CN106409302B - Audio-frequency water mark method and system based on insertion regional choice - Google Patents
Audio-frequency water mark method and system based on insertion regional choice Download PDFInfo
- Publication number
- CN106409302B CN106409302B CN201610458412.9A CN201610458412A CN106409302B CN 106409302 B CN106409302 B CN 106409302B CN 201610458412 A CN201610458412 A CN 201610458412A CN 106409302 B CN106409302 B CN 106409302B
- Authority
- CN
- China
- Prior art keywords
- frequency
- signal
- watermark
- frame
- audio
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000003780 insertion Methods 0.000 title claims abstract description 113
- 230000037431 insertion Effects 0.000 title claims abstract description 113
- 238000000034 method Methods 0.000 title claims abstract description 28
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 13
- 238000001514 detection method Methods 0.000 claims abstract description 45
- 230000001052 transient effect Effects 0.000 claims abstract description 20
- 230000008569 process Effects 0.000 claims abstract description 11
- 229910002056 binary alloy Inorganic materials 0.000 claims abstract description 10
- 238000009432 framing Methods 0.000 claims description 11
- 230000001427 coherent effect Effects 0.000 claims description 6
- 230000008825 perceptual sensitivity Effects 0.000 claims description 4
- 238000001914 filtration Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 13
- 230000005236 sound signal Effects 0.000 description 13
- 230000008447 perception Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 230000001154 acute effect Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/018—Audio watermarking, i.e. embedding inaudible data in the audio signal
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Editing Of Facsimile Originals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The present invention provides a kind of audio-frequency water mark method and system based on insertion regional choice, and telescopiny includes reading audio file, is first made whether as the judgement in insertion region, can then carry out the selection of the insertion frequency band of audio frequency watermark to every frame signal；Discrete Fourier Transform is carried out, the pseudorandom frequency expansion sequence of binary system is generated, carries out the insertion of watermark, transform to time domain；Detection process includes reading audio file to be detected, being made whether to every frame signal can be as the judgement in insertion region, calculate the starting point and frequency domain end point of detection range, it carries out Discrete Fourier Transform and generates the pseudorandom frequency expansion sequence of binary system, the sufficient statistic for calculating detection obtains the watermark bit detected.The invention proposes, than filtering out the accuracy rate that transient signal promotes watermark detection, promote the robustness of watermark by embedding a watermark in the significant frequency range of auditory perceptual by ceiling capacity in frame and least energy.
Description
Technical field
The present invention relates to Digital Audio Watermarking Techniques fields, more particularly to the audio-frequency water mark method based on insertion regional choice
And system.
Background technique
Digital audio frequency watermark is that certain digital informations are added into audio signal to reach the identification of the file true and false, copyright guarantor
The signal processing operations of the purpose of shield, Information hiding.The selection technique in audio frequency watermark insertion region, which refers to, is embedded into sound in watermark
Before frequency signal, appropriate audio region is selected to be embedded in watermark.Conventional audio digital watermark does not account for the spy of audio signal
Sign, the insertion of watermark is all carried out to entire audio file, after will lead to the low region insertion watermark of 1) audio frequency signal amplitude in this way,
Amplitude has exceeded masking threshold and generates noise, destroys the perception transparency；2) for occurring changing violent wink in audio signal
The variance of state signal, the audio signal in the region is very big, causes watermark bit error rate when detecting watermark very high after being embedded in watermark；3)
It is embedded in watermark in frequency domain, if the inapparent region of selection auditory perceptual is embedded in watermark, is damaged by signal processing or audio
After compression, watermark will be lost a part, cause the watermark detection bit error rate high.
Summary of the invention
The object of the present invention is to provide the audio watermarking techniques of selection region insertion, and watermark is enable to be embedded into suitable sound
In frequency domain, avoids the occurrence of unnecessary noise and reduce the generation of error code.
