CN107799121A - A kind of digital watermark embedding and method for detecting of radio broadcasting audio - Google Patents
A kind of digital watermark embedding and method for detecting of radio broadcasting audio Download PDFInfo
- Publication number
- CN107799121A CN107799121A CN201710977612.XA CN201710977612A CN107799121A CN 107799121 A CN107799121 A CN 107799121A CN 201710977612 A CN201710977612 A CN 201710977612A CN 107799121 A CN107799121 A CN 107799121A
- Authority
- CN
- China
- Prior art keywords
- audio
- data
- frequency
- heartbeat
- window
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000000605 extraction Methods 0.000 claims description 20
- 238000012795 verification Methods 0.000 claims description 17
- 230000001360 synchronised effect Effects 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 8
- 230000009977 dual effect Effects 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 5
- 230000005236 sound signal Effects 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 12
- 238000012544 monitoring process Methods 0.000 abstract description 9
- 238000003780 insertion Methods 0.000 abstract description 7
- 230000037431 insertion Effects 0.000 abstract description 7
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 5
- 241001269238 Data Species 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007639 printing Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000003139 buffering effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000036632 reaction speed Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000007306 turnover 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
-
- 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
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
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)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention discloses the digital watermark embedding of radio broadcasting audio and method for detecting, in insertion, by the phase difference between the ceiling capacity frequency of left and right acoustic channels to adjust each audio frame in the second audio window, and heartbeat watermark data is embedded into each audio frame in the second audio window.The legitimacy of audio is verified by extracting heartbeat watermark data in detection, realizes the content monitoring to broadcast singal.Using the embodiment of the present invention, help to carry out security control to the content that radio program broadcasts, reduce False Rate.
Description
Technical field
The present invention relates to radio broadcasting monitoring technical field, more particularly to a kind of digital watermark embedding of radio broadcasting audio
And method for detecting.
Background technology
Speech recognition technology had the development of abundance in recent years, in particular with the extensive general of the mobile terminal devices such as mobile phone
And but its be applied to radio broadcasting broadcast content monitoring on still have its limitation.First, speech recognition technology to music and
The judgement of uniformity can not be accomplished by singing the program of type, secondly relatively low to the phonetic recognization rate for the music that has powerful connections, speech recognition
As audio content monitoring key link be temperature-sensitive vocabulary full-text search, itself due to the range being related to, real-time etc. because
Element can all cause the error of judgement.
And in addition to speech recognition, also audio comparison technology, its be lean on the physics of two or more audio samples is special
Sign is compared, so as to obtain the whether consistent conclusion of sample.The maximum advantage of the technology is to be indifferent to the actual intension of sample,
The type of audio sample has been widened, basis for estimation is used as using the similarity of audio sample.But audio comparison technology is in the application
The situation of wrong report is had, the probability of wrong report changes with the difference of audio types, can have a strong impact on that it is monitored in radio broadcasting
Use value in system.
Therefore, monitored for the content that radio broadcasting is broadcasted, prior art has higher False Rate, can not be to broadcasting
Content carries out effective security control.
The content of the invention
The embodiment of the present invention proposes a kind of digital watermark embedding and method for detecting of radio broadcasting audio, contributes to wireless
The content of radio program transmission carries out security control, reduces False Rate.
The embodiment of the present invention provides a kind of data waterprint embedded method of radio broadcasting audio, including：
Audio signal is obtained, and the audio-frequency information is split, obtains dual channel data；
Audio frame window segmentation is carried out to the dual channel data, obtains the first audio window and the second audio window；Its
In, each audio window includes N number of sample data；N≥1；
Fast discrete Fourier's change is carried out to the sample data in first audio window and the second audio window respectively
Change, obtain the first frequency vector data and the second frequency vector data；
The energy amplitude of the first frequency vector data and the second frequency vector data is calculated respectively, and it is maximum to obtain first
Energy frequency and the second ceiling capacity frequency；
According between the heartbeat watermark data of default generation, the first ceiling capacity frequency and the second ceiling capacity frequency
Phase difference, adjust the second audio window in each audio frame, the heartbeat watermark data is embedded into the second audio window
In each audio frame in mouthful；What the heartbeat watermark data was made up of synchronous mark, Coordinated Universal Time(UTC) and verification data two enters
Data processed.
