CN101004830A - Robust watermark technique of digital image selected from space domain of images - Google Patents

Robust watermark technique of digital image selected from space domain of images Download PDF

Info

Publication number
CN101004830A
CN101004830A CN 200610167730 CN200610167730A CN101004830A CN 101004830 A CN101004830 A CN 101004830A CN 200610167730 CN200610167730 CN 200610167730 CN 200610167730 A CN200610167730 A CN 200610167730A CN 101004830 A CN101004830 A CN 101004830A
Authority
CN
China
Prior art keywords
image
watermark
blocks
initial
block
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
Application number
CN 200610167730
Other languages
Chinese (zh)
Inventor
费泽松
孟维娟
匡镜明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Institute of Technology BIT
Original Assignee
Beijing Institute of Technology BIT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Institute of Technology BIT filed Critical Beijing Institute of Technology BIT
Priority to CN 200610167730 priority Critical patent/CN101004830A/en
Publication of CN101004830A publication Critical patent/CN101004830A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Editing Of Facsimile Originals (AREA)
  • Image Processing (AREA)

Abstract

A method for embedding watermark on robust digital image of airspace image includes using chose sequence generated by cipher key to select initial image pixel block, seeking optimum matched block with said pixel block in complete image, recombining them as per their generated order and making DWT conversion, dividing watermark information as per their importance, realizing unequal error protection by utilizing orthogonal variable length OVSF code spread spectrum treatment and embedding spread spectrum watermark into DWT wavelet coefficient. The method for picking up watermark is also disclosed.

