CN102496135A - Deadweight tonnage (DWT) domain-based digital watermark method and system - Google Patents
Deadweight tonnage (DWT) domain-based digital watermark method and system Download PDFInfo
- Publication number
- CN102496135A CN102496135A CN2011104009618A CN201110400961A CN102496135A CN 102496135 A CN102496135 A CN 102496135A CN 2011104009618 A CN2011104009618 A CN 2011104009618A CN 201110400961 A CN201110400961 A CN 201110400961A CN 102496135 A CN102496135 A CN 102496135A
- Authority
- CN
- China
- Prior art keywords
- watermark
- image
- wavelet
- coefficient
- gray
- 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.)
- Granted
Links
Images
Landscapes
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a deadweight tonnage (DWT) domain-based digital watermark method. A watermark embedding process comprises the following steps that: inverse pretreatment of a gray-scale watermark image; luminance-bandwidth-chrominance (YUV) color space conversion of an original colorful image; wavelet secondary transformation of weight Y; wavelet transformation of gray-scale watermark image; difference value storage of watermark image wavelet coefficient; encryption arrangement of watermark image wavelet coefficient; insertion of watermark image wavelet coefficient and inverse coefficient; and wavelet inverse transformation of inserted watermark image; and the watermark extraction process comprises the following steps that: YUV color space conversion of the inserted watermark image; wavelet secondary transformation of weight Y of the inserted watermark image; wavelet coefficient restoration of the gray-scale watermark image according to the encryption arrangement of the gray-scale watermark image; all-zero compensation of a gray-scale watermark image wavelet coefficient diagonal band; and wavelet coefficient inverse transformation of the gray-scale watermark image. Due to the adoption of the method and the system, the invisibility of a watermark can be guaranteed, and good robustness on the attacks such as compression, noise plus and filtering of a joint photographic experts group (JPEG) can be realized.
Description
Technical Field
The invention belongs to the technical field of information hiding in the aspect of communication, and relates to a digital watermarking method and a digital watermarking system based on a DWT (discrete wavelet transform) domain.
Background
The digital watermarking technology is a technology which directly embeds some identification information into a digital carrier, does not affect the use value of the original carrier, is not easy to be detected and modified again, and can be identified and recognized by a production party. The information hidden in the carrier can be used for the purposes of confirming a content creator, a purchaser, transmitting secret information, judging whether the carrier is tampered or not, and the like. Digital watermarking is an important research direction of information hiding technology and an important branch of the research field of the information hiding technology.
In recent years, with the development of the digital age, more and more information materials are stored and transmitted in a digital form. Digital watermarking is more and more emphasized as a digital work protection means, and the application field of the digital watermarking is expanded accordingly. The basic application areas of digital watermarking today are copyright protection, hidden identification, authentication and secure invisible communication. When digital watermarking is applied to copyright protection, potential application markets are e-commerce, online or offline distribution of multimedia content, and large-scale broadcast services. When the digital watermark is used for hiding the identification, the digital watermark can be applied to the fields of medicine, drawing, digital imaging, digital image monitoring, multimedia indexing, content-based retrieval and the like. In the aspect of digital watermark authentication, the secure invisible communication of the digital watermark on a main ID card, a credit card, an ATM card and the like can be widely applied in national defense and intelligence departments.
In the prior art, most of digital watermarks embed watermark information into a spatial domain or a frequency domain of an image, and the method is widely used because wavelet transformation can decompose spatial and temporal signals at a plurality of different resolution scales and can process the signals according to the different resolution scales of the signals. There are many proposed methods for gray scale watermarking, blind watermarking and digital watermarking in color images based on wavelet transform. However, these methods all have some defects, and often only can consider the requirements of one or two points, and cannot completely realize the color image of the gray blind watermark.
Therefore, it is necessary to research and develop a solution to solve the above-mentioned defects in the prior art, and improve the stability of the system and ensure the concealment and security of the watermark while realizing blind watermark extraction.
Disclosure of Invention
In order to solve the above problems, an object of the present invention is to provide a DWT domain-based digital watermarking method and system, so as to implement embedding and extraction of a blind watermark image with gray scale, and ensure the concealment and security of the watermark.
In order to achieve the purpose, the technical scheme of the invention is as follows:
a digital watermarking method based on DWT domain includes embedding watermark and extracting watermark, wherein, the embedding of watermark includes the following steps:
s10: acquiring an original color image to be embedded with a watermark and a gray watermark image to be embedded;
s12: YUV color space conversion is carried out on the original color image, and secondary wavelet transformation is carried out on the Y component of the original color image after conversion;
s13: carrying out reverse color preprocessing on the gray watermark image, carrying out wavelet transformation on the gray watermark image subjected to the reverse color preprocessing, and discarding a high-frequency diagonal sub-band part;
s14: performing difference value storage on the wavelet coefficient low-frequency sub-band after wavelet transformation of the gray watermark image;
s15: multiplying the low-frequency coefficient stored by the difference value and other wavelet transform coefficients by respective coefficients, then sequencing the low-frequency coefficient and other wavelet transform coefficients together with watermark color reversal preprocessing variables, and taking the sequencing sequence as a key;
s16: determining the number of times of repeated embedding of the coefficient sequence by calculating the sizes of the original color image and the gray watermark image;
s17: embedding the sequence in the step S15 into the diagonal sub-band of the secondary wavelet transform coefficient of the original image according to the repetition times calculated in the step S16;
s18: performing secondary wavelet inverse transformation on the color image with the embedded coefficient to obtain a Y component, and performing color space transformation together with Cr and Cb components of the original image to obtain an image with the embedded watermark;
the extraction of the watermark comprises the following steps:
s20: carrying out YUV color space conversion on the image embedded with the watermark, and carrying out secondary wavelet transformation on a Y component in the image;
s21: the diagonal high-frequency subband portion of the secondary wavelet transform is subjected to the key extraction in step S15 to extract subbands of the grayscale watermark image and an inverse preprocessing decision variable epsilon'.
S22: and completing all zeros of the diagonal sub-bands of the wavelet coefficients of the gray watermark image, and performing wavelet inverse transformation on the obtained average value of the sub-band coefficients and the completely zeroed diagonal sub-bands to obtain the extracted gray watermark image.
A DWT domain-based digital watermarking system includes a watermark embedding section and a watermark extraction section, wherein the watermark embedding section includes: an image obtaining unit for obtaining an original color image to be embedded with a watermark and a gray watermark image to be embedded, a color space conversion and secondary wavelet conversion unit connected with the image obtaining unit and used for YUV color space conversion of the obtained original color image and carrying out secondary wavelet conversion on the Y component of the converted original color image, and a difference storage unit connected with the difference storage unit and used for storing the difference of a wavelet coefficient low-frequency sub-band after the gray watermark image is subjected to wavelet conversion, wherein the inverse wavelet conversion and primary wavelet conversion unit is used for carrying out inverse color pretreatment on the gray watermark image after the inverse color pretreatment and omitting the high-frequency diagonal sub-band part of the inverse color pretreatment and primary wavelet conversion unit, and the difference storage unit is connected with the inverse color pretreatment and primary wavelet conversion unit and used for storing the difference of the wavelet coefficient low-frequency sub-band after the gray watermark image is subjected to wavelet conversion, Other wavelet transform coefficients are multiplied by respective coefficients and then are subjected to sequence ordering together with watermark color reversal preprocessing variables, a sequence ordering unit taking the arrangement sequence as a key, a repeated embedding times calculating unit connected with the sequence ordering unit to calculate the size of the original color image and the gray watermark image, a sequence embedding unit connected with the repeated embedding times calculating unit to embed the sequence in the sequence ordering unit into the diagonal sub-band of the secondary wavelet transform coefficient of the original image according to the repeated times calculated by the repeated embedding times calculating unit, a sequence embedding unit connected with the sequence embedding unit to perform secondary wavelet inverse transform on the color image after embedding the coefficient to obtain Y component, and carry on the inverse transformation and color space conversion unit of the color space conversion together with Cr and Cb weight of the original image;
the watermark extraction part comprises a transformation unit, an extraction unit connected with the transformation unit, and a diagonal sub-band full-zero completion and wavelet inverse transformation unit connected with the extraction unit, wherein the transformation unit performs YUV color space transformation on the image embedded with the watermark, and performs secondary wavelet transformation on a Y component in the image; the extraction unit extracts each sub-band of the gray watermark image and the inverse color preprocessing judgment variable epsilon' from the diagonal high-frequency sub-band part of the secondary wavelet transform according to the key in the sequence ordering unit of the embedded watermark part, then the diagonal sub-band full-zero completion and wavelet inverse transform unit completes the full-zero completion of the wavelet coefficient diagonal sub-band of the gray watermark image, and the obtained sub-band coefficients and the full-zero diagonal sub-band are subjected to wavelet inverse transform once to obtain the extracted gray watermark image.
The digital watermarking method based on the DWT domain replaces the secondary wavelet transform diagonal component sub-band coefficient of the carrier image with the watermark image wavelet coefficient, embeds the watermark information into the diagonal component of the secondary wavelet transform, and uses the replacement watermark wavelet coefficient instead of the common quantization method, thereby realizing blind watermark extraction, improving the stability of the system and ensuring the concealment and safety of the watermark.
Drawings
Fig. 1 is a schematic diagram of the watermark embedding process according to the present invention.
Fig. 2 is a schematic diagram of the watermark extraction process of the present invention.
Fig. 3 is a schematic diagram of the secondary wavelet transform of the original color image according to the present invention.
Fig. 4 is a schematic diagram of wavelet transform of a watermark image.
Fig. 5(a) is an original color image Lena.
Fig. 5(b) shows a grayscale watermark image.
Fig. 6(a) shows a color image Lena after embedding a watermark.
Fig. 6(b) shows an extracted grayscale watermark image.
Fig. 7 is a block diagram of a DWT domain-based digital watermarking system according to another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1 and fig. 2, the DWT domain-based digital watermarking method of the present invention includes embedding of a watermark and extracting of the watermark. The watermark embedding process comprises the following steps: the method comprises the steps of converting an original color image in a YUV color space, performing secondary wavelet transformation on a Y component, performing inverse color preprocessing on a gray watermark image, storing wavelet transformation coefficients of the gray watermark image by difference values, replacing diagonal sub-band coefficients of the secondary wavelet transformation of the Y component by encryption arrangement, and performing secondary wavelet inverse transformation on the replaced Y component to obtain the color image embedded with the watermark. The extraction process of the watermark comprises the following steps: and performing YUV color space conversion on the image embedded with the watermark, performing secondary wavelet transformation on the Y component, obtaining each wavelet component of the gray watermark according to the encryption arrangement sequence in the embedding process, and performing wavelet inverse transformation to obtain the extracted gray watermark image.
As shown in fig. 1, which shows a flowchart of embedding a watermark, the embedding of the watermark includes the following steps:
s10: acquiring an original color image to be embedded with a watermark and a gray watermark image to be embedded;
s12: YUV color space conversion is carried out on the original color image, and secondary wavelet transformation is carried out on the Y component of the original color image after conversion;
s13: carrying out reverse color preprocessing on the gray watermark image, carrying out wavelet transformation on the gray watermark image subjected to the reverse color preprocessing, and discarding a high-frequency diagonal sub-band part;
s14: performing difference value storage on the wavelet coefficient low-frequency sub-band after wavelet transformation of the gray watermark image;
s15: multiplying the low-frequency coefficient stored by the difference value and other wavelet transform coefficients by respective coefficients, then sequencing the low-frequency coefficient and other wavelet transform coefficients together with watermark color reversal preprocessing variables, and taking the sequencing sequence as a key;
s16: determining the number of times of repeated embedding of the coefficient sequence by calculating the sizes of the original color image and the gray watermark image;
s17: embedding the sequence in the step S15 into the diagonal sub-band of the secondary wavelet transform coefficient of the original image according to the repetition times calculated in the step S16;
s18: and performing secondary wavelet inverse transformation on the color image with the embedded coefficient to obtain a Y component, and performing color space transformation together with Cr and Cb components of the original image to obtain an image with the embedded watermark.
Wherein,
in step S12, converting the color space of the original image of size M × N to convert the RGB color space to the YUV color space; the Y component is subjected to a secondary wavelet transform, and the resulting secondary wavelet transform subbands are denoted as LL2, HL2, LH2, and HH2, respectively, as shown in fig. 3.
In step S13, the grayscale watermark image of size m × n to be embedded is subjected to an inversion preprocessing, and whether or not the watermark image has been subjected to the inversion processing is recorded by positive and negative values of an inversion variable ∈. Then, the image after the reverse color processing is subjected to wavelet transform once, and the obtained wavelet-transformed low-frequency coefficient matrix is represented as wLL, and the high-frequency coefficient matrices are represented as wLH, wHL, and wHH, as shown in fig. 4. The gray watermark image is subjected to gray average calculation, and whether to perform the reverse color processing is determined according to whether the average is larger than 127. And the reverse color variable epsilon is a positive integer, if the watermark image is reversed, epsilon is set to be-epsilon, otherwise, the original value is kept. In practical application, the epsilon is 13, so that better effect can be achieved.
In steps S14 and S15, the differences between wLL are stored, and then each component is multiplied by the corresponding coefficient to obtain α · wLL, β · wLH, and γ · wHL. Then, the coefficients and epsilon in alpha wLL, beta wLH and gamma wHL are arranged together to form an mn × 1 sequence, wherein the arrangement order can be used as an encryption means to protect watermark information when extracting the watermark. The difference value storage is a method for storing the difference value of two coefficients before and after replacing the original coefficient value, and obtaining the original value again every delta coefficient values. In practical application, δ is taken to be 4. The permutation order encryption means refers to the permutation order of the coefficients and epsilon in alpha-wLL, beta-wLH and gamma-wHL, and different mn x 1 sequences can be generated through different permutation orders. When the extraction is performed, if the correct order is not known, the correct α · wLL, β · wLH, and γ · wHL cannot be obtained.
In steps S16, S17
The number of times the embedding should be repeated is calculated, and the mn × 1 sequence in step S15 is repeated N' times to replace the HH2 diagonal high-frequency subband of the secondary wavelet transform coefficient of the Y component obtained in step S12.
In step S18, the replaced image is subjected to inverse wavelet transform to obtain a Y component, and then the Y component is combined with the Cr and Cb components to perform color space conversion, so as to obtain an RGB watermark-embedded image.
As shown in fig. 2, which shows a schematic representation of a watermark extraction flow, the watermark extraction includes the following steps:
s20: carrying out YUV color space conversion on the image embedded with the watermark, and carrying out secondary wavelet transformation on a Y component in the image;
s21: the diagonal high-frequency subband portion of the secondary wavelet transform is subjected to the key extraction in step S15 to extract subbands of the grayscale watermark image and an inverse preprocessing decision variable epsilon'.
S22: and completing all zeros of the diagonal sub-bands of the wavelet coefficients of the gray watermark image, and performing wavelet inverse transformation on the obtained average value of the sub-band coefficients and the completely zeroed diagonal sub-bands to obtain the extracted gray watermark image.
Wherein,
in steps S20 and S21, the image embedded with the watermark is subjected to YUV color space conversion, the Y component is subjected to secondary wavelet transform, the HH2 diagonal subband component is extracted, and wavelet coefficients α · wLL ', β · wLH', γ · wHL 'and an inverse color processing parameter ∈' are extracted according to the key in step S15.
In step S22, an average value of all repeated embedding coefficients is obtained according to the number of repeated embedding, the average value corresponding to α · wLL ', β · wLH ', γ · wHL ' is multiplied by coefficients 1/α, 1/β, 1/γ to obtain corresponding watermark wavelet transforms wLL ', wLH ' and wHL ' subband portions, wLL ', wLH ', wHL ' are combined into corresponding watermark wavelet subbands, the non-embedded wHH portions are replaced by all-zero coefficient subbands, and these coefficients are subjected to a wavelet inverse transform to obtain a watermark image. And judging whether the watermark is inversed or not according to whether the average value of the obtained inversed color variable epsilon' is larger than zero or not, and determining whether the watermark image needs to be inversed or not according to the judgment result so as to obtain the gray watermark image needing to be extracted. The coefficients α, β, γ are used to multiply with the sub-band matrix coefficients wl, wLH, and wHL, and in practical applications, α is 0.08, and β is 0.12.
Referring to fig. 5 and fig. 6, the present invention will be described by taking a 512 × 512 Lena color image as an original image and a 32 × 32 grayscale image as a watermark image.
Firstly, embedding a watermark, and converting an original Lena image with the size of M multiplied by N (512 multiplied by 512) into a color space to convert the RGB color space into a YUV color space. The Y component in YUV color space is subjected to secondary wavelet transform to obtain four secondary wavelet transform sub-bands LL2, HL2, LH2 and HH2 with the size of 128 multiplied by 128, and the specific sub-band division is shown in FIG. 3. Calculating the average gray value of the m × n (64 × 64) gray watermark image, wherein the calculation formula is as follows:
wherein, YwAnd (i, j) represents the pixel gray scale value of the gray scale watermark image at the (i, j) position. If it is not
Greater than 127, then Yw(i，j)＝|255-Yw(i, j) |, and assigning the inverse variable as ∈ -13; if it is not
127 or less, Y is maintainedwThe value of (i, j) is unchanged and the inverse variable is assigned ε 13.
And performing wavelet transformation on the image after the reverse color processing, and recording the obtained low-frequency coefficient matrix after the wavelet transformation as wLL and the high-frequency coefficient matrix as wLH, wHL and wHH. The specific sub-band division is shown in fig. 4. The resulting low frequency coefficient matrix wLL is difference-preserved, where a variable δ is defined to represent the interval of the number of coefficients at which the non-difference watermark wavelet coefficients are again obtained. The specific method comprises the following steps: let the original coefficient sequence be Wi(i represents the i-th coefficient value in the sequence), δ is 4, and the difference-preserving coefficient sequence is Wi', then W1′＝W1，W2′＝W1-W2，W3′＝W2-W3，W4′＝W3-W4，W5′＝W5....... The watermark wavelet transform coefficient values are multiplied by corresponding coefficients alpha and gamma of 0.08 and beta and gamma of 0.12 to obtain beta 1 and wLL, beta 2 and wLH and gamma and wHL. Then, the coefficients in α · wLL, β · wLH, γ · wHL and the inverse color variable β 0 are arranged together as a sequence mn × 1. The arrangement sequence can be used as an encryption means, and watermark information can be protected when the watermark is extracted. In this example, α · wLL (1, 1), β · wLH (1, 1), γ · wHL (1, 1), ε, α · wLL (1, 2), β · wLH (1, 2) … … are taken <math>
<mrow>
<mi>α</mi>
<mo>·</mo>
</mrow>
</math> <math>
<mrow>
<mi>β</mi>
<mo>·</mo>
<mi>wLH</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mi>m</mi>
<mn>4</mn>
</mfrac>
<mo>,</mo>
<mfrac>
<mi>n</mi>
<mn>4</mn>
</mfrac>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
</math> <math>
<mrow>
<mi>γ</mi>
<mo>·</mo>
<mi>wHL</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mi>m</mi>
<mn>4</mn>
</mfrac>
<mo>,</mo>
<mfrac>
<mi>n</mi>
<mn>4</mn>
</mfrac>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
</math> ε is arranged (wLL (i, j) indicates the coefficient values wLH (i, j) in row i and column j in the coefficient matrix wLL, wHL (i, j) is the same). The resulting mn × 1 sequence is based on <math>
<mrow>
<mo>(</mo>
<msup>
<mi>N</mi>
<mo>′</mo>
</msup>
<mo>=</mo>
<mfrac>
<mi>M</mi>
<mrow>
<mn>4</mn>
<mi>m</mi>
</mrow>
</mfrac>
<mo>×</mo>
<mfrac>
<mi>N</mi>
<mrow>
<mn>4</mn>
<mi>n</mi>
</mrow>
</mfrac>
<mo>=</mo>
<mrow>
<mo>(</mo>
<mfrac>
<mn>512</mn>
<mrow>
<mn>4</mn>
<mo>×</mo>
<mn>64</mn>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mo>×</mo>
<mrow>
<mo>(</mo>
<mfrac>
<mn>512</mn>
<mrow>
<mn>4</mn>
<mo>×</mo>
<mn>64</mn>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</math> The resulting number of repeated embeddings is calculated, and the sequence is repeated 4 times replacing the HH2 diagonal subband coefficient value of the Y component of the original image. And performing secondary wavelet inverse transformation on the replaced wavelet coefficient image to obtain a Y component, and performing color space transformation by combining the Y component with the previous Cr and Cb components to obtain a watermark embedded image in an RGB (red, green and blue) form.
Secondly, extracting the watermark image, wherein the extraction process of the watermark image comprises the following steps: the water-containing print image is converted from RGB color space to YUV color space, and then the Y color component is subjected to secondary wavelet transformation. The corresponding LL2 ', HL 2', LH2 ', HH 2' were obtained. Wavelet coefficients are extracted according to the encryption arrangement sequence in the watermark embedding process. Since the embedding was repeated 4 times during the execution, α · wLL ', β · wLH ', γ · wHL ' were obtained, and the coefficients were averaged 4 times, whereas ∈ ' was embedded 4096 times instead of wHH in the embedding coefficient position, and similarly, an average of ∈ ' was obtained. The coefficients of wLL ', wLH', wHL 'wavelet transform sub-bands of the extracted watermark image are obtained by multiplying alpha-wLL', beta-wLH ', gamma-wHL' by 1/alpha, 1/beta, 1/gamma, and the part wHH which is not embedded is replaced by the sub-band of all-zero coefficients of equal size. And performing wavelet inverse transformation on the wavelet transformation sub-subband coefficient image to obtain a gray watermark image. According to epsilonAnd whether the average value is larger than zero or not is judged, so that whether the gray watermark image is inversed or not is judged. If reversed, Yw′(i，j)＝|255-Yw' (i, j) |, otherwise the original value is kept unchanged. Thus, the gray watermark image to be extracted is obtained. Wherein Y isw' (i, j) denotes a pixel gradation value of the extracted gradation watermark image at the (i, j) position. The extraction effect of the watermark image is shown in fig. 6 (b).
The results of this example were experimentally tested to evaluate the difference between the embedded watermark picture and the original image using peak signal-to-noise ratio (PSNR). PSNR is defined as:
wherein, I is an original image, I' is an image embedded with a watermark, and M and N are the number of pixels of the length and the width of the image respectively.
In the aspect of evaluating the watermark extraction effect, the similarity evaluation of the extracted watermark and the original watermark is carried out by using the normalized correlation coefficient NC in the test. The corresponding formula is defined as follows:
wherein, W represents the original watermark, W' represents the extracted watermark, and n and m are the number of pixels of the length and width of the watermark image, respectively.
The BMP images are subjected to embedding and extracting tests according to parameter values in the implementation process, the PSNR of the image after embedding the watermark is 37.03, and the NC of the extracted watermark is 0.998, so that the effect is shown in fig. 6 (b). As shown in tables 1, 2 and 3 below, the watermark image extraction effect of the digital watermarking system under JPEG compression, filtering, noise adding and 1/4 clipping attacks is shown. Where table 1 is the JPEG lossy compression test, table 2 is the noise attack test, and table 3 is the median filter and 1/4 clipping test.
TABLE 1
From the results in tables 1, 2 and 3, it can be seen that the digital watermarking method of the present invention has a good attack resistance effect on JPEG attack and noise attack, has good attack resistance capability on median filtering when the filter size is 3 × 3, and has good robustness on cropping attack.
Referring to fig. 7, as another embodiment of the present invention, there is provided a DWT domain-based digital watermarking system including a watermark embedding section and a watermark extracting section. The watermark embedding part comprises an image acquisition unit, a color space conversion and secondary wavelet transformation unit connected with the image acquisition unit, an inverse color preprocessing and primary wavelet transformation unit connected with the color space conversion and secondary wavelet transformation unit, a difference value storage unit connected with the inverse color preprocessing and primary wavelet transformation unit, a sequence sorting unit connected with the difference value storage unit, a repeated embedding time calculation unit connected with the sequence sorting unit, a sequence embedding unit connected with the repeated embedding time calculation unit, and an inverse conversion and color space conversion unit connected with the sequence embedding unit. The image acquisition unit is used for acquiring an original color image to be embedded with a watermark and a gray watermark image to be embedded, and the color space conversion and secondary wavelet transformation unit is used for performing YUV color space conversion on the acquired original color image and performing secondary wavelet transformation on a Y component of the converted original color image; meanwhile, the reverse color preprocessing and the primary wavelet transformation unit reversely preprocesses the gray watermark image, performs primary wavelet transformation on the gray watermark image after the reverse color preprocessing, and omits a high-frequency diagonal sub-band part therein, and then the difference value storage unit performs difference value storage on a wavelet coefficient low-frequency sub-band after the gray watermark image is wavelet transformed; the sequence ordering unit multiplies the low-frequency coefficient stored by the difference value and other wavelet transformation coefficients by respective coefficients, then carries out sequence ordering together with watermark color reversal preprocessing variables, and takes the ordering sequence as a key; the repeated embedding frequency calculation unit determines the times of repeated embedding of the coefficient sequence by calculating the sizes of the original color image and the gray watermark image; the sequence embedding unit embeds the sequence in the sequence ordering unit into an original image secondary wavelet transform coefficient diagonal sub-band according to the repetition times calculated by the repetition embedding time calculating unit, finally, the inverse transform and color space converting unit carries out secondary wavelet inverse transform on the color image embedded with the coefficient to obtain a Y component, and carries out color space conversion together with Cr and Cb components of the original image to obtain an image embedded with the watermark.
The watermark extraction part comprises a transformation unit, an extraction unit connected with the transformation unit, and a diagonal sub-band full-zero completion and wavelet inverse transformation unit connected with the extraction unit. The transformation unit carries out YUV color space transformation on the image embedded with the watermark and carries out secondary wavelet transformation on a Y component in the image; the extraction unit extracts each sub-band of the gray watermark image and an inverse color preprocessing judgment variable epsilon' from the diagonal high-frequency sub-band part of the secondary wavelet transform according to the key in the sequence ordering unit of the embedded watermark part, then the diagonal sub-band full-zero completion and wavelet inverse transform unit completes the full-zero completion of the diagonal sub-band of the wavelet coefficient of the gray watermark image, and the obtained sub-band coefficient and the full-zero diagonal sub-band are subjected to wavelet inverse transform once to obtain the extracted gray watermark image.
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 and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A digital watermarking method based on DWT domain includes watermark embedding and watermark extraction, and is characterized in that:
the embedding of the watermark comprises the following steps:
s10: acquiring an original color image to be embedded with a watermark and a gray watermark image to be embedded;
s12: YUV color space conversion is carried out on the original color image, and secondary wavelet transformation is carried out on the Y component of the original color image after conversion;
s13: carrying out reverse color preprocessing on the gray watermark image, carrying out wavelet transformation on the gray watermark image subjected to the reverse color preprocessing, and discarding a high-frequency diagonal sub-band part;
s14: performing difference value storage on the wavelet coefficient low-frequency sub-band after wavelet transformation of the gray watermark image;
s15: multiplying the low-frequency coefficient subjected to difference value storage and other wavelet transformation coefficients by respective coefficients, then carrying out sequence sequencing together with watermark color reversal preprocessing variables, and taking the sequence as a key;
s16: determining the number of times of repeated embedding of the coefficient sequence by calculating the sizes of the original color image and the gray watermark image;
s17: embedding the sequence in the step S15 into the diagonal sub-band of the secondary wavelet transform coefficient of the original image according to the repetition times calculated in the step S16;
s18: performing secondary wavelet inverse transformation on the color image with the embedded coefficient to obtain a Y component, and performing color space transformation together with Cr and Cb components of the original color image to obtain an image with embedded watermark;
the extraction of the watermark comprises the following steps:
s20: carrying out YUV color space conversion on the image embedded with the watermark, and carrying out secondary wavelet transformation on a Y component in the image;
s21: the diagonal high-frequency subband portion of the secondary wavelet transform is subjected to key extraction in step S15 to extract each subband coefficient of the grayscale watermark image and the inverse pre-processing decision variable epsilon'.
S22: and completing all zeros of the diagonal sub-bands of the wavelet coefficients of the gray watermark image, and performing wavelet inverse transformation on the obtained average value of the sub-band coefficients and the completely zeroed diagonal sub-bands to obtain the extracted gray watermark image.
2. The DWT domain based digital watermarking method of claim 1, wherein: in step S12, converting the color space of the original image to convert the RGB color space into the YUV color space; and performing secondary wavelet transform on the Y component to obtain secondary wavelet transform subbands which are respectively marked as LL2, HL2, LH2 and HH 2.
3. A DWT domain based digital watermarking method according to claims 1 and 2, characterized by: in step S13, an inverse color preprocessing is performed on the grayscale watermark image to be embedded, whether the watermark image has been subjected to the inverse color processing is recorded by positive and negative values of an inverse color variable epsilon, and then a wavelet transform is performed on the image after the inverse color processing, so as to obtain a wavelet-transformed low-frequency coefficient matrix wLL and high-frequency coefficient matrices wLH, wHL, and wHH.
4. A DWT domain based digital watermarking method according to claim 3, wherein: in the steps S14 and S15, the low frequency coefficient matrix wLL is subjected to difference storage, each component is multiplied by the corresponding coefficient to obtain α · wLL, β · wLH, and γ · wHL, and then the coefficients and epsilon in α · wLL, β · wLH, and γ · wHL are arranged together to form a sequence, wherein the arrangement order can be used as an encryption means to protect watermark information when extracting the watermark.
5. The DWT domain based digital watermarking method of claim 4, wherein: the difference value storage is a method for storing the difference values of the front coefficient and the back coefficient to replace the original coefficient values, and the original values are obtained again every delta coefficient values; the permutation order encryption means refers to the permutation order of the coefficients and epsilon in alpha-wLL, beta-wLH and gamma-wHL, and different sequences can be generated through different permutation orders.
6. The DWT domain based digital watermarking method of claim 5, wherein: in the above steps S16, S17, the number of times that embedding should be repeated is calculated, and the sequence arranged in step S15 is repeated N' times to replace the HH2 diagonal high-frequency subband of the secondary wavelet transform coefficient of the Y component obtained in step S12.
7. The DWT domain based digital watermarking method of claim 6, wherein: in step S18, the replaced image is subjected to inverse wavelet transform to obtain a Y component, and then the Y component is combined with the Cr and Cb components to perform color space transformation to obtain an RGB watermark embedded image.
8. The DWT domain based digital watermarking method of claim 7, wherein: in the steps S20 and S21, the image embedded with the watermark is subjected to YUV color space conversion, the Y component of the image is subjected to secondary wavelet transform, the HH2 ' diagonal subband component of the image is extracted, and wavelet coefficients α · wLL ', β · wLH ', γ · wHL ' and the inverse color processing parameter ∈ ' are extracted according to the key in the step S15.
9. The DWT domain based digital watermarking method of claim 8, wherein: in step S22, an average value of all repeated embedding coefficients is obtained according to the number of repeated embedding, the average value corresponding to α · wLL ', β · wLH ', γ · wHL ' is multiplied by coefficients 1/α, 1/β, 1/γ to obtain corresponding watermark wavelet transforms wLL ', wLH ', wHL ', wLL ', wLH ', wHL ' are combined into corresponding watermark wavelet subbands, the non-embedded wHH part is replaced by an all-zero coefficient subband, and the coefficients are subjected to a wavelet inverse transform to obtain a watermark image.
10. A digital watermarking system based on DWT domain includes a watermark embedding part and a watermark extracting part, and is characterized in that:
the watermark embedding section includes: an image obtaining unit for obtaining an original color image to be embedded with a watermark and a gray watermark image to be embedded, a color space conversion and secondary wavelet conversion unit connected with the image obtaining unit and used for YUV color space conversion of the obtained original color image and carrying out secondary wavelet conversion on the Y component of the converted original color image, and a difference storage unit connected with the difference storage unit and used for storing the difference of a wavelet coefficient low-frequency sub-band after the gray watermark image is subjected to wavelet conversion, wherein the inverse wavelet conversion and primary wavelet conversion unit is used for carrying out inverse color pretreatment on the gray watermark image after the inverse color pretreatment and omitting the high-frequency diagonal sub-band part of the inverse color pretreatment and primary wavelet conversion unit, and the difference storage unit is connected with the inverse color pretreatment and primary wavelet conversion unit and used for storing the difference of the wavelet coefficient low-frequency sub-band after the gray watermark image is subjected to wavelet conversion, Other wavelet transform coefficients are multiplied by respective coefficients and then are subjected to sequence ordering together with watermark color reversal preprocessing variables, a sequence ordering unit taking the arrangement sequence as a key, a repeated embedding times calculating unit connected with the sequence ordering unit to calculate the size of the original color image and the gray watermark image, a sequence embedding unit connected with the repeated embedding times calculating unit to embed the sequence in the sequence ordering unit into the diagonal sub-band of the secondary wavelet transform coefficient of the original image according to the repeated times calculated by the repeated embedding times calculating unit, a sequence embedding unit connected with the sequence embedding unit to perform secondary wavelet inverse transform on the color image after embedding the coefficient to obtain Y component, and carry on the inverse transformation and color space conversion unit of the color space conversion together with Cr and Cb weight of the original image;
the watermark extraction part comprises a transformation unit, an extraction unit connected with the transformation unit, and a diagonal sub-band full-zero completion and wavelet inverse transformation unit connected with the extraction unit, wherein the transformation unit performs YUV color space transformation on the image embedded with the watermark, and performs secondary wavelet transformation on a Y component in the image; the extraction unit extracts each sub-band of the gray watermark image and the inverse color preprocessing judgment variable epsilon' from the diagonal high-frequency sub-band part of the secondary wavelet transform according to the key in the sequence ordering unit of the embedded watermark part, then the diagonal sub-band full-zero completion and wavelet inverse transform unit completes the full-zero completion of the wavelet coefficient diagonal sub-band of the gray watermark image, and the obtained sub-band coefficients and the full-zero diagonal sub-band are subjected to wavelet inverse transform once to obtain the extracted gray watermark image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110400961 CN102496135B (en) | 2011-12-06 | 2011-12-06 | Deadweight tonnage (DWT) domain-based digital watermark method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110400961 CN102496135B (en) | 2011-12-06 | 2011-12-06 | Deadweight tonnage (DWT) domain-based digital watermark method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102496135A true CN102496135A (en) | 2012-06-13 |
| CN102496135B CN102496135B (en) | 2013-10-30 |
Family
ID=46187957
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110400961 Active CN102496135B (en) | 2011-12-06 | 2011-12-06 | Deadweight tonnage (DWT) domain-based digital watermark method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102496135B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104252694A (en) * | 2014-10-17 | 2014-12-31 | 苏州工业职业技术学院 | Embedding method and extraction method for image watermark based on DWT (Discrete Wavelet Transform) |
| CN107240060A (en) * | 2017-05-25 | 2017-10-10 | 上海地铁第运营有限公司 | Security against fire based on dynamic digital watermark checks picture method for anti-counterfeit and system |
| CN107426172A (en) * | 2017-06-05 | 2017-12-01 | 国政通科技股份有限公司 | The matching method for pushing and device of a kind of identity information |
| CN108230224A (en) * | 2017-12-29 | 2018-06-29 | 湖北工业大学 | A kind of Image Watermarking and extracting method for colour blindness image after correction |
| CN108564519A (en) * | 2017-12-22 | 2018-09-21 | 重庆邮电大学 | A kind of Watermarking for Color Images method based on different transform domain |
| CN109379508A (en) * | 2018-09-18 | 2019-02-22 | 湖北工程学院 | A kind of image protection method and storage medium of joint encryption and compression |
| CN110930288A (en) * | 2019-11-21 | 2020-03-27 | 上海鹰瞳医疗科技有限公司 | Method and device for adding and extracting watermark in fundus image |
| CN110969563A (en) * | 2018-09-28 | 2020-04-07 | 浙江大学 | Watermark embedding method, watermark extracting method, watermark embedding device, watermark extracting equipment and data processing method |
| CN112083894A (en) * | 2019-06-14 | 2020-12-15 | 森大（深圳）技术有限公司 | Method, device, device and medium for automatically setting the amount of white ink on an image pixel-by-pixel basis |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101272485A (en) * | 2008-04-07 | 2008-09-24 | 中国科学院计算技术研究所 | A digital watermark superposition and detection method for digital film copyright protection |
| US20090060257A1 (en) * | 2007-08-29 | 2009-03-05 | Korea Advanced Institute Of Science And Technology | Watermarking method resistant to geometric attack in wavelet transform domain |
| CN102147912A (en) * | 2011-03-30 | 2011-08-10 | 北京航空航天大学 | Adaptive difference expansion-based reversible image watermarking method |
-
2011
- 2011-12-06 CN CN 201110400961 patent/CN102496135B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090060257A1 (en) * | 2007-08-29 | 2009-03-05 | Korea Advanced Institute Of Science And Technology | Watermarking method resistant to geometric attack in wavelet transform domain |
| CN101272485A (en) * | 2008-04-07 | 2008-09-24 | 中国科学院计算技术研究所 | A digital watermark superposition and detection method for digital film copyright protection |
| CN102147912A (en) * | 2011-03-30 | 2011-08-10 | 北京航空航天大学 | Adaptive difference expansion-based reversible image watermarking method |
Non-Patent Citations (2)
| Title |
|---|
| 何冰: "一种基于DWT域的彩色图像数字水印算法", 《计算机与数字工程》 * |
| 刘奕: "基于DWT 的数字图像水印算法研究", 《计算机时代》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104252694A (en) * | 2014-10-17 | 2014-12-31 | 苏州工业职业技术学院 | Embedding method and extraction method for image watermark based on DWT (Discrete Wavelet Transform) |
| CN107240060A (en) * | 2017-05-25 | 2017-10-10 | 上海地铁第运营有限公司 | Security against fire based on dynamic digital watermark checks picture method for anti-counterfeit and system |
| CN107426172A (en) * | 2017-06-05 | 2017-12-01 | 国政通科技股份有限公司 | The matching method for pushing and device of a kind of identity information |
| CN108564519A (en) * | 2017-12-22 | 2018-09-21 | 重庆邮电大学 | A kind of Watermarking for Color Images method based on different transform domain |
| CN108564519B (en) * | 2017-12-22 | 2021-11-16 | 重庆邮电大学 | Color image digital watermarking method based on multiple transform domains |
| CN108230224A (en) * | 2017-12-29 | 2018-06-29 | 湖北工业大学 | A kind of Image Watermarking and extracting method for colour blindness image after correction |
| CN109379508B (en) * | 2018-09-18 | 2019-09-06 | 湖北工程学院 | A joint encryption and compression image protection method and storage medium |
| CN109379508A (en) * | 2018-09-18 | 2019-02-22 | 湖北工程学院 | A kind of image protection method and storage medium of joint encryption and compression |
| CN110969563A (en) * | 2018-09-28 | 2020-04-07 | 浙江大学 | Watermark embedding method, watermark extracting method, watermark embedding device, watermark extracting equipment and data processing method |
| CN110969563B (en) * | 2018-09-28 | 2022-06-24 | 浙江大学 | Watermark embedding and watermark extraction method, device and device and data processing method |
| CN112083894A (en) * | 2019-06-14 | 2020-12-15 | 森大（深圳）技术有限公司 | Method, device, device and medium for automatically setting the amount of white ink on an image pixel-by-pixel basis |
| CN112083894B (en) * | 2019-06-14 | 2024-12-17 | 森大（深圳）技术有限公司 | Method, device, equipment and medium for automatically setting white ink quantity of image pixel-by-pixel point |
| CN110930288A (en) * | 2019-11-21 | 2020-03-27 | 上海鹰瞳医疗科技有限公司 | Method and device for adding and extracting watermark in fundus image |
| CN110930288B (en) * | 2019-11-21 | 2023-09-08 | 上海鹰瞳医疗科技有限公司 | Method and apparatus for adding and extracting watermark in fundus image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102496135B (en) | 2013-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102496135B (en) | Deadweight tonnage (DWT) domain-based digital watermark method and system | |
| Dixit et al. | A review on digital image watermarking techniques | |
| Assini et al. | A robust hybrid watermarking technique for securing medical image | |
| CN100394443C (en) | A Reversible Watermarking Method for Image Authentication | |
| Divecha et al. | Implementation and performance analysis of DCT-DWT-SVD based watermarking algorithms for color images | |
| Singh et al. | A secure and robust block based DWT-SVD image watermarking approach | |
| CN110189243B (en) | Color image robust watermarking method based on tensor singular value decomposition | |
| Qu et al. | Optical color watermarking based on single-pixel imaging and singular value decomposition in invariant wavelet domain | |
| Khare et al. | Robust digital image watermarking scheme based on RDWT-DCT-SVD | |
| Shi et al. | An RST invariant image watermarking scheme using DWT-SVD | |
| Kekre et al. | Hybrid watermarking of color images using DCT-Wavelet, DCT and SVD | |
| Xie et al. | Toward improved wavelet-based watermarking using the pixel-wise masking model | |
| Sharma et al. | Robust technique for steganography on Red component using 3-DWT-DCT transform | |
| CN110120004A (en) | A kind of digital watermark method based on adaptive piecemeal and compressed sensing | |
| TW200849135A (en) | A robust watermarking method for high dynamic range images | |
| Moniruzzaman et al. | Robust RGB color image watermarking scheme based on DWT-SVD and chaotic system | |
| Chang et al. | A multipurpose wavelet-based image watermarking | |
| Kapadia et al. | Secured Reversible matrix embedding based on dual image using Integer wavelet and Arnold Transform | |
| Hu et al. | A blind watermarking algorithm for color image based on wavelet transform and Fourier transform | |
| Al-Shayea et al. | On the efficiency evaluation of a novel scheme based on Daubechies wavelet for watermarking in 5G | |
| Sever et al. | Block size analysis for discrete wavelet watermarking and embedding a vector image as a watermark. | |
| Ramaiya et al. | Digital security using watermarking techniques via discrete wavelet transform | |
| Joseph et al. | Singular Value Decomposition based wavelet domain watermarking | |
| Rao et al. | A Blind approach of QR Code based color Image watermarking using DWT | |
| Silja et al. | A watermarking algorithm based on contourlet transform and nonnegative matrix factorization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB02 | Change of applicant information |
Address after: Hangzhou City, Zhejiang Province, 310030 Xihu District science and Technology Park of West Lake eight West Road No. 2 block G Applicant after: Enjoyor Co., Ltd. Applicant after: Zhejiang Enjoyor Traffic Technology Co., Ltd. Applicant after: ZHEJIANG YINJIANG WISDOM TRANSPORT GROUP CO., LTD. Address before: Hangzhou City, Zhejiang Province, 310030 Xihu District science and Technology Park of West Lake eight West Road No. 2 block G Applicant before: Enjoyor Co., Ltd. Applicant before: Zhejiang Enjoyor Traffic Technology Co., Ltd. Applicant before: Hangzhou Yinjiang Wisdom Traffic Technology Co.,Ltd. |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 310030 block g, No. 2, Xiyuan 8th Road, Xihu science and technology economic Park, Xihu District, Hangzhou City, Zhejiang Province Patentee after: Yinjiang Technology Co.,Ltd. Patentee after: ZHEJIANG YINJIANG TRAFFIC TECHNOLOGY Co.,Ltd. Patentee after: ZHEJIANG YINJIANG ZHIHUI TRAFFIC GROUP Co.,Ltd. Address before: 310030 block g, No. 2, Xiyuan 8th Road, Xihu science and technology economic Park, Xihu District, Hangzhou City, Zhejiang Province Patentee before: ENJOYOR Co.,Ltd. Patentee before: ZHEJIANG YINJIANG TRAFFIC TECHNOLOGY Co.,Ltd. Patentee before: ZHEJIANG YINJIANG ZHIHUI TRAFFIC GROUP Co.,Ltd. |