Further, the heartbeat watermark data of the default generation, the first ceiling capacity frequency and the second maximum energy
The phase difference between frequency is measured, adjusts each audio frame in the second audio window, is specially：
As 90 ° of phase difference ＜ between the first ceiling capacity frequency and the second ceiling capacity frequency, and need to be embedded in described
When heartbeat watermark data is 0, then do not adjust；
As 90 ° of phase difference ＜ between the first ceiling capacity frequency and the second ceiling capacity frequency, and need to be embedded in described
When heartbeat watermark data is 1, the phase of audio frame to be embedded in the second audio window is adjusted, makes the phase of the audio frame to be embedded
Position is symmetrically overturn around Y-axis；
Phase difference >=90 ° between the first ceiling capacity frequency and the second ceiling capacity frequency, and need to be embedded in described
When heartbeat watermark data is 1, then do not adjust；
Phase difference >=90 ° between the first ceiling capacity frequency and the second ceiling capacity frequency, and need to be embedded in described
When heartbeat watermark data is 0, the phase of audio frame to be embedded in the second audio window is adjusted, makes the phase of the audio frame to be embedded
Position is symmetrically overturn around Y-axis.
Further, it is described by the heartbeat watermark data be embedded into each audio frame in second audio window it
Afterwards, in addition to：
It is inverse to first audio window and the second audio window progress fast Fourier for having been inserted into heartbeat watermark data
Conversion, obtains and exports the voice data with watermark.
Further, second audio window is right audio channel window.
Correspondingly, the embodiment of the present invention also provides a kind of digital watermarking method for detecting of radio broadcasting audio, including：
Receive and demodulate fm broadcast signal, obtain left and right acoustic channels analogue audio frequency；
Audio frame window segmentation is carried out to the left and right acoustic channels analogue audio frequency, obtains the 3rd audio window and the 4th audio window
Mouthful；Wherein, each audio window includes N number of sample data；N≥1；
Fast discrete Fourier's change is carried out to the sample data in the 3rd audio window and the 4th audio window respectively
Change, obtain the 3rd frequency vector data and the 4th frequency vector data；
The energy amplitude of the 3rd frequency vector data and the 4th frequency vector data is calculated respectively, and it is maximum to obtain the 3rd
Energy frequency and the 4th ceiling capacity frequency；
According to the phase difference between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency, the 4th sound is extracted
The heartbeat watermark data of each audio frame in frequency window；The heartbeat watermark data is by synchronous mark, Coordinated Universal Time(UTC) and verification
The binary data of data composition；
The legitimacy of the heartbeat watermark data of extraction is verified, and according to the legitimacy of the heartbeat watermark data, judges institute
State the legitimacy of fm broadcast signal.
Further, the phase difference according between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency,
The heartbeat watermark data of each audio frame in the 4th audio window is extracted, is specially：
90 ° of the phase difference absolute value ＜ between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency, then extract
Heartbeat watermark data is 0；
Phase difference absolute value >=90 ° between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency, then extract
Heartbeat watermark data is 1.
Further, the legitimacy of the heartbeat watermark data of the verification extraction, it is specially：
According to the synchronous mark in the heartbeat watermark data of extraction, the verification in the heartbeat watermark data of the extraction is determined
Data, and verify its validity；
If the verification data is valid data, it is determined that the heartbeat watermark data of the extraction is valid data；It is no
Then, the heartbeat watermark data for determining the extraction is invalid data.
Further, this method also includes：It is determined that the extraction heartbeat watermark data be invalid data after, give up institute
After 1/10th sample datas for stating left and right acoustic channels analogue audio frequency, return to step：Sound is carried out to the left and right acoustic channels analogue audio frequency
Frequency frame window is split, and obtains the 3rd audio window and the 4th audio window.
Implement the embodiment of the present invention, have the advantages that：
The embedding algorithm and method for detecting of a kind of radio broadcasting audio provided in an embodiment of the present invention, in insertion
When, each audio frame in the second audio window is adjusted by the phase difference between the ceiling capacity frequency of left and right acoustic channels, and will
Heartbeat watermark data is embedded into each audio frame in the second audio window.In detection by extracting heartbeat watermark data come school
The legitimacy of audio is tested, realizes the content monitoring to broadcast singal.The content that high False Rate compared to prior art be present monitors,
Technical solution of the present invention helps to carry out security control to the content that radio program broadcasts, and reduces False Rate.