In order to achieve the above objectives, technical solution provided by the invention provides a kind of audio frequency watermark based on insertion regional choice
Method, including telescopiny and detection process,
The telescopiny includes the following steps,
Step A1 reads audio file, obtains the signal x of n-th frame time-domain audio after sample rate f s1 and framingn, frame length is
N,
First to every frame signal xnBe made whether can as insertion region judgement,
Then being directed to can be as each frame signal x in insertion regionn, the selection of the insertion frequency band of audio frequency watermark is carried out,
If the start frequency according to the preset insertion of frequency-portions of auditory perceptual sensitivity is FWMIN, end frequency is FWMAX, a frame
Start be embedded in point freqmin1 and insertion end point freqmax1 seek it is as follows,
Freqmin1=floor ((FWMIN × 2.0/fs1) × N)
Freqmax1=floor ((FWMAX × 2.0/fs1) × N)
Wherein, floor is downward bracket function；
Step A2, to each frame signal x that can be embedded in watermarkn, carry out Discrete Fourier Transform and obtain frequency domain signal Xn；
Step A3, using key key as random number seed, generate that length is freqmax1-freqmin1+1 two into
Make pseudorandom frequency expansion sequence u；
Step A4, according to frequency expansion sequence u, frequency domain signal XnWith watermark bit b, the insertion of watermark is carried out, obtains insertion water
Frequency-region signal after print, calculating is as follows,
|X′n|=| Xn|+bαu
Wherein, α is constant, controls the embedment strength of watermark, | Xn| and | X 'n| the frequency domain width before respectively indicating insertion watermark
Frequency domain amplitude after value and insertion watermark, then obtains the frequency-region signal after insertion watermark by Euler's formula
Wherein, ∠ XnIndicate the phase of frequency-region signal, X 'nFrequency-region signal after indicating insertion watermark, e are that mathematics refers to naturally
Number；
Step A5, by be embedded in watermark after frequency domain signal X 'nTime domain is transformed to, the audio file of insertion watermark is generated；
The detection process includes the following steps,
Step B1 reads audio file to be detected, the n-th frame signal z after obtained time domain framingnWith sample rate f s2,
First to every frame signal xnBeing made whether can be as the judgement in insertion region；
For each frame signal x that can be used as insertion regionn, as signal to be detected, calculate the starting of detection range
Point freqmin2 and frequency domain end point freqmax2
Freqmin2=floor ((FWMIN × 2.0/fs2) × N)
Freqmax2=floor ((FWMAX × 2.0/fs2) × N)
Step B2 carries out Discrete Fourier Transform and obtains the frequency-region signal Z of signal to be detectedn, corresponding frequency domain range value note
For | Zn|；
Step B3, using key key as random number seed, generate that length is freqmax2-freqmin2+1 two into
Make pseudorandom frequency expansion sequence u；
Step B4, according to the frequency domain range value of frequency expansion sequence u and signal to be detected | Zn|, calculate the abundant statistics of detection
Measure rnIt is as follows,
If sufficient statistic rn>=0, then the watermark bit detected is b=1；Otherwise, the watermark bit detected
For b=0.
Moreover, in step A1 and step B1, to every frame signal xnIt is made whether to realize as the judgement in insertion region
Mode is as follows,
1) signal xnAverage energySize exceed preset respective threshold τ1, it is not allow embedding for mute area then
Enter watermark；
If 2) signal xnInterior includes transient signal, then does not allow to be embedded in watermark.
Moreover, signal xnWhether include inside transient signal, be judged by the following manner,
If a frame signal is decomposed into S block, the energy of S block is calculated separately out, compares the block and minimum of ceiling capacity
The energy ratio rate of energy block and preset respective threshold τ2If rate is greater than τ2Then think that the frame signal includes transient signal.
The present invention correspondingly provides a kind of audio frequency watermark system based on insertion regional choice, including audio frequency watermark insertion
System and watermark detection subsystem,
The audio frequency watermark insertion subsystem comprises the following modules,
Selection appropriate area insertion module obtains n-th frame time domain after sample rate f s1 and framing for reading audio file
The signal x of audion, frame length N,
First to every frame signal xnBe made whether can as insertion region judgement,
Then being directed to can be as each frame signal x in insertion regionn, the selection of the insertion frequency band of audio frequency watermark is carried out,
The selection of the insertion frequency band of audio frequency watermark is carried out, if according to the beginning of the preset insertion of frequency-portions of auditory perceptual sensitivity frequency
It is FWMAX that rate, which is FWMIN, terminates frequency, a frame start to be embedded in point freqmin1 and insertion end point freqmax1 seek as
Under,
Freqmin1=floor ((FWMIN × 2.0/fs1) × N)
Freqmax1=floor ((FWMAX × 2.0/fs1) × N)
Wherein, floor is downward bracket function；
First time-frequency convert module, for each frame signal x that can be embedded in watermarkn, carry out Discrete Fourier Transform and obtain
To frequency domain signal Xn；
First frequency expansion sequence generation module, for using key key as random number seed, generating length to be
The pseudorandom frequency expansion sequence u of the binary system of freqmax1-freqmin1+1；
Watermark embedding module, for according to frequency expansion sequence u, frequency domain signal XnWith watermark bit b, the insertion of watermark is carried out,
Frequency-region signal after obtaining insertion watermark, calculating is as follows,
|X′n|=| Xn|+bαu
Wherein, α is constant, controls the embedment strength of watermark, | Xn| and | X 'n| the frequency domain width before respectively indicating insertion watermark
Frequency domain amplitude after value and insertion watermark, then obtains the frequency-region signal after insertion watermark by Euler's formula
Wherein, ∠ XnIndicate the phase of frequency-region signal, X 'nFrequency-region signal after indicating insertion watermark, e are that mathematics refers to naturally
Number；
Time-frequency inverse transform module, for will be embedded in the frequency domain signal X after watermark 'nTime domain is transformed to, insertion watermark is generated
Audio file；
The watermark detection subsystem comprises the following modules,
Select appropriate area detection module, for reading audio file to be detected, the n-th frame after obtained time domain framing
Signal znWith sample rate f s2,
First to every frame signal xnBeing made whether can be as the judgement in insertion region；
For each frame signal x that can be used as insertion regionn, as signal to be detected, calculate the starting of detection range
Point freqmin2 and frequency domain end point freqmax2
Freqmin2=floor ((FWMIN × 2.0/fs2) × N)
Freqmax2=floor ((FWMAX × 2.0/fs2) × N)
Second time-frequency convert module obtains the frequency-region signal Z of signal to be detected for carrying out Discrete Fourier Transformn, phase
Frequency domain range value is answered to be denoted as | Zn|；
Second frequency expansion sequence generation module, for using key key as random number seed, generating length to be
The pseudorandom frequency expansion sequence u of the binary system of freqmax2-freqmin2+1；
Coherent detection module, for the frequency domain range value according to frequency expansion sequence u and signal to be detected | Zn|, calculate detection
Sufficient statistic rnIt is as follows,
If sufficient statistic rn>=0, then the watermark bit detected is b=1；Otherwise, the watermark bit detected
For b=0.