Description

The Robust digital figure watermark technology that a kind of spatial domain image is selected
Technical field
The present invention relates to a kind of Robust digital figure watermark technology of selecting based on the spatial domain image, is a kind of multimedia data protection method, belongs to field of multimedia signal processing.
Background technology
Along with the continuous development of multimedia technology, the copyright problem of copyright becomes a focus of present concern.In order to address this problem, digital watermark technology has been proposed in the world.Digital watermark technology is a kind of brand-new information security technology that combines technology such as signal Processing, digital communication, pattern-recognition, cryptography and Internet, and it provides a kind of new thinking for copyright protection.Digital watermark technology is one of effective ways of protection Digital Media copyright at present.
Digital watermark technology confirms the entitlement of these data by embed watermark in raw data (watermark) information. and do not influence the availability of raw data, and should be able to resist various have a mind to or attacks unintentionally, i.e. watermark should have robustness preferably.The watermark information that embeds can be the proprietorial literal of representative, product or everyone ID, two dimensional image, random series etc.
In recent years, the research of digital watermark technology has obtained very big progress, and some technology has been used for commercial object.Digital watermark technology has a variety of, according to different purposes, has a lot of different sorting techniques, for example: according to the purposes of carrier, can be divided into rest image, video, audio frequency, text document, binary code stream etc.; According to appearance characteristics, can be divided into visible watermark and invisible watermark (robust watermark, fragile watermark); According to detection method, be divided into secret and open watermark, private key and PKI watermark etc.; According to embedding strategy, be divided into spatial domain watermark, transform domain watermark etc.At present research is more transform domain digital watermarking algorithm, compression domain digital watermarking algorithm, physiological models algorithm, based on the digital watermarking algorithm of chaotic model and spread spectrum digital watermarking algorithm etc.
The most of digital watermarking schemes that propose mostly can only be resisted several frequently seen attack at present, as compression of images, and filtering, attacked by noise and other some signal processing operations.And for translation, rotation, stretch or even small its robust performance of geometric attack all relatively poor.The DFT conversion is adopted in the digital watermarking of most of resist geometric attacks, incompatible with present JPEG2000 Standard of image compression and JPEG compression standard, therefore in digital watermarking research, the digital watermark technology that obtains more efficiently resist geometric attacks and normal image processing operation becomes very important.
Summary of the invention
The objective of the invention is to propose a kind of digital watermarking embedding and extracting method that common image processing operations is had higher robustness as compression, noise etc. and geometric attack as rotation, translation etc.
The technical solution adopted for the present invention to solve the technical problems comprises that watermark embeds and extract the two large divisions.First is that watermark embeds, and comprising:
(1) original watermark information is divided into i grade by importance; carry out orthogonal spectrum expansion with the orthogonal variable spreading factor ovsf code and handle, to the long ovsf code spread spectrum of pith, according to importance; select the spreading code of unequal length successively for use, reach the effect of unequal error protection.
(2) be the chaotic maps initial value with the key, obtain the chaos random series, chaos sequence is carried out quantification treatment, as index, select the initial pictures block of pixels with quantized value.
(3) be starting point with the initial pictures block of pixels, form the block of pixels of N * N size,, in entire image, select the blocks and optimal matching blocks with size such as initial pixel piece, make the MSE that defines between itself and the initial pixel piece according to the similarity of image pixel information MReach minimum, select MSE MBe worth minimum p piece as blocks and optimal matching blocks, and blocks and optimal matching blocks is reconfigured, carry out the DWT wavelet transformation.
(4) watermark information behind the spread spectrum is embedded in the DWT wavelet conversion coefficient of blocks and optimal matching blocks.
Second portion is a watermark extracting, comprising:
(5) recovering key is the chaotic maps initial value, obtains the chaos random series, and chaos sequence is carried out quantification treatment, as index, selects the initial pictures block of pixels of watermarking images with quantized value.
(6) the initial pixel piece with watermarking images is a starting point, according to the MSE of definition MIn entire image, select blocks and optimal matching blocks, it is reconfigured, and carry out the DWT wavelet transformation.
(7) adopt correlation detection, the spread-spectrum watermark information that embeds is carried out despreading handle, recover original watermark information.
The spread spectrum coding The Application of Technology has improved the confidentiality and the security of system.Along with the increase of spreading code length, the security and the robustness of system improve thereupon.The present invention has adopted the unequal error technology, by importance original watermark information is divided, it is divided into different grades, part to outbalance in the raw information adopts long spreading code to carry out spread processing, the spread spectrum code length of other part successively decreases successively, has effectively guaranteed the accuracy of watermark information when detecting.
Aspect selection watermark information position,, calculate the square error function MSE of definition according to the similarity between the block of image pixels MIn entire image, select similar each other block of pixels, in the time of geometric attacks such as image is rotated, translation, because though the relative position between each block of image pixels has changed, but the similarity between each block of image pixels is not changed, and still can determine the position that watermark embeds comparatively accurately according to the similarity between the block of pixels, so can extract watermark information comparatively accurately, therefore aspect resist geometric attacks, have superiority.
