CN110599390A - Watermark embedding method based on Curvelet and RSA sequence - Google Patents
Watermark embedding method based on Curvelet and RSA sequence Download PDFInfo
- Publication number
- CN110599390A CN110599390A CN201910871201.1A CN201910871201A CN110599390A CN 110599390 A CN110599390 A CN 110599390A CN 201910871201 A CN201910871201 A CN 201910871201A CN 110599390 A CN110599390 A CN 110599390A
- Authority
- CN
- China
- Prior art keywords
- watermark
- coefficient matrix
- matrix
- curvelet
- image
- 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
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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Editing Of Facsimile Originals (AREA)
Abstract
The application provides a watermark embedding method based on Curvelet-DCT and RSA sequences, which comprises the following steps: acquiring image data; respectively carrying out full-image Curvelet transformation and DCT (discrete cosine transformation) transformation on the image data to obtain a second coefficient matrix; taking a submatrix with preset row and column numbers of the second coefficient matrix to obtain a third coefficient matrix; using perceptual hash and performing dimensionality reduction operation to obtain a visual feature vector; generating an encrypted watermark by using a pseudorandom sequence; and performing exclusive OR operation on the visual feature vector and the encrypted watermark bit by bit to insert the encrypted watermark into the medical image and obtain a logic key. By combining the watermarking technology with RSA large number decomposition encryption, the method has double confidentiality effects and can effectively protect the data security of private information of patients. The application also provides a watermark embedding system based on Curvelet-DCT and RSA sequences, a computer readable storage medium and a terminal, which have the beneficial effects.
Description
Technical Field
The present application relates to the field of image processing, and in particular, to a watermark embedding method and related apparatus based on Curvelet-DCT and RSA sequences.
Background
Medical development is gradually shifting from traditional medicine to telemedicine, which enables large amounts of medical images to be transmitted and shared in a network; in order to solve the problems that the medical image is possibly tampered in the transmission and sharing process and stolen, the original medical image needs to be processed; the zero watermark technology and the perceptual hash technology are combined to serve as the safety technology of information safety, so that the safety transmission can be guaranteed, the information authentication can be realized, and the method has very important application in practical application.
The digital watermarking technology is originally used for copyright protection of digital media, and the characteristics of invisibility, robustness and the like of the digital watermarking are utilized to hide personal information of a patient in a medical image of the patient so as to ensure the safe transmission of the personal information on the Internet. Therefore, in the case where digital images are widely used in network transmission, research on digital watermarking algorithms for medical images becomes extremely important; through the unique characteristics of invisibility, robustness and the like, the privacy of the patient is protected, and the zero watermark can avoid the tampered medical data, so that the relevant patient information required by remote medical diagnosis is realized.
However, the existing watermarking algorithm still has the problem of poor robustness in the face of geometric attack, and the research in the field of curvelet is less. Therefore, a technical problem to be solved by those skilled in the art is urgently needed how to achieve effective and reliable embedding of watermarks in medical images.
Disclosure of Invention
The application aims to provide a watermark embedding method based on Curvelet-DCT and RSA sequences, a watermark embedding system based on Curvelet-DCT and RSA sequences, a computer readable storage medium and a terminal, which can realize watermark embedding of medical images and protect privacy of patients and data security of the medical images.
In order to solve the technical problems, the application provides a watermark embedding method based on Curvelet-DCT and RSA sequences, and the specific technical scheme is as follows:
acquiring image data of a medical image;
carrying out full-image Curvelet transformation on the image data to obtain a first coefficient matrix;
performing DCT (discrete cosine transform) transformation on the first coefficient matrix at a Coarse layer to obtain a second coefficient matrix;
taking a submatrix with preset row and column numbers of the second coefficient matrix to obtain a third coefficient matrix, and calculating an average value of the third coefficient matrix;
converting the third coefficient matrix into a binary matrix by using perceptual hashing;
performing dimensionality reduction operation on the binary matrix to obtain a visual feature vector;
generating an encrypted watermark by using a pseudorandom sequence;
and performing exclusive OR operation on the visual feature vector and the encrypted watermark bit by bit to insert the encrypted watermark into the medical image and obtain a logic key for watermark extraction.
Wherein converting the third coefficient matrix into a binary matrix using perceptual hashing comprises:
and comparing the elements in the third coefficient matrix with the average value line by line, assigning 1 to the elements which are greater than or equal to the average value, and assigning 0 to the elements which are smaller than the average value to obtain a binary matrix.
Before generating the encrypted watermark by using the pseudorandom sequence, the method further comprises the following steps:
the pseudo-random sequence is generated using the RSA algorithm.
Wherein generating the encrypted watermark using the pseudorandom sequence comprises:
obtaining a num function by using a sequencing function for the pseudo-random sequence;
and scrambling and encrypting the binary watermark by using the num function to obtain an encrypted watermark.
The application also provides a watermark embedding system based on Curvelet-DCT and RSA sequences, which comprises:
an acquisition module for acquiring image data of a medical image;
the first transformation module is used for carrying out full-image Curvelet transformation on the image data to obtain a first coefficient matrix;
the second transformation module is used for performing DCT transformation on the first coefficient matrix at a Coarse layer to obtain a second coefficient matrix;
the third transformation module is used for taking a sub-matrix with the preset row and column number of the second coefficient matrix to obtain a third coefficient matrix and calculating the average value of the third coefficient matrix;
the binary conversion module is used for converting the third coefficient matrix into a binary matrix by using perceptual hash;
the eigenvector determining module is used for performing dimensionality reduction operation on the binary matrix to obtain a visual eigenvector;
the watermark generation module is used for generating an encrypted watermark by utilizing the pseudorandom sequence;
and the watermark embedding module is used for carrying out XOR operation on the visual feature vector and the encrypted watermark bit by bit so as to insert the encrypted watermark into the medical image and obtain a logic key for extracting the watermark.
The third variation module is specifically a module configured to compare the elements in the third coefficient matrix with the average value line by line, assign 1 to an element greater than or equal to the average value, and assign 0 to an element smaller than the average value, thereby obtaining a binary matrix.
Wherein, still include:
and the sequence generating module is used for generating the pseudo-random sequence by utilizing an RSA algorithm.
Wherein the watermark generation module comprises:
the sequence ordering unit is used for obtaining a num function by using an ordering function for the pseudo-random sequence;
and the watermark encryption unit is used for scrambling and encrypting the binary watermark by using the num function to obtain an encrypted watermark.
The present application also provides a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the watermark embedding method as described above.
The present application further provides a terminal, including a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the watermark embedding method when calling the computer program in the memory.
The watermark embedding method based on the Curvelet-DCT and RSA sequences provided by the application has the following specific technical scheme: acquiring image data of a medical image; carrying out full-image Curvelet transformation on the image data to obtain a first coefficient matrix; performing DCT (discrete cosine transform) transformation on the first coefficient matrix at a Coarse layer to obtain a second coefficient matrix; taking a submatrix with preset row and column numbers of the second coefficient matrix to obtain a third coefficient matrix, and calculating an average value of the third coefficient matrix; converting the third coefficient matrix into a binary matrix by using perceptual hashing; performing dimensionality reduction operation on the binary matrix to obtain a visual feature vector; generating an encrypted watermark by using a pseudorandom sequence; and performing exclusive OR operation on the visual feature vector and the encrypted watermark bit by bit to insert the encrypted watermark into the medical image and obtain a logic key for watermark extraction.
The method comprises the steps of extracting a medical image visual characteristic vector resisting geometric attack from a Curvelet-DCT coefficient based on full-image Curvelet transformation and Coarse layer DCT transformation; the encrypted watermark is embedded into the encrypted medical image, and the common watermark technology is combined with RSA large number decomposition encryption, so that the anti-geometric and conventional attack of the digital watermark is realized, the watermark image also has double confidentiality effects, and the watermark image has better robustness in the aspects of conventional attack and geometric attack, and can effectively protect the data security of private information of a patient. The application also provides a watermark embedding system based on Curvelet-DCT and RSA sequences, a computer readable storage medium and a terminal, which have the beneficial effects and are not described herein again.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a method for embedding a watermark in an image according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of an image watermark embedding system according to an embodiment of the present disclosure.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
At present, the main reasons of poor geometric attack resistance of most medical image watermarking algorithms are as follows: the slight geometric transformation of medical images often results in large changes in pixel or transform coefficient values, which makes the embedded watermark easily attacked. Therefore, if the feature vector reflecting the geometric characteristics of the medical image can be found, when the image is subjected to small geometric transformation, the feature value of the image is not subjected to obvious mutation, and the image can be watermarked through comparison of the feature vector, so that the watermark information authentication is completed.
It is to be readily understood that the image data is not particularly limited as long as it is an image requiring watermark embedding. The image data mainly refers to data carrying image content information, structural information and the like, and the structural information may include an image format, an image size and the like, and may further include pixel gray scale values and a peak signal-to-noise ratio and the like of the medical image. The peak signal-to-noise ratio is an engineering term representing the ratio of the maximum possible power of a signal and the power of destructive noise affecting its representation accuracy, and is generally used as an objective evaluation criterion for the quality of medical images.
Referring to fig. 1, fig. 1 is a flowchart of a method for embedding a watermark in an image according to an embodiment of the present application, where the method includes:
s101: acquiring image data of a medical image;
s102: carrying out full-image Curvelet transformation on the image data to obtain a first coefficient matrix;
s103: performing DCT (discrete cosine transform) transformation on the first coefficient matrix at a Coarse layer to obtain a second coefficient matrix;
s104: taking a submatrix with preset row and column numbers of the second coefficient matrix to obtain a third coefficient matrix, and calculating an average value of the third coefficient matrix;
s105: converting the third coefficient matrix into a binary matrix by using perceptual hashing;
s106: performing dimensionality reduction operation on the binary matrix to obtain a visual feature vector;
s107: generating an encrypted watermark by using a pseudorandom sequence;
s108: and performing exclusive OR operation on the visual feature vector and the encrypted watermark bit by bit to insert the encrypted watermark into the medical image and obtain a logic key for watermark extraction.
For a more detailed and clear description of the above process, the following description will be given with corresponding specific symbols. First, steps S101 to S106 are intended to perform feature extraction on the medical image I (I, j) in the transform domain, obtaining a visual feature vector V (I, j) of the image.
In S102, carrying out full-image Curvelet transformation on the original medical image I (I, j) to obtain a first coefficient matrix D0(I, j); i.e., D0(I, j) ═ currvelet (I, j));
in S103, performing DCT (discrete cosine transform) on the coefficient matrix D0(i, j) in a Coarse layer to obtain a second coefficient matrix D1(i, j); namely D1(i, j) ═ DCT (D0(i, j))
In S104, a submatrix with a preset number of rows and columns of the coefficient matrix D1(i, j) is taken, where the preset number of rows and columns is not limited, and the embodiment of the present application is described with 4 rows and 8 columns as an example, but a person skilled in the art may select other numbers of rows and columns according to actual situations. Then, calculating an average value 0 of the first 4 rows and the eight rows to obtain a third coefficient matrix D2(i, j);
i.e., D2(i, j) ═ D1(i, j) (first 4 rows, first eight columns)
In S105, the third coefficient matrix D2(i, j) is converted into a binary matrix K (i, j) by perceptual hashing, specifically, the elements of the coefficient matrix D2(i, j) are compared with the average value average0 line by line, the number greater than or equal to average0 is assigned as "1", and the rest are assigned as "0", so that the binary matrix K (i, j) is obtained.
In S106, obtaining a visual feature vector V (j) through dimensionality reduction operation on the binary matrix K (i, j);
it is also typically necessary to generate a pseudo-random sequence according to the RSA algorithm before performing S107. Specifically, two different prime numbers p and q are randomly selected, and an euler function of n which is p × q and n is calculatedThe following conditions are satisfied: d is prime number, (d × e) mod (phi (n)) ═ 1, e<Phi (n) and the greatest common factor of e and phi (n) is 1. There is a one-to-one correspondence of c and a (ranging from 2 to n-1) and c ═ ad) mod (n). In this way, a non-repeating pseudorandom sequence X (i) of length n-2 is generated. It should be noted that the pseudo-random sequence x (i) needs to be reserved for later watermark decryption.
Using the sorting function sort () on the pseudorandom sequence will have two return values, one being a return num function and the other being a function Y (i, j) sorted from small to large. Obtaining a binary watermark S (i, j) from the original watermark W (i, j), and scrambling and encrypting the binary watermark bit by using a num () function. Obtaining an encrypted watermark EW (i, j); ew (i) is S (num (i)).
Performing bitwise XOR operation on the feature vector V (j) and the encrypted watermark EW (i, j), and embedding the watermark into an encrypted image to obtain a logic Key Key (i, j);
and storing Key (i, j) which is used for extracting the watermark later. The Key (i, j) is used as a Key to apply to a third party, so that ownership and use right of the original medical image can be obtained, and the purpose of protecting the medical image is achieved.
The method comprises the steps of extracting a medical image visual characteristic vector resisting geometric attack from a Curvelet-DCT (discrete cosine transform) transform coefficient based on the whole-image Curvelet transform and Coarse layer DCT; the encrypted watermark is embedded into the encrypted medical image, and the common watermark technology is combined with RSA large number decomposition encryption, so that the anti-geometric and conventional attack of the digital watermark is realized, the watermark image also has double confidentiality effects, and the watermark image has better robustness in the aspects of conventional attack and geometric attack, and can effectively protect the data security of private information of a patient.
The following further describes the Curvelet transform, Coarse layer DCT transform and RSA algorithm in the process of the above embodiment:
curvelet transformation has better direction identification capability; the edge of the image can be effectively expressed, such as geometric characteristics of curves, straight lines and the like; the characteristics of multi-resolution (band-pass property), locality and anisotropy are provided;
the Curvelet coefficient can be obtained by the following formula, namely the inner product of the signal and the wavelet function:
in the frequency domain, the Curvelet-based support interval behaves as a "wedge," and in the spatial domain, a Cartesian grid is used. This "wedge" support region is in fact a manifestation of "directionality", and therefore the base is said to have "anisotropy", and in wedge partitions, the larger Curvelet coefficient is obtained only when the proximity base overlaps, i.e., its orientation matches, the geometry of the singularity feature. Curvelet decomposition is carried out on the medical image, and a low-frequency sub-band coefficient and high-frequency sub-band coefficients of all dimensions and all directions can be obtained. The low-frequency subband coefficient can well represent the characteristics of the medical image, and the high-frequency subband coefficient reflects important information such as details, textures and the like of each direction of the image.
The row m and the column n of the low-frequency sub-band are respectively
In the above two formulas, s represents the scale size of Curvelet decomposition, and the original image size is M × N. The decomposition scale s will influence the effect of the extraction of the medical image. If the number of layers of decomposition is too small, the amount of data is not compressed well and there will be much redundancy. If the number of layers of decomposition is too large, the features of a part of the image are lost. The decomposition scale s of Curvelet is defined as
Experiments prove that the number of Curvelet decomposition layers can better meet the requirement of feature extraction.
The principle of Discrete Cosine Transform (DCT) works by dividing an image into portions of different frequencies, including low, high and medium frequency coefficients. The discrete cosine transform is an orthogonal transform based on real numbers. The DCT domain has small calculation amount and strong energy concentration characteristic: most natural signals (including sound and images) concentrate energy in a low-frequency part after discrete cosine transform, are easy to extract visual feature vectors, and are compatible with international popular data compression standards (JPEG, MPEG, H261/263) and are convenient to realize in a compression domain. The method is widely applied to signal processing and image processing at present, and is also well applied to watermarking. The two-dimensional discrete cosine transform (2D-DCT) formula is as follows:
u＝0,1,...,M-1；v＝0,1,...,N-1；
wherein the content of the first and second substances,
wherein, x, y are spatial sampling frequency domains; u, v are frequency domain sample values, which are typically represented by a square matrix of pixels in digital image processing, i.e., M ═ N.
If a sequence is, on the one hand, predeterminable and is produced and reproduced repeatedly; on the one hand, it has the random property (i.e. statistical property) of a random sequence, so called pseudo-random sequence. Pseudorandom sequences have found wide application in cryptography, including communications cryptography, communications systems, and in radar signal design.
The RSA algorithm is one of the three major cryptographic algorithms (RSA, MD5, DES), and is an asymmetric encryption algorithm. Until now, RSA is the most widely used "asymmetric encryption algorithm", by which is meant that the encryption key does not coincide with the decryption key. The RSA algorithm essentially encrypts using the principle that large numbers are difficult to factorize. The whole RSA algorithm comprises three steps of key generation, cipher exchange and decryption transformation. The first step is key generation, which is used as a key to encrypt and decrypt data.
Specifically, the method comprises the following steps:
a) two prime numbers p and q are randomly selected, and the product n of the two is calculated to be p × q, wherein theoretically, the larger p and q are, the safer the operation is.
(b) Calculating the euler function of n:
φ(n)＝φ(p×q)＝φ(p-1)×φ(q-1)＝(p-1)(q-1)
wherein the Euler function phi (n) represents the number of numbers which form a relatively prime relationship with n in positive integers less than or equal to n.
(c) Randomly selecting an integer e, 1< e < phi (n), and making e and phi (n) be relatively prime integers.
(d) Calculating de;
so far, the public and private keys of the RSA algorithm have been generated. Wherein, (n, e) is public key, and (n, d) is private key, also can exchange.
The second step is encryption, where the actual process is performing a conversion of plaintext m into ciphertext c. In this process, the public key (n, e) and the formula c ≡ m are appliede(mod n) is complete.
The third step is decryption, where the encrypted text c is restored to m. In this process the private key (n, d) and the formula m ≡ c are appliedd(mod n)。
Based on the above embodiment, there should be a corresponding watermark extraction process after watermark embedding. It is easy to understand that the extraction and embedding processes of the watermark are inverse processes, and the extraction process and the embedding process are closely related, and the present application further provides a watermark extraction method for an image, which may specifically include:
s201: carrying out full-image Curvelet transformation on the medical image I '(I, j) to be extracted to obtain a fourth coefficient matrix D0' (I, j); d0'(I, j) ═ currvelet (I' (I, j));
s202: DCT transformation is carried out on the fourth coefficient matrix D0'(i, j) at the Coarse layer to obtain a fifth coefficient matrix D1' (i, j),
namely D1'(i, j) ═ DCT (D0' (i, j));
s203: the first 4 rows and 8 columns of the fifth coefficient matrix D1'(i, j) (corresponding to the first 4 rows and 8 columns adopted in the above embodiment, the first 4 rows and 8 columns are also taken as an example in the embodiment of the present application) are taken to obtain a sixth coefficient matrix D2' (i, j). Namely, the number of the preset rows and columns obtained in the extraction process and the embedding process is the same. Calculating an average value 1 of the sixth coefficient matrix;
i.e., D2'(i, j) ═ D1' (i, j) (first 4 rows, first eight columns)
S204: converting the coefficient matrix D2'(i, j) into a binary matrix K' (i, j) by perceptual hashing; specifically, the elements of the coefficient matrix D2'(i, j) are compared with the average value average1 line by line, the number greater than or equal to average1 is assigned as "1", and the rest are assigned as "0", so as to obtain a binary matrix K' (i, j);
s205: obtaining a visual characteristic vector V '(j) of the medical image to be extracted through dimensionality reduction operation on the binary matrix K' (i, j);
s206: extracting a watermark EW' (i, j);
carrying out bitwise XOR operation on the feature vector V '(j) of the encrypted image to be detected and the logic Key Key (i, j), and extracting an encrypted watermark EW' (i, j);
it can be seen that, in the embodiment of the application, only the Key (i, j) is needed when the watermark is extracted, and the original image is not needed to participate, so that the method is a blind watermark extraction algorithm.
Further, after the encrypted watermark is extracted, the encrypted watermark needs to be restored.
12) Obtaining a binary watermark matrix S' (i, j) according to the encrypted watermark;
obtaining a binary watermark matrix S' (i, j) by the extracted encrypted watermark by using the same method as the watermark scrambling encryption;
13) restoring the extracted encrypted watermark by using a binary watermark matrix S' (i, j)
Combining the binary watermark matrix S '(i, j) with the pseudo-random sequence X (i) and num function reserved in the process of the embodiment to obtain a restored watermark W' (i, j) bit by bit;
W'(num(i))＝S'(i)
by calculating the correlation coefficient NC of W (i, j) and W' (i, j), the ownership of the medical image and the embedded watermark information can be determined, thereby realizing the extraction of the watermark.
Normalization coefficient:
in the embodiment of the present application, the normalized correlation coefficient between the recovered watermark W' (i, j) and the original watermark W (i, j) can be evaluated by using the following formula:
in the embodiment of the application, the medical image quality can be evaluated by using a peak signal-to-noise ratio formula PSNR (dB) as follows:
PSNR represents the degree of distortion of the original image. Wherein, I (I, j) is the gray value of the pixel point with the coordinate (I, j) in the original image, I' (I, j) is the gray value of the pixel point with the coordinate (I, j) in the watermark image, and M and N are the pixel values of the row and column of the image respectively.
At present, the main reasons of poor geometric attack resistance of most medical image watermarking algorithms are as follows: the slight geometric transformation of medical images often results in large changes in pixel or transform coefficient values, which makes the embedded watermark easily attacked. If a feature vector reflecting the geometric characteristics of the medical image can be found, the feature value of the image basically does not have obvious mutation when the image has small geometric transformation. After Curvelet-DCT transformation of a large number of images, data (low-if) observations show that when performing medical image processing according to the algorithm of the present invention, a common geometric transformation is performed on a medical image, the low-if coefficients may change in magnitude, but the sign of the coefficients remains substantially unchanged. According to the human visual characteristics (HVS), the low-intermediate frequency signals have large influence on human vision and represent the main features of the image, so that the low-intermediate frequency coefficient symbol sequence of the medical image is selected as a visual feature vector.
Some experimental data were selected as shown in table 1, and the original medical image used for the test was image one, which is a slice image of the brain. Column 1 in table 1 shows the type of medical image attack, and columns 4 to 13 are C (1,1) to C (1,10) taken from the coefficient matrix after Curvelet-DCT transform, for a total of 10 low-if coefficients. Wherein the coefficient C (1,1) represents the dc component value of the medical image. As shown in table 1, for conventional attacks, these low-if coefficient values remain substantially unchanged, and are approximately equal to the original medical image values; for geometric attack, part of coefficients are greatly changed, but it can be found that when the medical image is subjected to geometric attack, the magnitude of part of Curvelet-DCT low-intermediate frequency coefficients is changed, but the sign of the part of Curvelet-DCT low-intermediate frequency coefficients is basically unchanged. Let coefficients positive and zero be denoted by "1" and negative values by "0", then for the original medical image, the C (1,1) -C (1,10) coefficients in the Curvelet-DCT coefficient matrix correspond to the coefficient symbol sequence: "1100001011". Observing the column, the symbol sequence and the original image can be similar whether the conventional attack or the geometric attack, and the normalized correlation coefficient with the original medical image is large (the NC value of the normalized correlation coefficient is 1).
TABLE 1 CURVEET-DCT coefficient variation of original medical image under different attacks
Coefficient unit: 1.0e +004
It should be noted that: the watermark embedding method provided by the embodiment of the application uses 32 bits of feature vector bits, and usually, the feature vector is used for sensing the hash, wherein the feature vector is greater than the average value and takes '1' and is smaller than the average value and takes '0', but the feature vector is not used for taking '1' and taking '0' according to a positive value or a zero value. The coefficients positive and zero are denoted by "1" and negative values by "0" for ease of observation only.
TABLE 2 relationship of feature vectors between different images (32bit)
| V1 | V2 | V3 | V4 | V5 | V6 | V7 | V8 | |
| V1 | 1.00 | 0.15 | 0.00 | 0.22 | 0.20 | -0.13 | 0.28 | 0.07 |
| V2 | 0.15 | 1.00 | 0.00 | 0.00 | 0.22 | -0.10 | -0.04 | 0.11 |
| V3 | 0.00 | 0.00 | 1.00 | 0.06 | -0.10 | -0.10 | -0.18 | 0.05 |
| V4 | 0.22 | 0.00 | 0.06 | 1.00 | 0.09 | 0.03 | -0.04 | -0.09 |
| V5 | 0.20 | 0.22 | -0.10 | 0.09 | 1.00 | -0.13 | 0.16 | 0.09 |
| V6 | -0.13 | -0.10 | -0.10 | 0.03 | -0.13 | 1.00 | -0.09 | 0.13 |
| V7 | 0.28 | -0.04 | -0.18 | -0.04 | 0.16 | -0.09 | 1.00 | 0.04 |
| V8 | 0.07 | 0.11 | 0.05 | -0.09 | 0.09 | 0.13 | 0.04 | 1.00 |
In summary, through the analysis of the Curvelet-DCT coefficients of the image, as shown in Table 2, it is found that the symbolic sequence of the Curvelet-DCT low-intermediate frequency coefficients can be used as the feature vector of the medical image. And embedding and extracting the zero watermark by using the feature vector of the image.
The following describes a watermark embedding system for an image provided by an embodiment of the present application, and the embedding system described below and the watermark embedding method described above may be referred to in correspondence.
Referring to fig. 2, fig. 2 is a schematic structural diagram of an image watermark embedding system according to an embodiment of the present application, and the present application further provides a watermark embedding system based on Curvelet-DCT and RSA sequences, including:
an acquisition module 100 for acquiring image data of a medical image;
a first transformation module 200, configured to perform a whole-image Curvelet transformation on the image data to obtain a first coefficient matrix;
a second transform module 300, configured to perform DCT transform on the first coefficient matrix at the Coarse layer to obtain a second coefficient matrix;
a third transformation module 400, configured to obtain a sub-matrix with a preset number of rows and columns of the second coefficient matrix, to obtain a third coefficient matrix, and calculate an average value of the third coefficient matrix;
a binary conversion module 500, configured to convert the third coefficient matrix into a binary matrix by using perceptual hashing;
the eigenvector determining module 600 is configured to perform dimension reduction operation on the binary matrix to obtain a visual eigenvector;
a watermark generation module 700, configured to generate an encrypted watermark using a pseudorandom sequence;
a watermark embedding module 800, configured to perform an exclusive or operation on the visual feature vector and the encrypted watermark bit by bit, so as to insert the encrypted watermark into the medical image, and obtain a logical key for watermark extraction.
Based on the foregoing embodiment, as a preferred embodiment, the binary conversion module 500 is specifically a module configured to compare the elements in the third coefficient matrix with the average value row by row, assign 1 to the elements greater than or equal to the average value, and assign 0 to the elements less than the average value, so as to obtain a binary matrix.
Based on the above embodiment, as a preferred embodiment, the method further includes:
and the sequence generating module is used for generating the pseudo-random sequence by utilizing an RSA algorithm.
Based on the foregoing embodiment, as a preferred embodiment, the watermark generating module 700 includes:
the sequence ordering unit is used for obtaining a num function by using an ordering function for the pseudo-random sequence;
and the watermark encryption unit is used for scrambling and encrypting the binary watermark by using the num function to obtain an encrypted watermark.
The present application also provides a computer readable storage medium having stored thereon a computer program which, when executed, may implement the steps provided by the above-described embodiments. The storage medium may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The application further provides a terminal, which may include a memory and a processor, where the memory stores a computer program, and the processor may implement the steps provided by the foregoing embodiments when calling the computer program in the memory. Of course, the terminal may also include various network interfaces, power supplies, and the like. The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system provided by the embodiment, the description is relatively simple because the system corresponds to the method provided by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present application are explained herein using specific examples, which are provided only to help understand the method and the core idea of the present application. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. A watermark embedding method based on Curvelet-DCT and RSA sequences is characterized by comprising the following steps:
acquiring image data of a medical image;
carrying out full-image Curvelet transformation on the image data to obtain a first coefficient matrix;
performing DCT (discrete cosine transform) transformation on the first coefficient matrix at a Coarse layer to obtain a second coefficient matrix;
taking a submatrix with preset row and column numbers of the second coefficient matrix to obtain a third coefficient matrix, and calculating an average value of the third coefficient matrix;
converting the third coefficient matrix into a binary matrix by using perceptual hashing;
performing dimensionality reduction operation on the binary matrix to obtain a visual feature vector;
generating an encrypted watermark by using a pseudorandom sequence;
and performing exclusive OR operation on the visual feature vector and the encrypted watermark bit by bit to insert the encrypted watermark into the medical image and obtain a logic key for watermark extraction.
2. The watermark embedding method of claim 1, wherein converting the third coefficient matrix into a binary matrix using perceptual hashing comprises:
and comparing the elements in the third coefficient matrix with the average value line by line, assigning 1 to the elements which are greater than or equal to the average value, and assigning 0 to the elements which are smaller than the average value to obtain a binary matrix.
3. The watermark embedding method of claim 1, wherein before generating the encrypted watermark using the pseudorandom sequence, further comprising:
the pseudo-random sequence is generated using the RSA algorithm.
4. The watermark embedding method of claim 3, wherein generating the encrypted watermark using the pseudorandom sequence comprises:
obtaining a num function by using a sequencing function for the pseudo-random sequence;
and scrambling and encrypting the binary watermark by using the num function to obtain an encrypted watermark.
5. A watermark embedding system based on Curvelet-DCT and RSA sequences is characterized by comprising:
an acquisition module for acquiring image data of a medical image;
the first transformation module is used for carrying out full-image Curvelet transformation on the image data to obtain a first coefficient matrix;
the second transformation module is used for performing DCT transformation on the first coefficient matrix at a Coarse layer to obtain a second coefficient matrix;
the third transformation module is used for taking a sub-matrix with the preset row and column number of the second coefficient matrix to obtain a third coefficient matrix and calculating the average value of the third coefficient matrix;
the binary conversion module is used for converting the third coefficient matrix into a binary matrix by using perceptual hash;
the eigenvector determining module is used for performing dimensionality reduction operation on the binary matrix to obtain a visual eigenvector;
the watermark generation module is used for generating an encrypted watermark by utilizing the pseudorandom sequence;
and the watermark embedding module is used for carrying out XOR operation on the visual feature vector and the encrypted watermark bit by bit so as to insert the encrypted watermark into the medical image and obtain a logic key for extracting the watermark.
6. The watermark embedding system according to claim 5, wherein the third variation module is specifically a module configured to compare the elements in the third coefficient matrix with the average value row by row, assign 1 to the elements greater than or equal to the average value, and assign 0 to the elements less than the average value, thereby obtaining a binary matrix.
7. The watermark embedding system of claim 5, further comprising:
and the sequence generating module is used for generating the pseudo-random sequence by utilizing an RSA algorithm.
8. The watermark embedding system of claim 7, wherein the watermark generation module comprises:
the sequence ordering unit is used for obtaining a num function by using an ordering function for the pseudo-random sequence;
and the watermark encryption unit is used for scrambling and encrypting the binary watermark by using the num function to obtain an encrypted watermark.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the watermark embedding method according to any one of claims 1-4.
10. A terminal, characterized in that it comprises a memory in which a computer program is stored and a processor which, when it is called up in said memory, implements the steps of the watermark embedding method according to any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910871201.1A CN110599390A (en) | 2019-09-16 | 2019-09-16 | Watermark embedding method based on Curvelet and RSA sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910871201.1A CN110599390A (en) | 2019-09-16 | 2019-09-16 | Watermark embedding method based on Curvelet and RSA sequence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110599390A true CN110599390A (en) | 2019-12-20 |
Family
ID=68859695
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910871201.1A Pending CN110599390A (en) | 2019-09-16 | 2019-09-16 | Watermark embedding method based on Curvelet and RSA sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110599390A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111625786A (en) * | 2020-05-07 | 2020-09-04 | 清华四川能源互联网研究院 | Time sequence data watermarking algorithm based on discrete cosine transform |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1750041A (en) * | 2005-10-20 | 2006-03-22 | 李京兵 | Digital watermark method capable of resisting geometric attack and conventional attack |
| CN102014266A (en) * | 2010-12-01 | 2011-04-13 | 华中科技大学 | Digital watermarking-based high-definition video encrypted transmitting method and system |
| CN102521796A (en) * | 2011-10-31 | 2012-06-27 | 南京大学 | Image block effect removing method on basis of curvelet transformation |
| CN103854251A (en) * | 2014-04-02 | 2014-06-11 | 海南大学 | Volume data multi-watermark method based on three-dimensional DWT-DCT (3D Wavelet Transform-Discrete Cosine Transformation) perceptual hashing |
| CN109727179A (en) * | 2018-12-29 | 2019-05-07 | 燕山大学 | A kind of zero watermarking generation method and system, extracting method and system |
| CN109741230A (en) * | 2018-12-13 | 2019-05-10 | 齐鲁工业大学 | Three image zero watermarking building methods and detection method based on quaternary number PHFM |
| CN110211020A (en) * | 2019-06-13 | 2019-09-06 | 海南大学 | A kind of watermark insertion of image and extracting method and relevant apparatus |
-
2019
- 2019-09-16 CN CN201910871201.1A patent/CN110599390A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1750041A (en) * | 2005-10-20 | 2006-03-22 | 李京兵 | Digital watermark method capable of resisting geometric attack and conventional attack |
| CN102014266A (en) * | 2010-12-01 | 2011-04-13 | 华中科技大学 | Digital watermarking-based high-definition video encrypted transmitting method and system |
| CN102521796A (en) * | 2011-10-31 | 2012-06-27 | 南京大学 | Image block effect removing method on basis of curvelet transformation |
| CN103854251A (en) * | 2014-04-02 | 2014-06-11 | 海南大学 | Volume data multi-watermark method based on three-dimensional DWT-DCT (3D Wavelet Transform-Discrete Cosine Transformation) perceptual hashing |
| CN109741230A (en) * | 2018-12-13 | 2019-05-10 | 齐鲁工业大学 | Three image zero watermarking building methods and detection method based on quaternary number PHFM |
| CN109727179A (en) * | 2018-12-29 | 2019-05-07 | 燕山大学 | A kind of zero watermarking generation method and system, extracting method and system |
| CN110211020A (en) * | 2019-06-13 | 2019-09-06 | 海南大学 | A kind of watermark insertion of image and extracting method and relevant apparatus |
Non-Patent Citations (2)
| Title |
|---|
| SY C.NGUYEN等: "An Improved Image Watermarking Scheme Using Selective Curvelet Scales", 《IEEE XPLORE》 * |
| 曲长波等: "基于Curvelet-DSVD和视觉密码的强鲁棒零水印算法", 《计算机应用研究》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111625786A (en) * | 2020-05-07 | 2020-09-04 | 清华四川能源互联网研究院 | Time sequence data watermarking algorithm based on discrete cosine transform |
| CN111625786B (en) * | 2020-05-07 | 2022-03-01 | 清华四川能源互联网研究院 | Time sequence data watermarking method based on discrete cosine transform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109360141B (en) | Image encryption method based on compressed sensing and three-dimensional cat mapping | |
| Zheng et al. | Walsh-Hadamard transform in the homomorphic encrypted domain and its application in image watermarking | |
| Bhatnagar et al. | A new robust adjustable logo watermarking scheme | |
| Ernawan | Tchebichef image watermarking along the edge using YCoCg-R color space for copyright protection | |
| CN111968026A (en) | Medical image robust watermarking method based on PHTs-DCT | |
| Adi et al. | High quality image steganography on integer Haar Wavelet Transform using modulus function | |
| CN111988492A (en) | Medical image robust watermarking method based on Gabor-DCT | |
| Alam et al. | Analysis of modified LSB approaches of hiding information in digital images | |
| CN110211020B (en) | Image watermark embedding and extracting method based on SWT-DFT | |
| CN111988490A (en) | Medical image robust watermarking method based on Tetrolet-DCT | |
| Yadav et al. | 3-Level security based spread spectrum image steganography with enhanced peak signal to noise ratio | |
| Shrestha et al. | Color image steganography technique using Daubechies discrete wavelet transform | |
| Devi et al. | A copyright protection scheme for digital images based on shuffled singular value decomposition and visual cryptography | |
| Nadiya et al. | Image steganography in DWT domain using double-stegging with RSA encryption | |
| CN110599390A (en) | Watermark embedding method based on Curvelet and RSA sequence | |
| Sun et al. | A Novel Image Steganography Based on Contourlet Transform and Hill Cipher. | |
| Park et al. | A new copyright protection scheme with visual cryptography | |
| Bhavani et al. | Image segmentation based hybrid watermarking algorithm for copyright protection | |
| Razak et al. | A review on digital image watermarking with cryptosystem techniques | |
| Habbouli et al. | A secure, self-recovery, and high capacity blind digital image information hiding and authentication scheme using DCT moments | |
| Sun et al. | A digital watermarking algorithm based on hyperchaos and discrete fractional Fourier transform | |
| CN116342357A (en) | Medical image robust watermarking method based on LPT-DCT | |
| Xing et al. | A new robust copyright protection scheme for digital image based on visual cryptography | |
| CN112907426A (en) | PHTs-DCT-based encrypted medical image digital watermark method | |
| Taouil et al. | High imperceptibility image steganography methods based on HAAR DWT |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191220 |