Moreover, in selection appropriate area insertion module and selection appropriate area detection module, to every frame signal xnIt is
It is no can as insertion region judgement, implementation is as follows,
1) signal xnAverage energySize exceed preset respective threshold τ1, it is not allow embedding for mute area then
Enter watermark；
If 2) signal xnInterior includes transient signal, then does not allow to be embedded in watermark.
Moreover, signal xnWhether include inside transient signal, be judged by the following manner,
If a frame signal is decomposed into S block, the energy of S block is calculated separately out, compares the block and minimum of ceiling capacity
The energy ratio rate of energy block and preset respective threshold τ2If rate is greater than τ2Then think that the frame signal includes transient signal.
The invention proposes by ceiling capacity in frame and least energy than promoting watermark detection to filter out transient signal
Accuracy rate promotes the robustness of watermark by embedding a watermark in the significant frequency range of auditory perceptual, further, proposes to utilize
Average energy promotes the perception transparency to filter out No Tooting Area.Technical solution of the present invention has important market value.
Detailed description of the invention
Fig. 1 is the insertion subsystem structure block diagram of the embodiment of the present invention.
Fig. 2 is the detection subsystem structure block diagram of the embodiment of the present invention.
Fig. 3 is the telescopiny flow chart of the embodiment of the present invention
Fig. 4 is the detection process flow chart of the embodiment of the present invention.
Specific embodiment
Technical solution of the present invention is described further with specific embodiment combination attached drawing below.
The embodiment of the present invention provides a kind of audio frequency watermark system based on insertion regional choice, including audio frequency watermark insertion
System and watermark detection subsystem.
Referring to Fig. 1, the audio watermarking technique of insertion regional choice provided in an embodiment of the present invention is embedded in subsystem, including choosing
Select appropriate area insertion module 1, the first time-frequency convert module 2, the first frequency expansion sequence generation module 3, watermark embedding module 4 and when
Frequency inverse transform module 5 can realize each module using software firming bechnology when specific implementation.
The selection appropriate area is embedded in module 1, judges the time-domain audio signal frame of reading, and when specific implementation can
To judge whether to can satisfy the condition of insertion watermark frame by frame: being unsatisfactory for just skipping this frame, continue the judgement of next frame；If full
Foot just exports signal to the first time-frequency conversion module 2, more according to the sample rate of the time-domain audio signal read and human ear
Sensitive frequency range calculates the range of this frequency-region signal insertion watermark, and the frequency-region signal in embeddable range is exported water supply
Print insertion module 4, the maximum value of the insertion range and minimum value are exported to the first frequency expansion sequence generation module 3；
The first time-frequency convert module 2, for the time-domain audio signal read to be converted to frequency-region signal, export to
Watermark embedding module 4；
The first frequency expansion sequence generation module 3, for defeated according to random number seed and selection appropriate area insertion module 1
It is 1 or -1 equally distributed stochastic ordering that the maximum value and minimum value of the insertion range entered, which generate and be embedded in amplitude of the range with length,
Column, and this random sequence is exported to watermark embedding module 4；
The watermark embedding module 4 generates the audio with watermark information of frequency domain for the amplitude spectrum in frequency-region signal
Signal is exported to time-frequency inverse transform module 5；
The time-frequency inverse transform module 5, the audio with watermark information of the frequency domain for embedding a watermark into the input of module 4
Signal is converted to the audio signal with watermark information of time domain, and the audio signal with watermark information of this time domain is generated
Audio file just obtains the audio file with watermark information.
Referring to fig. 2, watermark detection subsystem provided in an embodiment of the present invention, including selection appropriate area detection module 6, the
Two time-frequency convert modules 7, the second frequency expansion sequence generation module 8, coherent detection module 9, can be solid using software when specific implementation
Change technology realizes each module.
The selection appropriate area detection module 6 and the function of selecting appropriate area to be embedded in module 1 are essentially identical, are unsatisfactory for
Watermark is embedded in the region of condition, does not also contain watermark generally, can not have to when detection consider: can sentence frame by frame when specific implementation
It is disconnected, for being unsatisfactory for the frame of testing conditions, skips and do not detect, continue the judgement of next frame；Meet the audio signal of testing conditions
It exports to the second time-frequency conversion module 7, equally exports the maxima and minima in frequency detecting region to the second time-frequency convert
Module 7 and the second frequency expansion sequence generation module 8；
The second time-frequency convert module 7, for the time-domain audio signal read to be converted to frequency-region signal, export to
Coherent detection module 9；
The function of the second frequency expansion sequence generation module 8 and the first frequency expansion sequence generation module 3 is essentially identical, will generate
Result export to coherent detection module 9；
The coherent detection module 9, for the frequency domain amplitude signal and spread spectrum to be detected according to detection range to input
The frequency expansion sequence that sequence generating module 9 inputs calculates correlation according to the symbol of correlation and judges watermark.
Each module specific implementation is referring to method corresponding steps, and it will not go into details by the present invention.It is provided in an embodiment of the present invention to be based on
It is embedded in the audio-frequency water mark method of regional choice, including telescopiny and detection process.
Referring to Fig. 3, the audio frequency watermark telescopiny provided in an embodiment of the present invention based on selection region can be using calculating
Machine software technology means carry out process automatically, specifically includes the following steps:
Step A1, reads audio file, the audio signal x elder generation framing to time domain, and the after obtaining sample rate f s1 and framing
N frame time-domain audio signal xn(frame length N), to every frame signal xnBeing made whether can be as the judgement in insertion region, and judgement include
Both sides judgement:
1) judge xnThe size of average energy whether exceed the threshold value of setting, to judge present frame xnIt whether is mute area,
Insertion watermark is not allowed for if it is mute area, is not otherwise just mute area beyond threshold value, insertion can be can be carried out.By following
The average energy of formula calculating n-th frame
Wherein, N is frame length, i.e. sample points in a frame；I is the sample point index number in a frame, and value arrives N- 0
Between 1；xn 2(i) n-th frame time-domain signal x is indicatednI-th point of the energy in frame；τ1It is specific real for the decision threshold of average energy
Those skilled in the art can voluntarily preset value when applying, such as be empirically derived；If exceeding threshold value, meet condition 1),
Carry out following condition 2) judgement.
2) for there is the case where transient signal in a frame, due to its frequency acute variation, the biggish variance that will cause,
The error probability of watermark detection caused by signal variance is bigger when detecting is higher, and such case should not also be embedded in watermark.It is logical
It crosses and a frame is decomposed into S block, calculate separately out the energy of S block, pass through the block of ceiling capacity and the energy of least energy block
Than rate and threshold tau2Comparison, rate be greater than τ2Then it is considered that the frame signal includes that transient signal is not embedded in watermark, otherwise may be used
To be embedded in watermark.When it is implemented, those skilled in the art can voluntarily preset the value of S.
Specific implementation is as follows:
First by a frame signal xnIt is divided into S block, then the sample points M in each sub-block is
M=N/S (2)
Each piece of ENERGY EiIt calculates as follows
Wherein, i indicates the index number of intra block, and j indicates the index number of sample point in frame, xn 2(j) when indicating n-th frame
Domain signal xnThe energy of jth point in frame.
Find out the ceiling capacity E in block energyMaxWith least energy EMin
EMax=MAX { Ei, EMin=MIN { Ei, i ∈ [0, S-1] (4)
Wherein, MAX, MIN respectively indicate maximizing function and minimum value function.
The ratio rate of ceiling capacity and least energy calculates as follows:
If rate > τ2, it is considered as signal frame xnInside there is transient signal, which is not embedded in watermark；Otherwise, it can be embedded in
Watermark.Wherein τ2For threshold value, those skilled in the art can voluntarily preset value, such as τ when specific implementation2For the inspection of transient signal
Threshold value is surveyed, is empirically derived.
Then being directed to can be as each frame signal x in insertion regionn, for audio frequency watermark insertion frequency band selection,
Should be the more significant region of auditory perceptual, those skilled in the art can voluntarily preset according to auditory perceptual characteristic, such as
1000-7000Hz.Because the signal in these regions will not be removed after the attack such as filtering, audio compression.So by water
Print is embedded into the apparent region of perception, will not be erased after being subjected to some signal attacks, be able to detect out.If setting according to people
The start frequency of the sensitive preset insertion of frequency-portions of ear perception is FWMIN, end frequency is FWMAX, and a corresponding frame is opened
Begin insertion point freqmin1 and be embedded in end point freqmax1 seek it is as follows,
Freqmin1=floor ((FWMIN × 2.0/fs1) × N) (6)
Freqmax1=floor ((FWMAX × 2.0/fs1) × N) (7)
Wherein, floor is downward bracket function.
According to insertion point freqmin1 and insertion end point freqmax1 is started, the frequency domain audio letter within the scope of this is chosen
Number.
It can judge frame by frame when specific implementation, be unsatisfactory for skipping for condition, carry out the judgement of next frame.
Step A2, to the signal frame x that can be embedded in watermarkn, carrying out FFT transform (fast discrete Fourier transformation) is frequency domain
Signal Xn。
Step A3, using key key as random number seed, generate that length is freqmax1-freqmin1+1 two into
Pseudorandom spreading sequence u processed.
Detailed process is as follows for embodiment in MATLAB:
Firstly, calling RandStream function (random seed function) to rand function, (random number is raw using key key
At function) initialized, then call rand function generate random number, due to rand function generate random number be 0~1 it
Between number, also need these numbers round up become 0 and 1 binary pseudo-random sequence, then by this unipolar puppet
Random sequence switchs to the pseudo-random sequence u that bipolarity contains only+1 He -1.
Step A4, according to frequency expansion sequence u, frequency domain signal XnWith watermark bit b, watermark is carried out using following formula (8)
Insertion, obtain insertion watermark after frequency-region signal, calculate realize it is as follows
|X′n|=| Xn|+bαu (8)
Wherein, α is constant, controls the embedment strength of watermark, and those skilled in the art can preset value when specific implementation；|
Xn| and | X 'n| then the frequency domain amplitude after frequency domain amplitude and insertion watermark before respectively indicating insertion watermark passes through Euler's formula
Frequency-region signal after obtaining insertion watermark.
Wherein, ∠ XnIndicate the phase of frequency-region signal, X 'nFrequency-region signal after indicating insertion watermark, e are that mathematics refers to naturally
Number.
Step A5, by be embedded in watermark after frequency domain signal X 'nTime domain is transformed to, it is embedding to get arriving to ultimately produce audio file
Enter the audio file of watermark.
Referring to fig. 4, the audio frequency watermark detection process provided in an embodiment of the present invention based on selection region insertion, can use
Computer software technology means carry out process automatically, specifically includes the following steps:
Step B1 reads audio file to be detected, the n-th frame signal z after obtained time domain framingnWith sample rate f s2,
To each time-domain signal znTake steps the same decision method in A1,
Consider two following conditions,
1) signal xnAverage energySize exceed preset respective threshold τ1, it is not allow embedding for mute area then
Enter watermark；
If 2) signal xnInterior includes transient signal, then does not allow to be embedded in watermark.
Then it can not be embedded in watermark for mute area and not comprising the frame signal of transient signal and have to be detected.
It can judge frame by frame when specific implementation, be unsatisfactory for skipping for condition, carry out the judgement of next frame.
For each frame signal x that can be used as insertion regionn, as signal to be detected, calculate the frequency domain of detection range
Starting point freqmin2 and frequency domain end point freqmax2
Freqmin2=floor ((FWMIN × 2.0/fs2) × N) (10)
Freqmax2=floor ((FWMAX × 2.0/fs2) × N) (11)
Step B2, for meeting the signal z of testing conditionsn, carry out Discrete Fourier Transform and obtain the frequency of signal to be detected
Domain signal Zn, corresponding frequency domain range value is denoted as | Zn|。
Step B3 generates binary system frequency expansion sequence u (the u mode phase obtained with embedding grammar above using key key
Together), i.e., it using key key as random number seed, generates the binary system pseudorandom that length is freqmax2-freqmin2+1 and expands
Frequency sequence u.
Step B4, according to the frequency domain range value of frequency expansion sequence u and signal to be detected | Zn|, by calculate frequency expansion sequence u and
The frequency domain range value of signal to be detected | Zn| correlation, calculate the sufficient statistic r of detectionn
Wherein,<>indicates that the inner product of signal calculates.
If sufficient statistic rn>=0, then the watermark bit detected is b=1；Otherwise, the watermark bit detected
For b=0.
It is described in the present invention that specific embodiments are merely illustrative of the spirit of the present invention.Technology belonging to the present invention
The technical staff in field can make various modifications or additions to the described embodiments or by a similar method
Substitution, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Claims (6)
1. a kind of audio-frequency water mark method based on insertion regional choice, it is characterised in that: including telescopiny and detection process, institute
Telescopiny is stated to include the following steps,
Step A1 reads audio file, obtains the signal x of n-th frame time-domain audio after sample rate f s1 and framingn, frame length N, first
To every frame signal xnBe made whether can as insertion region judgement,
Then being directed to can be as each frame signal x in insertion regionn, the selection of the insertion frequency band of audio frequency watermark is carried out, if according to
The start frequency of the preset insertion of frequency-portions of auditory perceptual sensitivity is FWMIN, end frequency is FWMAX, the beginning of a frame
Insertion point freqmin1 and be embedded in end point freqmax1 seek it is as follows,
Freqmin1=floor ((FWMIN × 2.0/fs1) × N)
Freqmax1=floor ((FWMAX × 2.0/fs1) × N)
Wherein, floor is downward bracket function；
Step A2, to each frame signal x that can be embedded in watermarkn, carry out Discrete Fourier Transform and obtain frequency domain signal Xn；
It is pseudo- to generate the binary system that length is freqmax1-freqmin1+1 using key key as random number seed by step A3
Random frequency expansion sequence u；
Step A4, according to frequency expansion sequence u, frequency domain signal XnWith watermark bit b, the insertion of watermark is carried out, after obtaining insertion watermark
Frequency-region signal, calculating is as follows,
|X′n|=| Xn|+bαu
Wherein, α is constant, controls the embedment strength of watermark, | Xn| and | X 'n| respectively indicate insertion watermark before frequency domain amplitude and
Then frequency domain amplitude after being embedded in watermark obtains the frequency-region signal after insertion watermark by Euler's formula
Wherein, ∠ XnIndicate the phase of frequency-region signal, X 'nFrequency-region signal after indicating insertion watermark, e are mathematics natural Exponents, on
Mark j is imaginary unit；
Step A5, by be embedded in watermark after frequency domain signal X 'nTime domain is transformed to, the audio file of insertion watermark is generated；
The detection process includes the following steps,
Step B1 reads audio file to be detected, the n-th frame signal z after obtained time domain framingnIt is first right with sample rate f s2
Every frame signal xnBeing made whether can be as the judgement in insertion region；
For each frame signal x that can be used as insertion regionn, as signal to be detected, calculate the starting point of detection range
Freqmin2 and frequency domain end point freqmax2
Freqmin2=floor ((FWMIN × 2.0/fs2) × N)
Freqmax2=floor ((FWMAX × 2.0/fs2) × N)
Step B2 carries out Discrete Fourier Transform and obtains the frequency-region signal Z of signal to be detectedn, corresponding frequency domain range value is denoted as | Zn
|；
It is pseudo- to generate the binary system that length is freqmax2-freqmin2+1 using key key as random number seed by step B3
Random frequency expansion sequence u；
Step B4, according to the frequency domain range value of frequency expansion sequence u and signal to be detected | Zn|, calculate the sufficient statistic r of detectionn
It is as follows,
If sufficient statistic rn>=0, then the watermark bit detected is b=1；Otherwise, the watermark bit detected is b=
0。
2. the audio-frequency water mark method according to claim 1 based on insertion regional choice, it is characterised in that: step A1 and step
In B1, to every frame signal xnBeing made whether can be as the judgement in insertion region, and implementation is as follows,
1) signal xnAverage energySize exceed preset respective threshold τ1, it is then not allow to be embedded in water for mute area
Print；
The average energy of n-th frame is calculated by following formula
Wherein, N is frame length, and i is the sample point index number in a frame, xn 2(i) n-th frame time-domain signal x is indicatednI-th in frame
The energy of point；τ1For the decision threshold of average energy；
If 2) signal xnInterior includes transient signal, then does not allow to be embedded in watermark.
3. the audio-frequency water mark method according to claim 2 based on insertion regional choice, it is characterised in that: signal xnInside whether
Comprising transient signal, it is judged by the following manner,
If a frame signal is decomposed into S block, the energy of S block is calculated separately out, compares the block and least energy of ceiling capacity
The energy ratio rate of block and preset respective threshold τ2If rate is greater than τ2Then think that the frame signal includes transient signal.
4. it is a kind of based on insertion regional choice audio frequency watermark system, it is characterised in that: including audio frequency watermark insertion subsystem and
Watermark detection subsystem,
The audio frequency watermark insertion subsystem comprises the following modules,
Selection appropriate area insertion module obtains n-th frame time-domain audio after sample rate f s1 and framing for reading audio file
Signal xn, frame length N,
First to every frame signal xnBe made whether can as insertion region judgement,
Then being directed to can be as each frame signal x in insertion regionn, the selection of the insertion frequency band of audio frequency watermark is carried out, if according to
The start frequency of the preset insertion of frequency-portions of auditory perceptual sensitivity is FWMIN, end frequency is FWMAX, the beginning of a frame
Insertion point freqmin1 and be embedded in end point freqmax1 seek it is as follows,
Freqmin1=floor ((FWMIN × 2.0/fs1) × N)
Freqmax1=floor ((FWMAX × 2.0/fs1) × N)
Wherein, floor is downward bracket function；
First time-frequency convert module, for each frame signal x that can be embedded in watermarkn, carry out Discrete Fourier Transform and obtain frequency domain
Signal Xn；
First frequency expansion sequence generation module, for using key key as random number seed, generation length to be freqmax1-
The pseudorandom frequency expansion sequence u of the binary system of freqmin1+1；
Watermark embedding module, for according to frequency expansion sequence u, frequency domain signal XnWith watermark bit b, the insertion of watermark is carried out, is obtained embedding
Frequency-region signal after entering watermark, calculating is as follows,
|X′n|=| Xn|+bαu
Wherein, α is constant, controls the embedment strength of watermark, | Xn| and | X 'n| respectively indicate insertion watermark before frequency domain amplitude and
Then frequency domain amplitude after being embedded in watermark obtains the frequency-region signal after insertion watermark by Euler's formula
Wherein, ∠ XnIndicate the phase of frequency-region signal, X 'nFrequency-region signal after indicating insertion watermark, e are mathematics natural Exponents, on
Mark j is imaginary unit；
Time-frequency inverse transform module, for will be embedded in the frequency domain signal X after watermark 'nTime domain is transformed to, the audio of insertion watermark is generated
File；
The watermark detection subsystem comprises the following modules,
Select appropriate area detection module, for reading audio file to be detected, the n-th frame signal after obtained time domain framing
znWith sample rate f s2,
First to every frame signal xnBeing made whether can be as the judgement in insertion region；
For each frame signal x that can be used as insertion regionn, as signal to be detected, calculate the starting point of detection range
Freqmin2 and frequency domain end point freqmax2
Freqmin2=floor ((FWMIN × 2.0/fs2) × N)
Freqmax2=floor ((FWMAX × 2.0/fs2) × N)
Second time-frequency convert module obtains the frequency-region signal Z of signal to be detected for carrying out Discrete Fourier Transformn, corresponding frequency domain
Range value is denoted as | Zn|；
Second frequency expansion sequence generation module, for using key key as random number seed, generation length to be freqmax2-
The pseudorandom frequency expansion sequence u of the binary system of freqmin2+1；
Coherent detection module, for the frequency domain range value according to frequency expansion sequence u and signal to be detected | Zn|, calculate filling for detection
Divide statistic rnIt is as follows,
If sufficient statistic rn>=0, then the watermark bit detected is b=1；Otherwise, the watermark bit detected is b=
0。
5. the audio frequency watermark system according to claim 4 based on insertion regional choice, it is characterised in that: selection appropriate area
It is embedded in module and selection appropriate area detection module, to every frame signal xnBe made whether can as insertion region judgement,
Implementation is as follows,
1) signal xnAverage energySize exceed preset respective threshold τ1, it is then not allow to be embedded in water for mute area
Print；
The average energy of n-th frame is calculated by following formula
Wherein, N is frame length, and i is the sample point index number in a frame, xn 2(i) n-th frame time-domain signal x is indicatednI-th in frame
The energy of point；τ1For the decision threshold of average energy；
If 2) signal xnInterior includes transient signal, then does not allow to be embedded in watermark.
6. the audio frequency watermark system according to claim 5 based on insertion regional choice, it is characterised in that: signal xnInside whether
Comprising transient signal, it is judged by the following manner,
If a frame signal is decomposed into S block, the energy of S block is calculated separately out, compares the block and least energy of ceiling capacity
The energy ratio rate of block and preset respective threshold τ2If rate is greater than τ2Then think that the frame signal includes transient signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610458412.9A CN106409302B (en) | 2016-06-22 | 2016-06-22 | Audio-frequency water mark method and system based on insertion regional choice |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610458412.9A CN106409302B (en) | 2016-06-22 | 2016-06-22 | Audio-frequency water mark method and system based on insertion regional choice |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106409302A CN106409302A (en) | 2017-02-15 |
| CN106409302B true CN106409302B (en) | 2019-07-09 |
Family
ID=58005751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610458412.9A Expired - Fee Related CN106409302B (en) | 2016-06-22 | 2016-06-22 | Audio-frequency water mark method and system based on insertion regional choice |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106409302B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109714284B (en) * | 2018-11-27 | 2020-06-30 | 华中科技大学 | Radio frequency watermark detection method based on K-S detection |
