CN110047495B - High-capacity audio watermarking algorithm based on 2-level singular value decomposition - Google Patents
High-capacity audio watermarking algorithm based on 2-level singular value decomposition Download PDFInfo
- Publication number
- CN110047495B CN110047495B CN201910211717.3A CN201910211717A CN110047495B CN 110047495 B CN110047495 B CN 110047495B CN 201910211717 A CN201910211717 A CN 201910211717A CN 110047495 B CN110047495 B CN 110047495B
- Authority
- CN
- China
- Prior art keywords
- watermark
- matrix
- svd
- audio
- singular value
- 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
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 title description 18
- 239000011159 matrix material Substances 0.000 claims abstract description 73
- 230000005236 sound signal Effects 0.000 claims abstract description 40
- 230000009466 transformation Effects 0.000 claims abstract description 20
- 238000000034 method Methods 0.000 claims abstract description 16
- 238000006243 chemical reaction Methods 0.000 claims abstract description 11
- 238000000605 extraction Methods 0.000 claims abstract description 11
- 238000009432 framing Methods 0.000 claims abstract description 10
- 230000008569 process Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000000903 blocking effect Effects 0.000 claims abstract description 5
- 238000005516 engineering process Methods 0.000 description 7
- 238000011160 research Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000012952 Resampling Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000013432 robust analysis Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/0021—Image watermarking
-
- 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
- G10L19/0018—Speech coding using phonetic or linguistical decoding of the source; Reconstruction using text-to-speech synthesis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of information security and audio processing, and discloses a high-capacity audio watermark embedding and extracting method based on 2-level singular value decomposition, which aims to embed a gray image into carrier audio, has larger embedding capacity, can meet the requirements of imperceptibility and robustness and has higher security, and comprises the following steps: firstly, watermark embedding process: (1) watermark image blocking processing: firstly, SVD conversion is carried out for the first time; second SVD conversion; decomposing original audio wavelet; (2) audio signal framing processing: firstly, SVD conversion is carried out for the first time; second SVD conversion; embedding watermark information; fourthly, SVD inverse transformation; replacing the matrix H; sixthly, carrying out SVD inverse transformation for the second time; seventhly, constructing an audio signal containing a watermark by inverse transformation; and II, watermark extraction. The invention is mainly applied to the audio processing occasion.
Description
Technical Field
The invention belongs to the technical field of information security and audio processing, and particularly relates to a high-capacity audio watermarking algorithm based on 2-level singular value decomposition
Background
The rapid development of internet technology provides great convenience for the dissemination of audio digital content, and also causes piracy infringement and an increasing flooding of audio digital content tampering behaviors. How to manage copyright and protect integrity of audio digital contents becomes a research hotspot in academia. In recent years, audio watermarking technology has been proposed by scholars, and the audio watermarking technology achieves the purpose of protecting the copyright ownership and integrity of audio digital content by embedding watermark information in the audio digital content in an imperceptible manner. The digital audio watermarking technology needs to meet the requirements of three aspects, mainly including (1) imperceptibility, (2) robustness and (3) embedding capacity, and the three are in a mutual contradictory relationship, and the improvement of one aspect inevitably causes the reduction of the performances of the other two aspects. Therefore, the current research on audio digital watermarking mainly focuses on how to design an audio watermarking algorithm so that the audio watermarking algorithm has good robustness, imperceptibility and high embedding capacity.
A series of new audio watermarking methods and new technologies are created based on the technology. However, most algorithms embed a pseudo-random sequence and a binary image as watermarks in an audio carrier, and the research on grayscale images is less, grayscale images have more visual and rich image information compared with binary images, and the requirements on embedding capacity of the watermarking algorithms are higher, and meanwhile, the imperceptibility and robustness are required to be met, which is a challenge to be overcome by the audio watermarking technology. Based on the purpose, the invention aims to provide a new watermarking algorithm, and the gray level image is embedded into the carrier audio as the watermark, so that the embedding capacity is greatly improved, and meanwhile, the invention also has good imperceptibility and robustness.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a new audio watermarking algorithm to realize the embedding of the gray level image into the carrier audio, the method has larger embedding capacity, and simultaneously can meet the requirements of imperceptibility and robustness, in addition, the algorithm also has higher safety, and the original watermarking information can not be decrypted under the condition of not knowing a correct key, thereby achieving the purpose of protecting the copyright of the audio information. The invention relates to a high-capacity audio watermark embedding and extracting method based on 2-level singular value decomposition, which comprises the following steps:
1. watermark embedding process
(1) Watermark image blocking process
Selecting a grayscale image B as a watermark image, wherein the grayscale image B has a size of M × M, and the original image is divided into 8 × 8 small blocks Ci,1≤i≤n,n＝(M/8)2；
First SVD conversion
After the watermark image is divided into blocks, C is applied to each small blockiUsing SVD decomposition to select the maximum singular value S obtained after each small block is decomposedi(1,1) forming them into a new matrix, called W, with a size of (M/8) × (M/8);
second SVD transform
For the matrix W, a second SVD transformation is performed:
[Uw,Sw,Vw T]＝SVD(W) (1)
u is a left singular value matrix, Vw TFor the matrix of right singular values, the matrix U isw，VwStored for later extraction;
original audio wavelet decomposition
One-dimensional two-level discrete wavelet transform is applied to the audio signal A to obtain an approximate component cA2And a detail component cD1,cD2：
[cA2,cD2,cD1]＝DWT(A) (2)
(2) Audio signal framing
Selecting approximate components to perform framing processing, and then reconstructing each frame into a two-dimensional matrix, wherein the number of frames is (M/8) × (M/8);
first SVD conversion
Performing SVD conversion on each frame, selecting the maximum singular value S (1,1) obtained after audio decomposition of each frame, and forming a new matrix, namely H, with the size of (M/8) x (M/8);
second SVD transform
For matrix H, we apply a second SVD transform
[Uh,Sh,Vh T]＝SVD(H) (3)
Embedding watermark information
The strategy adopted by the watermark information embedding is to modify the maximum singular value of each frame of the audio signal, and the embedding formula is as follows:
Sh'＝Sh+α·Sw (4)
wherein Sh' is a singular value matrix in which the watermark has been embedded, and α is the embedding strength, usually with a value range of (0, 1), the choice of which should satisfy the balance of imperceptibility and robustness;
iv inverse SVD transform
Will matrix Sh',Uh，Vh TMultiplication, generating a new matrix H':
H'＝Uh×Sh'×Vh T (5)
replacing matrix H
Each element of matrix H is S '(1,1), then each S (1,1) from matrix H is replaced by S' (1, 1):
S(1,1)←S'(1,1) (6)
sixth, the second SVD inverse transformation
For each frame of audio, applying an inverse SVD transform;
seventhly, inverse transformation constructs audio signal containing watermark
Reconstructing the audio signal by applying two-level discrete inverse wavelet transform to obtain an audio signal A1 containing a watermark;
2. the watermark extraction process comprises the following specific steps:
(1) reading audio signals containing watermarks
Reading the audio signal a1 after embedding the watermark;
(2) wavelet decomposition
Applying one-dimensional two-level discrete wavelet transform to the watermark-containing audio signal to obtain an approximate component and a detail component;
(3) watermark-containing audio framing
Dividing the approximate components into non-overlapping frames, and then reconstructing each frame into a two-dimensional matrix;
(4) SVD transform
Performing SVD transformation on each two-dimensional matrix, forming a new matrix by the maximum singular value S' (1,1) of each frame, and then applying the second singular value transformation to the new matrix;
(5) extracting watermarks
Watermark information is extracted according to the following formula:
Sw＝(Sh'-Sh)/α (7)
Swis the singular value, S, of the extracted watermark matrixhIs the singular value of the matrix H;
(6) inverse SVD transform
W'＝Uw×Sw×Vw T (8)
Uw，Vw TRespectively storing a left singular value matrix and a right singular value matrix in the embedding process;
(7) replacement matrix W'
Each element of the matrix W is replaced by each element of W, W coming from the embedding step 2;
(8) inverse SVD transform
The inverse SVD transform is applied to each 8 x 8 patch and then these patches are fused into a new image.
The invention has the characteristics and beneficial effects that:
the invention takes the gray level image as the watermark to be embedded into the carrier audio, and has richer and more visual image information compared with the prior algorithm that most of the embedded watermarks are pseudo-random sequences and binary images. The innovation of the invention is that 2-level SVD transformation is respectively applied to the carrier audio and the watermark image, the maximum singular value obtained by decomposing the watermark image is embedded into the maximum singular value of the carrier audio, the SVD transformation is adopted to compress the watermark information, so that the embedding capacity is improved, and meanwhile, the SVD transformation has better stability and can resist the attack of conventional signal processing. The experimental result shows that the robustness is improved by applying the 2-level SVD transformation under the same perception constraint as the first-level SVD transformation, and even some desynchronization attacks can be resisted. In terms of security, the matrix U is divided intow，VwThe watermark image is stored as a secret key and transmitted to an extraction terminal, and the original replacement scrambling encryption preprocessing is not needed to be adopted for the watermark image, so that the complexity of the algorithm is reduced, and the consumption of time and space is saved.
Description of the drawings:
fig. 1 is a schematic view of an embedded portion.
FIG. 2 is a schematic diagram of the extraction section.
Fig. 3 shows the original audio signal, the watermarked audio, and the difference of two audios, the audio being music of a duration of 95 s.
Fig. 4 is an original watermark image, whose pixels are 256 × 256.
Fig. 5 is a watermark image extracted without adding any attack.
Fig. 6 is a watermark image extracted after adding gaussian noise with a signal-to-noise ratio of 10dB to watermarked audio.
Fig. 7 is a watermark image extracted after adding gaussian noise with a signal-to-noise ratio of 20dB to watermarked audio.
Fig. 8 is a watermark image extracted after adding gaussian noise with a signal-to-noise ratio of 30dB to watermarked audio.
Fig. 9 is a watermark image extracted after watermarked audio has been filtered using a Butterworth filter.
Fig. 10 is a watermark image extracted after a resampling attack.
Fig. 11 is a watermark image extracted after a weighting attack.
Fig. 12 is a watermark image extracted after a cropping attack.
Detailed Description
The invention provides a high-capacity audio watermarking algorithm based on 2-level singular value decomposition, and robustness is improved, the scheme of the embodiment 1 is described in detail by combining the design principles of figures 1 and 2, and the specific steps are as follows:
example 1
1. Watermark embedding algorithm
(1) Watermark image blocking process
Selecting a grayscale image B as a watermark image, wherein the size of the grayscale image B is M × M (M is 256), and dividing the original image into 8 × 8 small blocks Ci(1≤i≤n),n＝(M/8)2＝1024。
(2) First SVD transform
After the watermark image is divided into blocks, C is applied to each small blockiApplying SVD transform Ci＝Ui×Si×Vi TThe maximum singular value S (1,1) obtained after the decomposition of each small block is selected and is formed into a new matrix, which is called W, and the size of the matrix is (M/8) × (M/8).
(3) Second SVD transform
For matrix W, we apply a second SVD transform
[Uw,Sw,Vw T]＝SVD(W) (9)
Will matrix UwAnd Vw TSaved for later extraction.
(4) Wavelet decomposition of audio signals
One-dimensional two-level discrete wavelet transform is applied to the audio signal A to obtain an approximate component cA2And a detail component cD1,cD2.