Brief description of the drawings
Fig. 1 is a kind of flow signal of embodiment of the data waterprint embedded method of radio broadcasting audio provided by the invention
Figure；
Fig. 2 is the energy amplitude schematic diagram of frequency vector provided by the invention；
Fig. 3 is a kind of flow signal of embodiment of the digital watermarking method for detecting of radio broadcasting audio provided by the invention
Figure.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
It is a kind of stream of embodiment of the data waterprint embedded method of radio broadcasting audio provided by the invention referring to Fig. 1
Journey schematic diagram, the method comprising the steps of 101 to step 105.Each step is specific as follows：
Step 101：Audio signal is obtained, and audio-frequency information is split, obtains dual channel data.
In the present embodiment, after obtaining audio signal, split signal source and obtain dual channel data.This paper audio frequency watermark is calculated
It is owned by France in blind watermatking type, monitored applied to frequency modulation broadcasting content.Conventional blind watermatking needs to judge continuous audio window when detecting
Mouth intermediate-frequeney point phase difference search synchronizing information, speed can influence the actual effect of content monitoring more slowly.
Step 102：Audio frame window segmentation is carried out to dual channel data, obtains the first audio window and the second audio window
Mouthful；Wherein, each audio window includes N number of sample data；N≥1.
In the present embodiment, a section audio is divided into several audio windows successively according to sample number.Enter for convenience
The follow-up Fast Fourier Transform (FFT) (FFT) of row, audio window are split according to 2 power (selection 128) number of samples, because
This uses different split windows for the signal source of different sample rates：
| Sample rate | Window sample number | Bit rate is transmitted in watermark |
| 48K | 128 | 375 |
| 44.1K | 128 | 344.5 |
| 32K | 128 | 250 |
1 different sampling rate signals of table sample index
In the present embodiment, insert data bit in the watermarking algorithm of the present embodiment, using in left and right acoustic channels audio window most
The size of big energy frequency phase difference represents data 0 and 1, therefore the transmission bit rate of watermark and the split window number one of audio per second
Cause.
In the present embodiment, the first audio window is L channel, and the second audio window is R channel.User can also basis
Actual conditions are changed, and such as the first audio window is R channel, and the second audio window is L channel.
Step 103：The sample data in the first audio window and the second audio window is carried out in fast discrete Fu respectively
Leaf transformation, obtain the first frequency vector data and the second frequency vector data.
Step 104：The energy amplitude of the first frequency vector data and the second frequency vector data is calculated respectively, obtains first
Ceiling capacity frequency and the second ceiling capacity frequency.
In the present embodiment, Fast Fourier Transform (FFT) is carried out for the audio window sample data of the sound channel of left and right two, obtained
Two groups of frequency vector datas of left and right acoustic channels are obtained, description below is just for L channel, and R channel performs identical processing.If in Fu
The frequency Vector Groups obtained after leaf transformation are：
(a0, b0), (a1, b1) ... (ak, bk)；K=128；
Calculate the energy amplitude Mn and phase Pn of each frequency vector.(n=1,2 ... k/2),(n=1,2 ... k/2).Due to after Fourier transformation frequency spectrum intermediate-frequeney point energy have using centre frequency as
The symmetrical characteristic of axle, therefore need to only calculate the data of scanning half front axle, i.e. n=0,1 ... k/2.As shown in Fig. 2 Fig. 2 is horizontal
Axle represents Frequency point, and if sample rate is 48KHZ, then frequency unit precision is FREQ/k=48KHZ/128=375HZ；Longitudinal axis table
Show the energy amplitude of vector field homoemorphism, i.e. frequency.0 point is DC component, represents the amplitude aggregate-value of each frequency, and the frequency is not joined
With calculating below.Energy amplitude finds amplitude maximum frequency (frequency of indicia circle in Fig. 2) in search frequency spectrum, records this frequency
The phase of point is PL.Right data is calculated according to same method, the phase for obtaining R channel ceiling capacity frequency is PR.
Step 105：According to the heartbeat watermark data of default generation, the first ceiling capacity frequency and the second ceiling capacity frequency
Between phase difference, adjust the second audio window in each audio frame, heartbeat watermark data is embedded into the second audio window
Each audio frame in；The binary data that heartbeat watermark data is made up of synchronous mark, Coordinated Universal Time(UTC) and verification data.
In the present embodiment, heartbeat watermark data, by synchronous mark (2 byte), UTC time (4 byte), verification data (2
Byte) amount to the binary data string that 8 bytes constitute 64.When watermark is embedded in, an audio frame is embedded in 1 bit
According to 64 audio frames just can completely be embedded in a watermark information.
In the present embodiment, the data for forming watermark can be a LOGO two-values picture or the character of copyright
String information.The watermarking algorithm for being applied to monitoring purpose discussed herein, in order to improve watermark detector efficiency, after watermark is lost
Can and alarm, then design used core data of the seldom UTC time of data volume as heartbeat watermark.Watermark detects
Front end can judge the reasonability of heartbeat using this data, while function when can also realize local school.Synchronous mark be 55AA,
0xaa, it detects front end as watermark, originates and judges for data.Verification data is CRC16 data, to since 55AA
Sync byte data are in order to which watermark detection front end carries out validity judgement to the check value of UTC last byte number evidence
It is used.
In the present embodiment, according to the heartbeat watermark data of default generation, the first ceiling capacity frequency and the second maximum energy
The phase difference between frequency is measured, adjusts each audio frame in the second audio window, is specially：
As 90 ° of phase difference ＜ between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in
When watermark data is 0, then do not adjust；
As 90 ° of phase difference ＜ between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in
When watermark data is 1, the phase of audio frame to be embedded in the second audio window is adjusted, makes the phase of audio frame to be embedded around Y-axis
Symmetrical upset；
Phase difference >=90 ° between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in
When watermark data is 1, then do not adjust；
Phase difference >=90 ° between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in
When watermark data is 0, the phase of audio frame to be embedded in the second audio window is adjusted, makes the phase of audio frame to be embedded around Y-axis
Symmetrical upset.
When i.e. watermark data need to be embedded in 0, then need to meet
When watermark data need to be embedded in 1, then need to meet
The present invention can meet above-mentioned want by overturning the sign bit of energy maximum frequency vector (an, bn) in right audio spectrum
Ask.What be should be noted is a little that if necessary to overturn vector, then all frequency vector couples in audio window need synchronously to turn over
Turn, to reduce influence of the vector difference of adjacent frequency to auditory perceptual.This paper blind watermatking core algorithm is using L channel as mark
Standard, watermark data is inserted in R channel, left and right acoustic channels are naturally synchronous, conveniently realize data detection.Methods herein belongs to blind
Watermarking algorithm, it but make use of the technology of non-blind watermatking.
One kind as the present embodiment is illustrated, after step 105, in addition to：To the first audio window and have been inserted into the heart
The second audio window for jumping watermark data carries out inverse fast Fourier transform, obtains and exports the voice data with watermark.Exist
After left and right acoustic channels are respectively completed inverse fast Fourier transform, left and right acoustic channels are merged and re-encoded, output carries the sound of watermark information
Frequently.
Correspondingly, the invention provides a kind of digital watermarking method for detecting of radio broadcasting audio.Referring to Fig. 3, Fig. 3 is this
A kind of schematic flow sheet of embodiment of the digital watermarking method for detecting of the radio broadcasting audio provided is provided.As shown in figure 3, should
Method for detecting includes step 301 to step 306.
Step 301：Receive and demodulate fm broadcast signal, obtain left and right acoustic channels analogue audio frequency.
In the present embodiment, antenna is empty receives fm broadcast signal, and left and right acoustic channels analogue audio frequency, AD samplings can be obtained after demodulation
Rate is preferably set for and is embedded in the equal speed of watermark, can so improve watermark data extraction validity (radio reception head chip is all
There is the function of setting sample rate).
Step 302：Audio frame window segmentation is carried out to left and right acoustic channels analogue audio frequency, obtains the 3rd audio window and the 4th sound
Frequency window；Wherein, each audio window includes N number of sample data；N≥1.
Step 303：The sample data in the 3rd audio window and the 4th audio window is carried out in fast discrete Fu respectively
Leaf transformation, obtain the 3rd frequency vector data and the 4th frequency vector data.
Step 304：The energy amplitude of the 3rd frequency vector data and the 4th frequency vector data is calculated respectively, is obtained
3rd ceiling capacity frequency and the 4th ceiling capacity frequency.
In the present embodiment, the realization principle of the step 302 of method for detecting to 304 and the step 102 of embedding grammar are to 104
It is identical.
Step 305 extracts the 4th sound according to the phase difference between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency
The heartbeat watermark data of each audio frame in frequency window；Heartbeat watermark data is by synchronous mark, Coordinated Universal Time(UTC) and verification data
The binary data of composition.
In the present embodiment, step 305 is specially：When between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency
90 ° of phase difference absolute value ＜, then the heartbeat watermark data extracted are 0；When the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency
Phase difference absolute value >=90 ° between point, then the heartbeat watermark data extracted are 1.
Step 306：The legitimacy of the heartbeat watermark data of extraction is verified, and according to the legitimacy of heartbeat watermark data, is sentenced
Determine the legitimacy of fm broadcast signal.
In the present embodiment, after step 305 extracts some bit binary datas, it is saved into buffering area, so
Search for 0x55 0xaa synchronous head in the buffer afterwards and carry out CRC16 verifications to determine the validity of watermark detection.Step
The legitimacy of the heartbeat watermark data of verification extraction in 306, it is specially：Synchronous mark in the heartbeat watermark data of extraction
Will, it is determined that the verification data in the heartbeat watermark data of extraction, and verify its validity；If verification data is valid data,
The heartbeat watermark data for then determining extraction is valid data；Otherwise, it determines the heartbeat watermark data of extraction is invalid data.
In the present embodiment, the digital watermarking method for detecting of radio broadcasting audio also includes：It is determined that the heartbeat water of extraction
Printing according to be invalid data after, after giving up 1/10th sample datas of left and right acoustic channels analogue audio frequency, return to step 302：To a left side
Right analog audio carries out audio frame window segmentation, obtains the 3rd audio window and the 4th audio window.So repeat, pass through
A period of time can stablize acquisition watermark data after adjusting.When stably continuously obtaining correct heartbeat watermark data, such as
Fruit occurs the undetectable situation of watermark and continues the regular hour suddenly, mean that watermark lose, program illegally intercutted or
Destroy.
The present embodiment technical scheme has advantages below：
1st, the right audio channel and L channel reference audio simultaneous transmission of watermark data are embedded into, keeps synchronous, water naturally
It is very convenient during print detection, there is the characteristics of non-blind watermatking；
Substantially it is each independent sound although being related when the 2, the audio of the left and right acoustic channels of frequency modulation broadcasting is most
Frequency evidence, embedded watermark data has the characteristics of blind watermatking naturally in certain road sound channel, i.e., unauthorized user cracks watermark
Difficulty is very big, discounting for efficiency of transmission in the case of, digital signature is inserted in watermark, watermark data is nothing in theory
What method cracked；
3rd, left and right acoustic channels, which exchange, does not influence watermark detection function, convenient construction and deployment；
4th, the multichannel characteristic insertion watermark data of left and right acoustic channels is taken full advantage of, left and right acoustic channels are naturally synchronous, so as to not
The adjacent audio window frequency domain characteristic of monophonic must be relied on and complete data insertion, but compare the frequency of left and right acoustic channels isochronous audio window
Domain characteristic leans on phase difference embedding data, therefore substantially increases valid data load, splits by 48KHZ sample rates, 128 sample numbers
Window calculation, watermark bit rate have reached 375bits/s；
5th, pi/2 phase adjustment is carried out when watermark data is embedded in, the inversion operation of sign bit need to be only completed, very high speed, have
Effect reduces CPU usage, can realize the broadcast program watermark insertion work(more covered under same hardware condition than conventional algorithm
Can, reduce cost；
6th, techniques described herein main purpose is for monitoring intercutting for illegal programs, therefore the watermark of detection equipment
Judge that speed will as much as possible quick and precisely.The present invention is more carefully dived print, and the top load rate that algorithm has using data volume,
The reaction speed judged for detection equipment watermark provides guarantee.By 48KHZ sample rates, 128 sample number split windows, heartbeat water
Printing calculates according to 8 bytes, under non-mute theory state each second can detect 5.8 watermarks (in the case of Jing Yin, due to each frequency
Point energy amplitude difference is smaller, and embedded watermark is easily destroyed, and watermark can be caused to detect mistake, thus it is Jing Yin when sentence without watermark
It is disconnected), the alarm response of second level can be also realized under conditions of appropriate mistake detection number is set as thresholding；
7th, watermark data insertion is realized using phase adjustment techniques.The masking effect of auditory perceptual is taken full advantage of, it is real
The embedding data transparency is showed；
8th, wide spectrum method of adjustment (all frequencies of full acoustic frequency window all adjust) is employed in adjustment phase place, it is not only synchronous
The phase difference (human ear to phase-unsensitive, but more sensitive to phase difference) of all frequencies is adjusted, reduces perceptibility, and can compare
The attack of AD conversion is preferably resisted, detection accuracy is improved, that is, improves robustness.
In summary, a kind of embedding algorithm of radio broadcasting audio provided in an embodiment of the present invention and detection side
Method, in insertion, each sound in the second audio window is adjusted by the phase difference between the ceiling capacity frequency of left and right acoustic channels
Frequency frame, and heartbeat watermark data is embedded into each audio frame in the second audio window.High miss be present compared to prior art
Sentence the content monitoring of rate, technical solution of the present invention helps to carry out security control to the content that radio program broadcasts, reduced
False Rate.
One of ordinary skill in the art will appreciate that realize all or part of flow in above-described embodiment method, being can be with
The hardware of correlation is instructed to complete by computer program, described program can be stored in a computer read/write memory medium
In, the program is upon execution, it may include such as the flow of the embodiment of above-mentioned each method.Wherein, described storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random Access
Memory, RAM) etc..
Described above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (8)
- A kind of 1. data waterprint embedded method of radio broadcasting audio, it is characterised in that including：Audio signal is obtained, and the audio-frequency information is split, obtains dual channel data；Audio frame window segmentation is carried out to the dual channel data, obtains the first audio window and the second audio window；Wherein, often Individual audio window includes N number of sample data；N≥1；Fast discrete Fourier transformation is carried out to the sample data in first audio window and the second audio window respectively, obtained To the first frequency vector data and the second frequency vector data；The energy amplitude of the first frequency vector data and the second frequency vector data is calculated respectively, obtains the first ceiling capacity Frequency and the second ceiling capacity frequency；According to the phase between the heartbeat watermark data of default generation, the first ceiling capacity frequency and the second ceiling capacity frequency Potential difference, each audio frame in the second audio window is adjusted, the heartbeat watermark data is embedded into second audio window Each audio frame in；The binary number that the heartbeat watermark data is made up of synchronous mark, Coordinated Universal Time(UTC) and verification data According to.
- 2. the data waterprint embedded method of radio broadcasting audio according to claim 1, it is characterised in that the default life Into heartbeat watermark data, the phase difference between the first ceiling capacity frequency and the second ceiling capacity frequency, adjustment second Each audio frame in audio window, it is specially：As 90 ° of phase difference ＜ between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in When watermark data is 0, then do not adjust；As 90 ° of phase difference ＜ between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in When watermark data is 1, adjust the phase of audio frame to be embedded in the second audio window, make the phase of the audio frame to be embedded around Y-axis is symmetrically overturn；Phase difference >=90 ° between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in When watermark data is 1, then do not adjust；Phase difference >=90 ° between the first ceiling capacity frequency and the second ceiling capacity frequency, and the heartbeat that need to be embedded in When watermark data is 0, adjust the phase of audio frame to be embedded in the second audio window, make the phase of the audio frame to be embedded around Y-axis is symmetrically overturn.
- 3. the data waterprint embedded method of radio broadcasting audio according to claim 1 or 2, it is characterised in that described The heartbeat watermark data is embedded into after each audio frame in second audio window, in addition to：Inverse fast Fourier transform is carried out to first audio window and the second audio window for having been inserted into heartbeat watermark data, Obtain and export the voice data with watermark.
- 4. the data waterprint embedded method of radio broadcasting audio according to claim 3, it is characterised in that second sound Frequency window is right audio channel window.
- A kind of 5. digital watermarking method for detecting of radio broadcasting audio, it is characterised in that including：Receive and demodulate fm broadcast signal, obtain left and right acoustic channels analogue audio frequency；Audio frame window segmentation is carried out to the left and right acoustic channels analogue audio frequency, obtains the 3rd audio window and the 4th audio window； Wherein, each audio window includes N number of sample data；N≥1；Fast discrete Fourier transformation is carried out to the sample data in the 3rd audio window and the 4th audio window respectively, obtained To the 3rd frequency vector data and the 4th frequency vector data；The energy amplitude of the 3rd frequency vector data and the 4th frequency vector data is calculated respectively, obtains the 3rd ceiling capacity Frequency and the 4th ceiling capacity frequency；According to the phase difference between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency, the 4th audio window is extracted The heartbeat watermark data of each audio frame in mouthful；The heartbeat watermark data is by synchronous mark, Coordinated Universal Time(UTC) and verification data The binary data of composition；The legitimacy of the heartbeat watermark data of extraction is verified, and according to the legitimacy of the heartbeat watermark data, judges the tune The legitimacy of frequency broadcast singal.
- 6. the digital watermarking method for detecting of radio broadcasting audio according to claim 5, it is characterised in that described according to institute The phase difference between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency is stated, extracts each audio in the 4th audio window The heartbeat watermark data of frame, it is specially：90 ° of the phase difference absolute value ＜ between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency, the then heartbeat extracted Watermark data is 0；Phase difference absolute value >=90 ° between the 3rd ceiling capacity frequency and the 4th ceiling capacity frequency, the then heartbeat extracted Watermark data is 1.
- 7. the digital watermarking method for detecting of radio broadcasting audio according to claim 6, it is characterised in that the verification carries The legitimacy of the heartbeat watermark data taken, it is specially：According to the synchronous mark in the heartbeat watermark data of extraction, the check number in the heartbeat watermark data of the extraction is determined According to, and verify its validity；If the verification data is valid data, it is determined that the heartbeat watermark data of the extraction is valid data；Otherwise, really The heartbeat watermark data of the fixed extraction is invalid data.
- 8. the digital watermarking method for detecting of radio broadcasting audio according to claim 7, it is characterised in that also include： After the heartbeat watermark data for determining the extraction is invalid data, give up 1/10th samples of the left and right acoustic channels analogue audio frequency After data, return to step：Audio frame window segmentation is carried out to the left and right acoustic channels analogue audio frequency, obtains the 3rd audio window and the Four audio windows.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710977612.XA CN107799121A (en) | 2017-10-18 | 2017-10-18 | A kind of digital watermark embedding and method for detecting of radio broadcasting audio |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710977612.XA CN107799121A (en) | 2017-10-18 | 2017-10-18 | A kind of digital watermark embedding and method for detecting of radio broadcasting audio |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107799121A true CN107799121A (en) | 2018-03-13 |
Family
ID=61533474
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710977612.XA Pending CN107799121A (en) | 2017-10-18 | 2017-10-18 | A kind of digital watermark embedding and method for detecting of radio broadcasting audio |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107799121A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09261070A (en) * | 1996-03-22 | 1997-10-03 | Sony Corp | Digital audio signal processing unit |
| CN1808970A (en) * | 2006-01-26 | 2006-07-26 | 西南交通大学 | Non-uniform discrete Fourier transform based robust digital watermark method for audio |
| CN101263552A (en) * | 2005-09-16 | 2008-09-10 | 汤姆森许可贸易公司 | Blind watermarking of audio signals by using phase modifications |
| CN101918999A (en) * | 2007-11-12 | 2010-12-15 | 尼尔森（美国）有限公司 | Methods and apparatus to perform audio watermarking and watermark detection and extraction |
| CN102132340A (en) * | 2008-08-15 | 2011-07-20 | Dts（Bvi）有限公司 | Parametric stereo conversion system and method |
| CN103237271A (en) * | 2013-04-02 | 2013-08-07 | 北京中视广信科技有限公司 | Method and system for embedding and detecting large-capacity video digital watermarks |
| CN204350002U (en) * | 2015-01-26 | 2015-05-20 | 浙江省广电科技股份有限公司 | A kind of digital watermark detector of frequency modulation broadcast system |
-
2017
- 2017-10-18 CN CN201710977612.XA patent/CN107799121A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09261070A (en) * | 1996-03-22 | 1997-10-03 | Sony Corp | Digital audio signal processing unit |
| CN101263552A (en) * | 2005-09-16 | 2008-09-10 | 汤姆森许可贸易公司 | Blind watermarking of audio signals by using phase modifications |
| CN1808970A (en) * | 2006-01-26 | 2006-07-26 | 西南交通大学 | Non-uniform discrete Fourier transform based robust digital watermark method for audio |
| CN101918999A (en) * | 2007-11-12 | 2010-12-15 | 尼尔森（美国）有限公司 | Methods and apparatus to perform audio watermarking and watermark detection and extraction |
| CN102132340A (en) * | 2008-08-15 | 2011-07-20 | Dts（Bvi）有限公司 | Parametric stereo conversion system and method |
| CN103237271A (en) * | 2013-04-02 | 2013-08-07 | 北京中视广信科技有限公司 | Method and system for embedding and detecting large-capacity video digital watermarks |
| CN204350002U (en) * | 2015-01-26 | 2015-05-20 | 浙江省广电科技股份有限公司 | A kind of digital watermark detector of frequency modulation broadcast system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102394062B (en) | Method and system for automatically identifying voice recording equipment source | |
| EP1522064B1 (en) | Audio identification system | |
| CN102265536B (en) | Methods and apparatus to perform audio watermarking and watermark detection and extraction | |
| CN101271690B (en) | Audio spread-spectrum watermark processing method for protecting audio data | |
| CN108597505B (en) | Voice recognition method and device and terminal equipment | |
| CN102074240A (en) | Digital audio watermarking algorithm for copyright management | |
| CN112116742B (en) | Identity authentication method, storage medium and equipment fusing multi-source sound production characteristics of user | |
| CN109117622B (en) | Identity authentication method based on audio fingerprints | |
| CN104851428A (en) | Method for realizing watermark encryption and decryption based on audio signal | |
| CN111653283B (en) | Cross-scene voiceprint comparison method, device, equipment and storage medium | |
| CN102237093B (en) | Echo hiding method based on forward and backward echo kernels | |
| CN110163787A (en) | Digital audio Robust Blind Watermarking Scheme embedding grammar based on dual-tree complex wavelet transform | |
| CN104992713A (en) | Fast audio comparing method | |
| Faundez-Zanuy et al. | Speaker identification security improvement by means of speech watermarking | |
| CN105283916A (en) | Digital-watermark embedding device, digital-watermark embedding method, and digital-watermark embedding program | |
| CN100545834C (en) | Method and apparatus based on the identification of the audio content of feature | |
| CN102915740A (en) | Phonetic empathy Hash content authentication method capable of implementing tamper localization | |
| CN109935233A (en) | A kind of recording attack detection method based on amplitude and phase information | |
| CN1848829B (en) | Method for automatic synchronizing of audio-frequency watermark | |
| JP4770194B2 (en) | Information embedding apparatus and method for acoustic signal | |
| CN107799121A (en) | A kind of digital watermark embedding and method for detecting of radio broadcasting audio | |
| CN109920447A (en) | Recording fraud detection method based on sef-adapting filter Amplitude & Phase feature extraction | |
| CN103236265A (en) | MP3Stegz steganography detecting method | |
| CN106205627B (en) | Digital audio reversible water mark algorithm based on side information prediction and histogram translation | |
| Wu et al. | Audio watermarking algorithm with a synchronization mechanism based on spectrum distribution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Hao Inventor after: Deng Jin Inventor after: Zou Wenfeng Inventor after: He Yongxing Inventor after: Zhang Ke Inventor after: Huang Ren Inventor after: Zheng Yongxi Inventor before: Chen Hao |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20210312 |