| CN111292756B (en) * | 2020-01-19 | 2023-05-26 | 成都潜在人工智能科技有限公司 | Compression-resistant audio silent watermark embedding and extracting method and system |
| CN113362835B (en) * | 2020-03-05 | 2024-06-07 | 杭州网易云音乐科技有限公司 | Audio watermarking method, device, electronic equipment and storage medium |
| CN111883108A (en) * | 2020-07-06 | 2020-11-03 | 珠海格力电器股份有限公司 | Password embedding method and device, password matching method and device and control system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101185121A (en) * | 2005-06-02 | 2008-05-21 | 汤姆森许可贸易公司 | Method and apparatus for watermarking an audio or video signal with watermark data using a spread spectrum |
| CN102142255A (en) * | 2010-07-08 | 2011-08-03 | 北京三信时代信息公司 | Method for embedding and extracting digital watermark in audio signal |
| CN102664013A (en) * | 2012-04-18 | 2012-09-12 | 南京邮电大学 | Audio digital watermark method of discrete cosine transform domain based on energy selection |
| CN104658542A (en) * | 2015-03-16 | 2015-05-27 | 武汉大学 | Additive spread spectrum audio watermarking embedding method, additive spread spectrum audio watermarking detection method and additive spread spectrum audio watermarking embedding system based on orthogonality |
| CN104700841A (en) * | 2015-02-10 | 2015-06-10 | 浙江省广电科技股份有限公司 | Watermark embedding and detecting method based on audio content classification |
| CN105374360A (en) * | 2015-11-25 | 2016-03-02 | 武汉大学 | Interleaved additive spread spectrum audio watermark embedding method and detection method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100611094B1 (en) * | 2000-06-15 | 2006-08-09 | 주식회사 케이티 | Apparatus and method for inserting/detecting watermark based stochastic model |
-
2016
- 2016-06-22 CN CN201610458412.9A patent/CN106409302B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101185121A (en) * | 2005-06-02 | 2008-05-21 | 汤姆森许可贸易公司 | Method and apparatus for watermarking an audio or video signal with watermark data using a spread spectrum |
| CN102142255A (en) * | 2010-07-08 | 2011-08-03 | 北京三信时代信息公司 | Method for embedding and extracting digital watermark in audio signal |
| CN102664013A (en) * | 2012-04-18 | 2012-09-12 | 南京邮电大学 | Audio digital watermark method of discrete cosine transform domain based on energy selection |
| CN104700841A (en) * | 2015-02-10 | 2015-06-10 | 浙江省广电科技股份有限公司 | Watermark embedding and detecting method based on audio content classification |
| CN104658542A (en) * | 2015-03-16 | 2015-05-27 | 武汉大学 | Additive spread spectrum audio watermarking embedding method, additive spread spectrum audio watermarking detection method and additive spread spectrum audio watermarking embedding system based on orthogonality |
| CN105374360A (en) * | 2015-11-25 | 2016-03-02 | 武汉大学 | Interleaved additive spread spectrum audio watermark embedding method and detection method and system |
Non-Patent Citations (2)
| Title |
|---|
| A Spread Spectrum watermarking Algorithm based Local Instruction Statistic;Tianwen Feng et.al;《9th International Conference on Intelligent Information Hiding and Multimedia Signal Processing》;20131016;全文 |
| Spread-Spectrum Robust Image Watermarking for Ownership Protection;Ansari，A ET AL;《22nd Iranian Conference on Electrical Engineering》;20140522;全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106409302A (en) | 2017-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106409302B (en) | Audio-frequency water mark method and system based on insertion regional choice | |
| KR102386155B1 (en) | How to protect your voice assistant from being controlled by machine learning-based silent commands | |
| Lei et al. | Robust SVD-based audio watermarking scheme with differential evolution optimization | |
| CN105976823B (en) | Adaptive audio water mark method and system based on phase code | |
| Avcibas et al. | Steganalysis of watermarking techniques using image quality metrics | |
| WO2016077760A1 (en) | Determining media device activation based on frequency response analysis | |
| CN104658542B (en) | Based on orthogonal additivity spread spectrum audio frequency watermark embedding grammar, detection method and system | |
| CN101494051A (en) | Detection method for time-domain audio LSB hidden write | |
| CN113782041B (en) | Method for embedding and positioning watermark based on audio variable frequency domain | |
| CN108682425B (en) | Robust digital audio watermark embedding system based on constant watermark | |
| CN110163787A (en) | Digital audio Robust Blind Watermarking Scheme embedding grammar based on dual-tree complex wavelet transform | |
| CN102074238A (en) | Linear interference cancellation-based speech secrete communication method | |
| Sundaram et al. | Audio scene segmentation using multiple features, models and time scales | |
| CN111613243A (en) | Voice detection method and device | |
| CN105374360B (en) | Intersect additivity spread spectrum audio frequency watermark embedding grammar, detection method and system | |
| CN101350198B (en) | Method for compressing watermark using voice based on bone conduction | |
| Zeng et al. | An algorithm of echo steganalysis based on Bayes classifier | |
| KR20070061285A (en) | Digital audio watermarking method using hybrid transform | |
| Khademi et al. | Audio watermarking based on quantization index modulation in the frequency domain | |
| Wu et al. | Adaptive audio watermarking based on SNR in localized regions | |
| Panda et al. | Application of energy efficient watermark on audio signal for authentication | |
| Youssef | HFSA-AW: a hybrid fuzzy self-adaptive audio watermarking | |
| Chen et al. | Multipurpose audio watermarking algorithm | |
| CN108877819A (en) | A kind of voice content evidence collecting method based on coefficient correlation | |
| Wang et al. | An audio watermarking scheme with neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190709 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |