level instead of the digital watermark information detecting device 400, the plurality of adjacent frequency components are A digital watermark information detecting device that detects digital watermark information by detecting that the digital watermark information has a specific level may be used.

[0134]

As described above, according to the digital signal processing apparatus and method of the present invention, digital watermark information can be embedded in a digital signal without substantially lowering the quality of the digital signal. Further, according to the digital signal processing apparatus and method of the present invention, it is possible to detect digital watermark information embedded by changing a transform coefficient indicating a level of a predetermined frequency component. Further, according to the digital signal processing device of the present invention, it is possible to detect embedded digital watermark information by changing a conversion coefficient indicating a level of a predetermined frequency component, and control the output of a digital signal based on the detection result. According to the digital signal processing device of the present invention, digital watermark information embedded by changing a conversion coefficient indicating a level of a predetermined frequency component is detected, and a digital signal is recorded on a recording medium based on the detection result. Can be controlled.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a digital watermark information embedding device according to a first embodiment of the present invention.

FIG. 2 is a diagram for explaining processing in the FFT circuit shown in FIG. 1;

FIG. 3 is a diagram for explaining a process in the FFT circuit shown in FIG. 1;

FIG. 4 is a flowchart for explaining an operation of the digital watermark information embedding device shown in FIG. 1 and a process of embedding digital watermark information in an MPU;

FIG. 5 is a flowchart for explaining an operation of the digital watermark information embedding device shown in FIG. 1 and a process of embedding digital watermark information in the MPU;

FIG. 6 is a configuration diagram of a digital watermark information embedding device according to a second embodiment of the present invention.

FIG. 7 is a flowchart for explaining the operation of the digital watermark information embedding device shown in FIG. 6 and a process of embedding digital watermark information in the MPU;

FIG. 8 is a flowchart for explaining the operation of the digital watermark information embedding device shown in FIG. 6 and a process of embedding digital watermark information in the MPU;

FIG. 9 is a configuration diagram of a digital watermark information embedding device according to a third embodiment of the present invention.

FIG. 10 is a flowchart for explaining the operation of the digital watermark information embedding device shown in FIG. 9 and the processing for embedding the digital watermark information in the MPU.

FIG. 11 is a flowchart for explaining the operation of the digital watermark information embedding device shown in FIG. 9 and the processing for embedding digital watermark information in the MPU.

FIG. 12 is a configuration diagram of a digital watermark information detection device according to a fourth embodiment of the present invention.

FIG. 13 is a diagram for explaining an example of a screen displayed on the display shown in FIG. 12;

FIG. 14 is a flowchart for explaining the operation of the digital watermark information detecting device shown in FIG. 12 and a process of detecting digital watermark information in the MPU;

FIG. 15 is a flowchart for explaining an operation of the digital watermark information detecting device shown in FIG. 12 and a process of detecting digital watermark information in the MPU.

FIG. 16 is a configuration diagram of a digital audio signal reproducing device according to an embodiment of the present invention.

FIG. 17 is a configuration diagram of a digital audio signal recording device according to an embodiment of the present invention.

[Explanation of symbols]

11, 411: digital input I / F circuit, 12: analog input I / F circuit, 13: A / D conversion circuit, 14, 41
2: memory, 15,413: FFT circuit, 16: IFF
T circuit, 17 ... Signal compression circuit, 18,415 ... Digital output I / F circuit, 19,219,319,422 ... MP
U, 20, 420 ... internal bus, 21, 423 ... operation unit,
30: recording medium, 416, 417: D / A conversion circuit, 4
21: display, 418, 419: analog output I
/ F circuit

Continued on the front page (72) Inventor Yumi Kato 6-7-35 Kita-Shinagawa, Shinagawa-ku, Tokyo Sony Corporation F-term (reference) 5D044 AB05 AB07 EF03 FG10 FG18 GK07 GK17 HL08 HL09 HL11

Claims (34)

[Claims] A digital signal is frequency-analyzed and converted from a time domain to a frequency domain to generate a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency components included in the digital signal. A digital signal processing apparatus, comprising: an orthogonal transformation unit that performs a transformation; and an information embedding unit that embeds digital watermark information into the digital signal by changing a value of at least one of the plurality of transformation coefficients. 2. An orthogonal inverse transform for orthogonally transforming a digital signal represented in a frequency domain into a time domain by a transform coefficient whose value has been changed and a transform coefficient whose value has not been changed among the plurality of transform coefficients. 2. The digital signal processing device according to claim 1, further comprising means. 3. The digital signal processing apparatus according to claim 1, wherein said information embedding means changes a predetermined value of said conversion coefficient when embedding said digital watermark information. 4. The information embedding means according to claim 3, wherein said plurality of electronic watermark information is embedded in said digital signal by changing a value of said predetermined conversion coefficient for each of said plurality of electronic watermark information. Digital signal processing equipment. 5. The digital signal processing apparatus according to claim 1, wherein said orthogonal transformation means transforms said digital signal from said time domain to said frequency domain in units of a transformation processing block comprising a predetermined amount of digital signal. 6. The information embedding means, when the digital watermark information is embedded by changing a value of a conversion coefficient indicating a level of a predetermined frequency component obtained from one of the conversion processing blocks, 7. The digital signal processing device according to claim 6, wherein a value of a conversion coefficient indicating a level of the predetermined frequency component of the conversion processing block processed before and after the processing block is not changed. 7. The digital signal processing apparatus according to claim 1, wherein said information embedding means sets one value selected from a plurality of values to said at least one transform coefficient. 8. The information embedding means includes: a ratio of a value of the one transform coefficient to a value obtained by adding a plurality of values of the transform coefficients obtained from a transform processing block to which the transform is integrally performed; The digital signal processing device according to claim 7, wherein the one value is selected from the plurality of values based on a magnitude relationship with a predetermined value. 9. The digital signal according to claim 1, wherein said information embedding means determines a transform coefficient whose value is changed when embedding the digital watermark information, based on the generated values of the plurality of transform coefficients. Processing equipment. 10. The information embedding means converts the digital watermark information used for recording control when the orthogonally inversely transformed digital signal is recorded on a recording medium, into at least one of the plurality of transform coefficients. 3. The digital signal processing device according to claim 2, wherein the embedding is performed by changing the value of. 11. The information embedding means embeds the digital watermark information by changing values of a plurality of transform coefficients indicating adjacent frequency components in a predetermined pattern among the plurality of generated transform coefficients. Item 2. The digital signal processing device according to item 1. 12. The digital signal processing apparatus according to claim 1, wherein said information embedding means embeds digital watermark information by setting a value of at least one of the plurality of transform coefficients to 0. 13. The digital signal processing device according to claim 1, wherein said orthogonal transform means converts at least one of a digital audio signal and a digital image signal as the digital signal from a time domain to a frequency domain. 14. A trigger signal detecting means for detecting a trigger signal included in the input digital signal based on an amplitude level of the input digital signal, and a frequency analysis of the input digital signal to convert a time domain to a frequency domain. And orthogonal transform means for generating a plurality of transform coefficients each indicating the level of a corresponding frequency component among the levels of the plurality of frequency components included in the digital signal; and, after the trigger signal is detected, An information embedding unit that embeds digital watermark information by changing a value of at least one of the plurality of transformed coefficients. 15. An orthogonal inverse transform for orthogonally transforming a digital signal represented in a frequency domain by a transform coefficient whose value is changed and a transform coefficient of the plurality of transform coefficients whose value is not changed into a time domain. 15. The method of claim 14, further comprising:
3. The digital signal processing device according to claim 1. 16. The information embedding means, when the digital signal after the orthogonal inverse transform has an amplitude level other than the trigger signal that is equal to or greater than the amplitude level of the trigger signal, the information embedding means converts the amplitude level to the trigger signal. The digital signal processing device according to claim 15, wherein the amplitude level is lower than the amplitude level of the digital signal. 17. The information embedding means according to claim 1, wherein said digital watermark information used for recording control when recording said orthogonally inversely converted digital signal on a recording medium is at least one of said plurality of transform coefficients. The digital signal processing device according to claim 15, wherein the embedded value is changed. 18. A digital signal is subjected to frequency analysis to convert from a time domain to a frequency domain, and to generate a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency components included in the digital signal. A digital signal processing apparatus, comprising: an orthogonal transformation unit that performs a transformation; and an information detection unit that detects whether or not digital watermark information is embedded based on the generated change pattern of the plurality of transformation coefficients. 19. The information detecting means as to whether digital watermark information used for controlling the recording of the digital signal on a recording medium is embedded based on the generated change pattern of the plurality of transform coefficients. The digital signal processing device according to claim 18, wherein the digital signal processing device detects whether or not the digital signal is present. 20. The digital signal according to claim 19, further comprising: an output unit that outputs the digital signal; and a control unit that controls whether the output unit can output the digital signal based on a result of the detection. Processing equipment. 21. A digital signal is frequency-analyzed in units of a conversion processing block composed of a predetermined amount of digital signals and converted from a time domain to a frequency domain, and a corresponding one of a plurality of frequency component levels included in the digital signal is provided. Orthogonal transform means for generating a plurality of transform coefficients each indicating a level of a frequency component; and a plurality of transform processing blocks successively processed by the orthogonal transform means. And an information detecting means for detecting whether or not the digital watermark information is embedded based on a result of the detection. 22. The information detecting means, for the predetermined transform coefficient, processes the value of the transform coefficient obtained from one transform processing block by the orthogonal transform means before and after the one transform processing block. Determining whether digital watermark information is embedded in the conversion coefficient of the one conversion processing block based on a value obtained by dividing by a sum of the values of the conversion coefficients obtained from the two conversion processing blocks. Item 2
2. The digital signal processing device according to 1. 23. An input unit for inputting the digital signal, and a trigger signal detecting unit for detecting a trigger signal included in the input digital signal based on an amplitude level of the input digital signal, 22. The digital signal processing device according to claim 21, wherein the orthogonal transform unit transforms the input digital signal from a time domain to a frequency domain after the trigger signal is detected. 24. A digital signal is frequency-analyzed and converted from a time domain to a frequency domain, and a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency components included in the digital signal are generated. Orthogonal transform means, and detects a difference between values of a plurality of transform coefficients respectively indicating levels of a plurality of frequency components continuously located on a frequency among the plurality of generated transform coefficients, based on the difference. A digital signal processing device having information detection means for detecting whether or not the digital watermark information is embedded. 25. A digital signal is frequency-analyzed and converted from a time domain to a frequency domain, and a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. Orthogonal transformation means, information detection means for detecting whether or not digital watermark information is embedded based on the generated change pattern of the plurality of transformation coefficients, and output means for outputting in accordance with the digital signal. And a control means for controlling whether the output means outputs the digital signal based on a result of the detection. 26. A digital signal is frequency-analyzed and converted from a time domain to a frequency domain, and a plurality of transform coefficients each representing a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. Orthogonal transforming means, information detecting means for detecting whether digital watermark information is embedded based on the generated change pattern of the plurality of transform coefficients, and recording means for recording the digital signal on a recording medium. And a control unit for controlling whether or not the digital signal can be recorded on the recording medium by the recording unit based on a result of the detection. 27. A digital signal is frequency-analyzed and converted from a time domain to a frequency domain, and a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. And a digital signal processing method for embedding digital watermark information by changing a value of at least one of the plurality of transform coefficients. 28. An orthogonal inverse transform of a digital signal represented in a frequency domain by a transform coefficient whose value has been changed and a transform coefficient whose value has not been changed among the plurality of transform coefficients in a time domain. 3. The digital signal processing method according to 1. 29. The digital signal processing method according to claim 28, wherein the digital watermark information is information used when controlling recording of the orthogonally inverted digital signal on a recording medium. 30. A trigger signal included in the input digital signal is detected based on an amplitude level of the input digital signal, and the input digital signal is frequency-analyzed and converted from a time domain to a frequency domain. Generating a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal, and detecting at least one of the plurality of transform coefficients after detecting the trigger signal. Digital signal processing method for embedding digital watermark information by changing the value of 31. A digital signal is frequency-analyzed and converted from a time domain to a frequency domain, and a plurality of transform coefficients each representing a level of a corresponding frequency component among a plurality of frequency component levels included in the digital signal are generated. And a digital signal processing method for detecting whether or not digital watermark information is embedded based on the generated change patterns of the plurality of transform coefficients. 32. The digital signal processing method according to claim 31, wherein the digital watermark information is information used when controlling recording of the digital signal on a recording medium. 33. A digital signal is frequency-analyzed in units of a conversion processing block composed of a predetermined amount of digital signals and converted from a time domain to a frequency domain, and a corresponding one of a plurality of frequency component levels included in the digital signal is provided. A plurality of transform coefficients each indicating a level of a frequency component are generated, and a difference of a predetermined transform coefficient among the generated transform coefficients between the plurality of transform processing blocks that have been continuously converted is determined. A digital signal processing method for detecting and detecting whether or not the digital watermark information is embedded based on a result of the detection. 34. A digital signal is frequency-analyzed and converted from a time domain to a frequency domain, and a plurality of transform coefficients each indicating a level of a corresponding frequency component among a plurality of frequency components included in the digital signal are generated. And detecting a difference between values of a plurality of transform coefficients each indicating a level of a plurality of frequency components successively located on a frequency among the generated plurality of transform coefficients, and determining the digital watermark information based on the difference. A digital signal processing method for detecting whether or not embedded.