[cA2,cD2,cD1]＝DWT(A) (10)
(5) Audio signal framing
We select the approximate components for framing and then reconstruct each frame into a two-dimensional matrix, which is divided into (M/8) × (M/8) — 1024 frames.
(6) First SVD transform
SVD conversion is carried out on each frame of audio, the maximum singular value S (1,1) obtained after decomposition of each frame of audio signal is selected, and the maximum singular value S are combined into a new matrix, which is called H, and the size of the matrix is (M/8) x (M/8), and when the watermark is extracted, only the matrix is needed.
(7) Second SVD transform
For matrix H, we apply a second SVD transform
[Uh,Wh,Vh T]＝SVD(H) (11)
(8) Watermark information embedding
The strategy adopted by the watermark information embedding is to modify the maximum singular value of each frame of the audio signal, and the embedding formula is as follows:
Sh'＝Sh+α·Sw (12)
wherein Sh' is the singular value matrix in which the watermark has been embedded, and α is the embedding strength, usually with a value in the range of (0, 1), and its choice should satisfy the balance between imperceptibility and robustness, where we take α to be 0.20.
(9) Inverse SVD transform
Will matrix Sh',Uh，VhThe multiplication produces a new matrix H'.
H'＝Uh×Sh'×Vh T (5)
(10) Replacement matrix H
Each element of matrix H is S '(1,1), and then each S (1,1) from matrix H is replaced by S' (1, 1).
S(1,1)←S'(1,1) (6)
(11) Second inverse SVD transform
For each frame of audio, the inverse SVD transform is applied.
(12) Inverse transformation to construct watermarked audio signals
And (3) reconstructing the audio signal by applying two-level discrete inverse wavelet transform to obtain the audio signal A1 containing the watermark.
2. Watermark extraction algorithm
The watermark extraction is the inverse process of watermark embedding, and the specific steps are as follows:
(1) reading audio signals containing watermarks
The watermarked audio file a1 is read.
(2) Wavelet decomposition of audio signal containing watermark
And applying one-dimensional two-level discrete wavelet transform to the watermark-containing audio signal to obtain an approximate component and a detail component.
(3) Watermark-containing audio framing
The approximation components are divided into non-overlapping frames, and each frame is then reconstructed into a two-dimensional matrix.
(4) SVD transform
The SVD transform is applied to each two-dimensional matrix, the maximum singular values S' (1,1) of each frame constitute a new matrix, to which the second SVD transform is then applied.
(5) Extracting watermarks
Watermark information is extracted according to the following formula:
Sw＝(Sh'-Sh)/α (13)
wherein ShIs the singular value of the matrix H, SwAre the singular values of the extracted watermark matrix.
(6) Inverse SVD transform
An SVD inverse transform is applied to the extracted matrix singular values.
W'＝Uw×Sw×Vw T (14)
(7) Replacement matrix W'
Each element of the matrix W is replaced by each element of W, W from the embedding step 2.
(8) Inverse SVD transform
The inverse SVD transform is applied to each 8 x 8 small block and these small blocks are then merged into a new image of size M x M (M256), i.e. the watermark image B1 we are to extract.
Example 2
In order to further verify the technical effect of the technical scheme of the present invention, the embodiment of the processing method for audio watermark embedding and watermark extraction of the present invention is subjected to simulation testing, which is described in detail in the following description:
(1) imperceptibility analysis verification
The signal-to-noise ratio SNR is adopted to verify the imperceptibility of the method, and the specific formula is as follows:
where S (i) is the original audio signal and S' (i) is the audio with embedded watermark, the larger the SNR value, the better the imperceptibility of the solution.
After the original audio signal shown in fig. 3 is processed by the processing method for embedding an audio watermark of the present invention, the audio signal with a watermark shown in fig. 3 is generated, as can be seen from fig. 3, the original audio and the audio after embedding the watermark are not different, and no difference can be heard audibly, and the SNR obtained by calculation is 26.9596dB, which indicates that the technical solution has good imperceptibility.
(2) Robustness analysis verification
The robustness is used for measuring the anti-attack capability of an audio watermarking algorithm, the robustness of the algorithm is measured by calculating the similarity between an original watermark and an extracted watermark, and a normalized correlation coefficient NC is selected as a robustness evaluation standard, wherein the specific formula is as follows:
where w (i, j) is the original watermark image and w x (i, j) is the extracted watermark image.
The watermark image adopted by the invention is shown in fig. 4, the watermark image extracted under the condition of no attack is shown in fig. 5, and NC-1 shows that the watermark image is the same as the original image and has no difference.
To verify the robustness of the algorithm, the following attacks are taken on the watermarked audio signal:
1) gaussian white noise with different signal-to-noise ratios is added, the watermark images extracted from the audio after the signal-to-noise ratios are respectively added to 10dB, 20dB and 30dB in the images 6-8, and the NC values are respectively 0.9998, 1.0 and 1.0, so that when the signal-to-noise ratio is added to 10dB, little noise can be seen in the extracted watermark images, and the rest two images cannot see any noise and have no difference with the original image.
2) And (4) filtering attack, in fig. 9, a watermark image is extracted after filtering by using a Butterworth digital filter, and NC is 1.0, and basically no difference from the original image exists.
3) The resampling attack firstly down-samples the audio signal embedded with the watermark to 22050Hz, then restores to 44100Hz, and extracts the watermark image as shown in fig. 10, where NC is 1.0, and it can be seen that there is no difference from the original image.
4) The weighting attack is to re-quantize the audio signal embedded with the watermark, where the quantization bit number is 8 bits, and the extracted watermark image is shown in fig. 11, where NC is 1.0, which is seen to be no different from the original image.
5) Shear attack
The audio signal with the watermark is subjected to a 1% cropping operation, that is, 41000 samples are removed from a 95s signal by randomly selecting different positions, and the extracted watermark image is as shown in fig. 12, where NC is 0.9925, and it can be seen that the extracted watermark image has a little noise, but the watermark image can still be distinguished.
(3) Watermark embedding capacity analysis verification
Watermark embedding capacity (Payload) refers to the amount of data that can be embedded in an audio signal per unit length, and is usually expressed in terms of bit rate (bit/s), i.e., how many bits of watermark information can be embedded in each second of audio, and the international association for phonography (IFPI) requires that the embedded watermark capacity be at least 20 bits/s. The watermark embedding capacity depends on the number of bits of the watermark and the length of the embedded audio (Time), and is specifically disclosed as follows:
wherein N iswThe number of watermark bits is N, the Time is the length of the audio signalw256 × 256 to 65536 bits, the watermark length is 95s, so the watermark embedding capacity is 689.85 bits/s, and the invention has larger embedding capacity.
Those skilled in the art will appreciate that the drawings are only schematic illustrations of preferred embodiments, and the above-described embodiments of the present invention are merely provided for description and do not represent the merits of the embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (1)
1. A high-capacity audio watermark embedding and extracting method based on 2-level singular value decomposition is characterized by comprising the following steps:
watermark embedding process
(1) Watermark image blocking process
Selecting a grayscale image B as a watermark image, wherein the grayscale image B has a size of M × M, and the original image is divided into 8 × 8 small blocks Ci,1≤i≤n,n＝(M/8)2；
First SVD conversion
After the watermark image is divided into blocks, C is applied to each small blockiUsing SVD decomposition to select the maximum singular value S obtained after each small block is decomposedi(1,1) forming them into a new matrix, called W, with a size of (M/8) × (M/8);
second SVD transform
For the matrix W, a second SVD transformation is performed:
[Uw,Sw,Vw T]＝SVD(W)
Uwis a matrix of left singular values, Vw TIs a matrix of the right singular values,will matrix Uw，VwStored for later extraction;
original audio wavelet decomposition
One-dimensional two-level discrete wavelet transform is applied to the audio signal A to obtain an approximate component cA2And a detail component cD1,cD2：
[cA2,cD2,cD1]＝DWT(A)
(2) Audio signal framing
Selecting approximate components to perform framing processing, and then reconstructing each frame into a two-dimensional matrix, wherein the number of frames is (M/8) × (M/8);
first SVD conversion
Performing SVD conversion on each frame, selecting the maximum singular value S (1,1) obtained after audio decomposition of each frame, and forming a new matrix, namely H, with the size of (M/8) x (M/8);
second SVD transform
Applying a second SVD transform to matrix H
[Uh,Sh,Vh T]＝SVD(H)
Embedding watermark information
The strategy adopted by the watermark information embedding is to modify the maximum singular value of each frame of the audio signal, and the embedding formula is as follows:
Sh′＝Sh+α·Sw (4)
wherein Sh' is a singular value matrix in which a watermark has been embedded, alpha is embedding strength, the value range is (0, 1), and the selection should satisfy the balance of imperceptibility and robustness;
iv inverse SVD transform
Will matrix Sh′,Uh，Vh TMultiplication, generating a new matrix H':
H′＝Uh×Sh′×Vh T
replacing matrix H
Each element of matrix H is S '(1,1), then each S (1,1) from matrix H is replaced by S' (1, 1):
S(1,1)←S′(1,1)
sixth, the second SVD inverse transformation
For each frame of audio, applying an inverse SVD transform;
seventhly, inverse transformation constructs audio signal containing watermark
Reconstructing the audio signal by applying two-level discrete inverse wavelet transform to obtain an audio signal A1 containing a watermark;
secondly, the watermark extraction process comprises the following specific steps:
(1) reading audio signals containing watermarks
Reading the audio signal a1 after embedding the watermark;
(2) wavelet decomposition
Applying one-dimensional two-level discrete wavelet transform to the watermark-containing audio signal to obtain an approximate component and a detail component;
(3) watermark-containing audio framing
Dividing the approximate components into non-overlapping frames, and then reconstructing each frame into a two-dimensional matrix;
(4) SVD transform
Performing SVD transformation on each two-dimensional matrix, forming a new matrix by the maximum singular value S' (1,1) of each frame, and then applying the second singular value transformation to the new matrix;
(5) extracting watermarks
Watermark information is extracted according to the following formula:
Sw＝(Sh′-Sh)/α (7)
Swis the singular value, S, of the extracted watermark matrixhIs the singular value of the matrix H;
(6) inverse SVD transform
W′＝Uw×Sw×Vw T (8)
Uw，Vw TRespectively storing a left singular value matrix and a right singular value matrix in the embedding process;
(7) replacement matrix W'
Each element of the matrix W is replaced by each element of W', and W is from the first SVD transformation in the watermark image blocking process;
(8) inverse SVD transform
The inverse SVD transform is applied to each 8 x 8 patch and then these patches are fused into a new image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910211717.3A CN110047495B (en) | 2019-03-20 | 2019-03-20 | High-capacity audio watermarking algorithm based on 2-level singular value decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910211717.3A CN110047495B (en) | 2019-03-20 | 2019-03-20 | High-capacity audio watermarking algorithm based on 2-level singular value decomposition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110047495A CN110047495A (en) | 2019-07-23 |
| CN110047495B true CN110047495B (en) | 2021-04-27 |
Family
ID=67273908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910211717.3A Expired - Fee Related CN110047495B (en) | 2019-03-20 | 2019-03-20 | High-capacity audio watermarking algorithm based on 2-level singular value decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110047495B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112927700B (en) * | 2021-02-06 | 2024-03-19 | 兰州理工大学 | Blind audio watermark embedding and extracting method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140005815A1 (en) * | 2012-06-27 | 2014-01-02 | Krishna Rao KAKKIRALA | Method and System for Blind Audio Watermarking |
| CN103955879A (en) * | 2014-04-11 | 2014-07-30 | 杭州电子科技大学 | DWT-SVD robust blind watermark method based on multilevel DCT |
| CN106611601A (en) * | 2017-01-23 | 2017-05-03 | 天津大学 | Audio watermarking method based on DWT-SVD and PSO (Particle Swarm Optimization) |
-
2019
- 2019-03-20 CN CN201910211717.3A patent/CN110047495B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140005815A1 (en) * | 2012-06-27 | 2014-01-02 | Krishna Rao KAKKIRALA | Method and System for Blind Audio Watermarking |
| CN103955879A (en) * | 2014-04-11 | 2014-07-30 | 杭州电子科技大学 | DWT-SVD robust blind watermark method based on multilevel DCT |
| CN106611601A (en) * | 2017-01-23 | 2017-05-03 | 天津大学 | Audio watermarking method based on DWT-SVD and PSO (Particle Swarm Optimization) |
Non-Patent Citations (3)
| Title |
|---|
| 《SVD Based Audio Watermarking》;Masato Ogura et al.;《International Conference on Electrical, Computer and Communication Engineering(ECCE)》;20170218;第119-122页 * |
| 《基于LWT-SVD的鲁棒音频水印算法》;杨志疆等;《光电子·激光》;20181130;第29卷(第11期);第1221-1227页 * |
| 《离散小波变换与奇异值分解的音频信号水印算法》;冯小明等;《华侨大学学报(自然科学版)》;20161130;第37卷(第6期);第770-773页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110047495A (en) | 2019-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Arora | A DWT-SVD based robust digital watermarking for digital images | |
| Kaur et al. | Steganographic approach for hiding image in DCT domain | |
| Amirgholipour et al. | Robust digital image watermarking based on joint DWT-DCT | |
| Kasmani et al. | A new robust digital image watermarking technique based on joint DWT-DCT transformation | |
| Divecha et al. | Implementation and performance analysis of DCT-DWT-SVD based watermarking algorithms for color images | |
| Malonia et al. | Digital image watermarking using discrete wavelet transform and arithmetic progression technique | |
| EP1695539B1 (en) | Method for robust lossless data hiding and recovering from the integer wavelet representation | |
| CN110047495B (en) | High-capacity audio watermarking algorithm based on 2-level singular value decomposition | |
| Zeng et al. | Extraction of multiresolution watermark images for resolving rightful ownership | |
| Sharma et al. | Robust technique for steganography on Red component using 3-DWT-DCT transform | |
| Parameswaran et al. | Content-based watermarking for image authentication using independent component analysis | |
| Goel et al. | ADCT-based robust methodology for image steganography | |
| Pawar et al. | Histogram-based reversible data hiding using block division | |
| Liao et al. | A novel watermark embedding scheme using compressive sensing in wavelet domain | |
| Mirza et al. | Color image watermarking and self-recovery based on independent component analysis | |
| Yadav | Dwt–svd–wht watermarking using varying strength factor derived from means of the wht coefficients | |
| Sharma et al. | Robust image watermarking technique using contourlet transform and optimized edge detection algorithm | |
| Jagadeesh et al. | A genetic algorithm based oblivious image watermarking scheme using singular value decomposition (SVD) | |
| Ma et al. | A robust watermarking scheme based on dual quantization of wavelet significant difference | |
| Barkat et al. | Time-frequency and time-scale-based fragile watermarking methods for image authentication | |
| Phadikar et al. | Multibit QIM watermarking using M-ary modulation and lifting | |
| Shiju et al. | Performance analysis of high dynamic range image watermarking based on quantization index modulation | |
| Sukumar et al. | Multi Image-Watermarking scheme based on Framelet and SVD | |
| Kumari et al. | Robust image watermarking algorithm DWT-SVD-AES for telemedicine application | |
| You et al. | Two-Stage Watermark Removal Framework for Spread Spectrum Watermarking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210427 |