In addition, the application of chaos random series has increased the security of algorithm, and wavelet transformation has the advantage of time-space domain and the conversion of DCT territory concurrently.The high-frequency region of DWT wavelet transformation has all shown weak robustness at aspects such as opposing JPEG compression, low-pass filtering, attacked by noise, but its invisibility is good; And low frequency region has robustness preferably aspect the various signal processing operations of opposing, but embeddable capacity is less, and invisibility is poor.Watermark information is embedded in DWT intermediate frequency zone, both guarantees the invisibility of the image of embed watermark, strengthened the robustness of various common signal Processing such as watermaking system opposing compression, filtering, noise again.
The geometric attack that the present invention suffers carrier image has resistibility preferably, particularly attacks such as rotation, translation, and general signal Processing such as operations such as filtering, compression also had good robustness.
Description of drawings
The match block of Fig. 1-initial pixel piece is selected
Fig. 2-watermark embed process
Fig. 3-watermark detection process
Fig. 4-DWT wavelet transformation
Embodiment
The invention will be further described below in conjunction with drawings and Examples:
At first with one group of binary sequence B={b that can represent copyright information 0, b 1..., b L-1, b i∈ 1,1} is as digital watermarking, the original image note do X={x (i, j) | i≤M 1, j≤M 2, wherein, M 1* M 2Be the image size, x (i j) is the pixel value of image, and the telescopiny of watermark is as follows as shown in Figure 2:
Step 1. couple original watermark information B={b 0, b 1..., b L-1Be divided into i grade G by importance 1, G 2G i, importance is successively decreased successively with the increase of i, wherein G 1For
Figure A20061016773000051
Most important, G 2For
Figure A20061016773000052
Inferior important, G iImportance minimum.Carry out orthogonal spectrum expansion with the orthogonal variable spreading factor ovsf code and handle, to pith G 1With long ovsf code spread spectrum, its spreading factor is C r=2 k, G 2And G iThe part spreading factor is decremented to 2 K-1, 2 K-i+1(k is a positive integer) obtains the watermark information W behind the spread spectrum.
Step 2. is the chaotic maps initial value with the key, obtains the chaos random series, and chaos sequence is carried out quantification treatment, as index, selects the initial pictures block of pixels with quantized value.
Chaos is nonlinear equation (group) separating under given conditions, and its output at random determines that by one equation (group) determines.Its initial parameter can be used as system key, the characteristics that chaos system has good performance at random and regenerates easily, and this has increased the security of system.With the one dimension chaotic maps is example f:U → U, U ∈ R
z(n+1)＝f(λ，z(n))，λ∈R，z(n)∈U
Wherein, n is a positive integer, the expression iterations, and λ is a systematic parameter.The real-valued sequence that the following formula iteration obtains for n time for z (1), z (2) ..., z (n) }.Choose num different chaos value, each chaos value is mapped as col (i) ∈ M1, two vectors of row (i) ∈ M2, i={1 wherein, 2 ..., num}, num are the number of initial pixel piece, col and row are the position at initial pixel piece place.
Step 3. as shown in Figure 1, (col row) is starting point with the initial pictures block of pixels, form the block of pixels of N * N size, according to the similarity of Pixel Information between image, in entire image, select the blocks and optimal matching blocks with size such as initial pixel piece, make the definition MSE between itself and the initial pixel piece MReach minimum, select MSE MMinimum p as blocks and optimal matching blocks.
The upper left corner of image initial block of pixels is positioned at that (i, j), then the block of pixels of N * N size can be expressed as { x I+m, j+n| 0≤m≤N-1,0≤n≤N-1}.The spectrum transfer function is defined as x I+m, j+n→ v (x I+m, j-n) here, v is an one dimension real number equation.V is defined as
V (z)=a 0+ a 1Z or v (z)=z
Wherein a 1 = N 2 Σ m = 0 N - 1 Σ n = 0 N - 1 ( X k + m , l + n ) ( X i + m , j + n ) - Σ m = 0 N - 1 Σ n = 0 N - 1 X k + m , l + n Σ m = 0 N - 1 Σ n = 0 N - 1 X i + m , j + n N 2 Σ m = 0 N - 1 Σ n = 0 N - 1 X i + m , j + n 2 - ( Σ m = 0 N - 1 Σ n = 0 N - 1 X i + m , j + n ) 2
Figure A20061016773000062
The block of pixels of sizes such as calculating and initial pixel piece and the MSE between the initial pixel piece M, MSE wherein MBe defined as:
MSE M = 1 N 2 Σ m = 0 N - 1 Σ n = 0 N - 1 ( v ( X k + m , l + n ) - X i + m , j + n ) 2
When choosing, select MSE for use MP block of pixels of value minimum is as initial pixel piece (i, blocks and optimal matching blocks j).
(col row) all does above-mentioned identical processing, obtains a series of blocks and optimal matching blocks, reconfigures according to the order that produces to be subimage X to each initial pixel piece New, and to subimage X NewCarry out the secondary wavelet transformation.
The watermark information of step 4. after with spread spectrum is embedded in subimage X NewThe intermediate frequency HL of secondary wavelet conversion coefficient 1, LH 1, HL 2, LH 2In the sub-band coefficients.
Adopt embedding grammar commonly used, embedding formula is X New=X New+ α W, wherein α is an embedment strength.
Amended coefficient is carried out the IDWT inverse transformation obtain X New, with X NewTo original image, obtain the image X ' behind the embed watermark by the initial arrangement order recovery.
The leaching process of watermark is as follows:
It is the chaotic maps initial value that step 1. is recovered key, obtains the chaos random series, and chaos sequence is carried out quantification treatment, as index, selects the initial pictures block of pixels of watermarking images with quantized value.
After the one dimension chaotic maps being carried out n iteration, from the real-valued sequence that produces, find num identical when embedding chaos value, and it is mapped as two-dimensional vector col (i) ∈ M 1, row (i) ∈ M 2, i={1 wherein, 2 ..., num}, num are the number of initial pixel piece, col and row are the position at initial pixel piece place.
Step 2. is a starting point with the initial pixel piece of watermarking images, forms the block of pixels of N * N size, according to definition MSE MIn entire image, select MSE MBe worth p minimum block of pixels as blocks and optimal matching blocks, and reconfigure according to the order that produces.
Image after step 3. pair reconfigures carries out secondary DWT wavelet transformation, extracts the intermediate frequency subband wavelet coefficient HL that embeds, LH 1, HL 2, LH 2
Step 4. adopts correlation detection to extract watermark information, does relevant despreading with the orthogonal variable spreading factor ovsf code that produces with the regional wavelet coefficient of embedding and handles.
Coherent detection is: ρ = 1 C r Σ i C r X new ′ ( i ) PN ( i ) , X wherein New(i) for watermarking images embeds regional wavelet coefficient values, PN (i) is an ovsf code.If ρ＞0 b '=1, otherwise b '=0.
Be easy to realize the extraction and the detection of digital watermarking according to above introduction.Use the present invention and can resist geometric attacks such as translation, rotation effectively, and image processing operations such as general Flame Image Process such as JPEG compression are also had good robustness.

Claims (5)

1. Robust digital figure watermark technology that the spatial domain image is selected is characterized in that: comprise following concrete steps:
First is the embedding of watermark information, mainly comprises:
1) with the initial value of key, obtains the chaos random series, the chaos random series is quantized, and select the initial pictures block of pixels with quantized value as chaotic maps;
2) be initial point with the initial pictures block of pixels, form the block of pixels of N * N size,, in entire image, seek the blocks and optimal matching blocks with size such as initial pixel piece, promptly similar, make the MSE between itself and the initial pixel piece according to the similarity between the image block MReach minimum, select MSE MThe p piece of value minimum is reassembled into subimage with blocks and optimal matching blocks in proper order by generation, and carries out the DWT wavelet transformation as blocks and optimal matching blocks;
3) original watermark information is carried out unequal error protection; it is divided according to importance; and carry out orthogonal variable spreading factor sign indicating number spread processing by importance; select the spreading code of unequal length successively for use; important information is used than long spreading code spread spectrum, and the watermark information behind the spread spectrum is embedded in the carrier image match block DWT wavelet coefficient.
Second portion is the extraction of watermark information, mainly comprises:
4) be the chaotic maps initial value with the key equally, obtain the chaos random series, chaos sequence is carried out quantification treatment, as index, select the initial pictures block of pixels of watermarking images with quantized value;
5) the initial pixel piece with watermarking images is a starting point, according to the MSE of definition MIn entire image, select blocks and optimal matching blocks, it is reconfigured, and carry out the DWT wavelet transformation;
6) adopt correlation detection, the spread-spectrum watermark information that embeds is carried out despreading handle, recover original watermark information.
2. the Robust digital figure watermark technology that a kind of spatial domain according to claim 1 image is selected; it is characterized in that: original watermark information is carried out unequal error protection; it is divided into i grade according to importance; and, select for use the spreading code of unequal length to carry out the spread spectrum protection successively by importance.
3. the Robust digital figure watermark technology that a kind of spatial domain according to claim 1 image is selected, it is characterized in that: according to the susceptibility of chaos sequence initial value, when embedding, produce chaos sequence with key, and it is quantized the initial pixel piece position that image is selected in the back.
4. the Robust digital figure watermark technology that a kind of spatial domain according to claim 1 image is selected, it is characterized in that: with the initial pictures block of pixels is initial point, form the block of pixels of N * N size, according to the similarity between the block of image pixels, the blocks and optimal matching blocks of sizes such as searching and initial pixel piece in entire image, be similar, make the MSE that defines between itself and the initial pixel piece MReach minimum, select MSE MThe p piece of value minimum is as blocks and optimal matching blocks.
5. the Robust digital figure watermark technology that a kind of spatial domain according to claim 1 image is selected is characterized in that: the blocks and optimal matching blocks of selecting is reassembled into a new subimage according to the order that produces, and subimage is carried out the DWT conversion.
CN 200610167730 2006-12-20 2006-12-20 Robust watermark technique of digital image selected from space domain of images Pending CN101004830A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200610167730 CN101004830A (en) 2006-12-20 2006-12-20 Robust watermark technique of digital image selected from space domain of images

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200610167730 CN101004830A (en) 2006-12-20 2006-12-20 Robust watermark technique of digital image selected from space domain of images

Publications (1)

Publication Number Publication Date
CN101004830A true CN101004830A (en) 2007-07-25

Family

ID=38703944

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200610167730 Pending CN101004830A (en) 2006-12-20 2006-12-20 Robust watermark technique of digital image selected from space domain of images

Country Status (1)

Country Link
CN (1) CN101004830A (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101847251A (en) * 2010-06-13 2010-09-29 电子科技大学 Digital watermark embedding and extraction method based on chaos sequences
CN101916427A (en) * 2010-08-10 2010-12-15 浙江大学 Spatial domain-based image watermark adding method
CN102082582A (en) * 2011-02-17 2011-06-01 清华大学 Chaos direct sequence spread spectrum signal transmitting and receiving devices
CN101782947B (en) * 2008-12-19 2011-12-14 天津易步科技发展有限公司 Multimedia copyright protection method
CN103024244A (en) * 2008-03-14 2013-04-03 弗劳恩霍夫应用研究促进协会 Embedder for embedding a watermark in an information presentation, detector for detecting a watermark in an information presentation, method and computer program
CN106961602A (en) * 2017-03-30 2017-07-18 福建北卡科技有限公司 A kind of cross-platform incompressible color image information hidden algorithm based on RS and Hamming code
CN107067363A (en) * 2017-04-18 2017-08-18 海南大学 A kind of non-blind robust Spatial Domain Image Watermark method constant based on pixel order relation
CN111127291A (en) * 2019-12-30 2020-05-08 山东师范大学 Image watermark embedding and extracting method and system based on spatial frequency domain JND conversion
CN111145070A (en) * 2019-12-12 2020-05-12 中国科学院深圳先进技术研究院 Safe and robust watermarking algorithm
CN111640051A (en) * 2019-03-01 2020-09-08 浙江大学 Image processing method and device
CN111881929A (en) * 2020-06-09 2020-11-03 长江大学 Method and device for detecting large-period state of Duffing system based on chaotic image pixel identification
CN112669191A (en) * 2019-10-15 2021-04-16 国际关系学院 Anti-overflow reversible digital watermark embedding and extracting method based on image content identification

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103024244A (en) * 2008-03-14 2013-04-03 弗劳恩霍夫应用研究促进协会 Embedder for embedding a watermark in an information presentation, detector for detecting a watermark in an information presentation, method and computer program
US9037453B2 (en) 2008-03-14 2015-05-19 Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. Embedder for embedding a watermark into an information representation, detector for detecting a watermark in an information representation, method and computer program
CN103024244B (en) * 2008-03-14 2015-08-26 弗劳恩霍夫应用研究促进协会 Detector, the method and computer program of the watermark during the embedded device in being represented to information by watermark embedment, Detection Information represent
CN101782947B (en) * 2008-12-19 2011-12-14 天津易步科技发展有限公司 Multimedia copyright protection method
CN101847251A (en) * 2010-06-13 2010-09-29 电子科技大学 Digital watermark embedding and extraction method based on chaos sequences
CN101916427A (en) * 2010-08-10 2010-12-15 浙江大学 Spatial domain-based image watermark adding method
CN102082582A (en) * 2011-02-17 2011-06-01 清华大学 Chaos direct sequence spread spectrum signal transmitting and receiving devices
CN102082582B (en) * 2011-02-17 2013-11-06 清华大学 Chaos direct sequence spread spectrum signal transmitting and receiving devices
CN106961602A (en) * 2017-03-30 2017-07-18 福建北卡科技有限公司 A kind of cross-platform incompressible color image information hidden algorithm based on RS and Hamming code
CN106961602B (en) * 2017-03-30 2019-10-18 北卡科技有限公司 A kind of cross-platform incompressible color image information hidden algorithm based on RS and Hamming code
CN107067363A (en) * 2017-04-18 2017-08-18 海南大学 A kind of non-blind robust Spatial Domain Image Watermark method constant based on pixel order relation
CN111640051A (en) * 2019-03-01 2020-09-08 浙江大学 Image processing method and device
CN112669191A (en) * 2019-10-15 2021-04-16 国际关系学院 Anti-overflow reversible digital watermark embedding and extracting method based on image content identification
CN111145070A (en) * 2019-12-12 2020-05-12 中国科学院深圳先进技术研究院 Safe and robust watermarking algorithm
CN111145070B (en) * 2019-12-12 2024-04-05 中国科学院深圳先进技术研究院 Safe and robust watermarking algorithm
CN111127291A (en) * 2019-12-30 2020-05-08 山东师范大学 Image watermark embedding and extracting method and system based on spatial frequency domain JND conversion
CN111881929A (en) * 2020-06-09 2020-11-03 长江大学 Method and device for detecting large-period state of Duffing system based on chaotic image pixel identification
CN111881929B (en) * 2020-06-09 2023-12-08 长江大学 Method and device for detecting large-period state of Duffing system based on chaotic image pixel identification

Similar Documents

Publication Publication Date Title
CN101004830A (en) Robust watermark technique of digital image selected from space domain of images
Sari et al. A good performance OTP encryption image based on DCT-DWT steganography
CN102156955B (en) Robust reversible watermark embedding and extracting method based on histogram neighborhood
US20080226124A1 (en) Method For Inserting and Extracting Multi-Bit Fingerprint Based on Wavelet
CN102750660A (en) Method and device for embedding and extracting digital watermarking
CN113989096B (en) Robust image watermarking method and system based on deep learning and attention network
CN104933668A (en) Processing method of reversible watermark on the basis of local positioning and ALattar integer transform
Chammem et al. DWT-based stereoscopic image watermarking
Khalifa et al. A robust non-blind algorithm for watermarking color images using multi-resolution wavelet decomposition
Kumar et al. A hybrid digital watermarking approach using wavelets and LSB
Bekkouch et al. Robust and reversible image watermarking scheme using combined DCT-DWT-SVD transforms
Xie et al. A blind wavelet based digital signature for image authentication
Hazim et al. A Novel Method of Invisible Video Watermarking Based on Index Mapping and Hybrid DWT-DCT.
CN100547609C (en) A kind of bittorrent Robust digital figure watermark embedding and detection method and device
Kumar et al. Digital image steganography based on combination of DCT and DWT
Ghosh et al. Spread spectrum image watermarking with digital design
Chalamala et al. Analysis of wavelet and contourlet transform based image watermarking techniques
Bouzidi et al. Contourlet domain feature extraction for image content authentication
Shahraeini et al. A robust digital image watermarking approach against jpeg compression attack based on hybrid fractal-wavelet
CN101093576A (en) Digital watermark method of serial multiple robustnesses based on multiple copyright authentications
Lee et al. Genetic algorithm-based watermarking in discrete wavelet transform domain
Fan et al. ROI-based watermarking scheme for JPEG 2000
Wu et al. Robust watermarking for text images based on Arnold scrambling and DWT-DCT
Jiao et al. Dual-tree complex wavelet transform blind video watermark based on complete complementary codes
CN104143173A (en) Image self-adaption blind watermarking algorithm based on DWT-DCT

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication