WO2011026365A1 - 一种图像数字水印嵌入和提取的方法及系统 - Google Patents

一种图像数字水印嵌入和提取的方法及系统 Download PDF

Info

Publication number
WO2011026365A1
WO2011026365A1 PCT/CN2010/074143 CN2010074143W WO2011026365A1 WO 2011026365 A1 WO2011026365 A1 WO 2011026365A1 CN 2010074143 W CN2010074143 W CN 2010074143W WO 2011026365 A1 WO2011026365 A1 WO 2011026365A1
Authority
WO
WIPO (PCT)
Prior art keywords
frequency coefficient
low frequency
embedded
coefficient matrix
watermark
Prior art date
Application number
PCT/CN2010/074143
Other languages
English (en)
French (fr)
Inventor
胥虎军
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to PL10813287T priority Critical patent/PL2428930T3/pl
Priority to EP10813287.9A priority patent/EP2428930B1/en
Priority to ES10813287T priority patent/ES2431595T3/es
Priority to US13/258,003 priority patent/US8615103B2/en
Publication of WO2011026365A1 publication Critical patent/WO2011026365A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/32101Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title
    • H04N1/32144Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
    • H04N1/32149Methods relating to embedding, encoding, decoding, detection or retrieval operations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00General purpose image data processing
    • G06T1/0021Image watermarking
    • G06T1/005Robust watermarking, e.g. average attack or collusion attack resistant
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/32101Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title
    • H04N1/32144Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
    • H04N1/32149Methods relating to embedding, encoding, decoding, detection or retrieval operations
    • H04N1/32154Transform domain methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2201/00General purpose image data processing
    • G06T2201/005Image watermarking
    • G06T2201/0052Embedding of the watermark in the frequency domain
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2201/00General purpose image data processing
    • G06T2201/005Image watermarking
    • G06T2201/0083Image watermarking whereby only watermarked image required at decoder, e.g. source-based, blind, oblivious

Definitions

  • the present invention relates to the field of computer information security and copyright protection, and in particular, to a method and system for image digital watermark embedding and extraction.
  • Digital Watermark technology is the use of signal processing techniques to embed hidden marks in digitized multimedia data. Such marks are usually invisible and can only be extracted by a dedicated detector or reader. Information embedded in digital media must have the following two basic characteristics to be called a digital watermark: easy to detect.
  • Robustness means that a digital watermark containing digital watermarks remains intact or can still be accurately identified after undergoing a variety of unintentional or intentional signal processing. Possible signal processing includes channel noise, filtering, digital-to-analog conversion, analog-to-digital conversion, resampling, shearing, displacement, scale variation, and lossy compression coding.
  • the transform domain watermark becomes a more important watermarking technique because it can embed a large amount of data without reducing the concealment of the watermark.
  • Such techniques are generally based on common image transformations, based on local or total transformations, including Discrete Cosine Transform (DCT), Discrete Wavelet Transform (DWT), Discrete Fourier Transform (Discrete Fourier) Transform, DFT), Singular value decomposition (SVD), Discrete Hadamard Transform (DHT), etc.
  • DCT Discrete Cosine Transform
  • DWT Discrete Wavelet Transform
  • DFT Discrete Fourier Transform
  • SVD Singular value decomposition
  • DHT Discrete Hadamard Transform
  • block-based DCT is one of the most commonly used transforms.
  • JPEG currently used is also based on block DCT.
  • One of the earliest digital watermarking schemes based on block DCT is to randomly select some blocks of the image by a key, and slightly change a triplet in the IF coefficients of the frequency domain coefficients to hide the binary sequence information. Choosing to encode in the IF component is because it is easy to be processed by various signals in high frequency encoding. The method is destroyed, while in low frequency encoding, since human vision is sensitive to low frequency components, changes to low frequency components are easily perceived. This digital watermarking technique is robust to lossy compression and low pass filtering.
  • Another DCT digital watermarking technique first divides the image into 8x8 non-overlapping blocks. After the block DCT, the frequency block composed of DCT coefficients is obtained, and then some frequency blocks are randomly selected to embed the watermark signal into the key. Control some of the selected DCT coefficients.
  • the method is to represent a bit of information by performing a small transformation on the selected DCT coefficients to satisfy a particular relationship. When the watermark information is extracted, the same DCT coefficients are selected, and the bit information is extracted according to the relationship between the coefficients.
  • the technical problem to be solved by the present invention is to propose a method and system for image digital watermark embedding and extracting, which is used for copyright protection of digital images, and can also hide hidden data, and can be widely used in the field of copyright protection and information security. .
  • the present invention provides an image digital watermark embedding method, comprising: performing DWT on a digital image of an original carrier to obtain two intermediate frequency coefficient matrices;
  • DWT is performed on another intermediate frequency coefficient matrix, and then the obtained low frequency coefficient matrix is selected as the first Two low frequency coefficient matrix;
  • the obtained two intermediate frequency coefficient matrices are the horizontal intermediate frequency coefficient matrix (LH1) and the vertical intermediate frequency coefficient matrix (HL1).
  • the method further comprises: scrambling and/or encrypting the watermark.
  • the present invention also provides an image digital watermark embedding system, comprising a matrix transformation processing module, a storage module and a watermark embedding processing module, wherein:
  • the matrix transformation processing module is configured to: perform DWT on the original carrier digital image to obtain two intermediate frequency coefficient matrices; perform DCT on one of the intermediate frequency coefficient matrices, then perform DWT, and then select the obtained low frequency coefficient matrix as the first low frequency coefficient. Matrix; D WT is performed on another intermediate frequency coefficient matrix, and then the obtained low frequency coefficient matrix is selected as the second low frequency coefficient matrix;
  • the storage module is configured to: store a preset embedded value and a correspondence relationship between the first low frequency coefficient and the second low frequency coefficient obtained by the matrix transformation processing module; the watermark embedding processing module is configured The first low frequency coefficient matrix and the second low frequency coefficient matrix are respectively at the same embedding position, and the first low frequency coefficient and the second low frequency coefficient are adjusted according to the watermark value to be embedded at the embedding position, so that after the adjustment The relationship between the magnitudes of the two low frequency coefficients corresponds to the correspondence between the preset embedded value in the storage module and the size relationship of the two low frequency coefficients.
  • the watermark embedding processing module is further configured to: scramble and/or encrypt the watermark before adjusting the size of the first low frequency coefficient and the second low frequency coefficient according to the watermark value to be embedded at the embedded position deal with.
  • the present invention also provides an image digital watermark extraction method, comprising: performing D WT on an image embedded with a watermark, and obtaining two intermediate frequency coefficient matrices;
  • DWT is performed on another intermediate frequency coefficient matrix, and then the obtained low frequency coefficient matrix is selected as the second low frequency coefficient matrix;
  • the obtained two intermediate frequency coefficient matrices are the horizontal intermediate frequency coefficient matrix (LH1_Y) and the vertical intermediate frequency coefficient matrix (HL1_Y).
  • the above image digital watermark extraction method may further include:
  • the reordering and/or decryption process is performed after the embedded digital watermark image is obtained.
  • the present invention also provides an image digital watermark extraction system, comprising a matrix transformation processing module, a storage module and a watermark extraction processing module, wherein:
  • the matrix transformation processing module is configured to: perform DWT on the image embedded with the watermark to obtain two intermediate frequency coefficient matrices; perform DCT on one of the intermediate frequency coefficient matrices, then perform DWT, and then select the obtained low frequency coefficient matrix as the first low frequency. Coefficient matrix; D WT is performed on another intermediate frequency coefficient matrix, and then the obtained low frequency coefficient matrix is selected as the second low frequency coefficient matrix;
  • the storage module is configured to: store a preset embedded value and a correspondence relationship between the first low frequency coefficient and the second low frequency coefficient obtained by the matrix transformation processing module;
  • the watermark extraction processing module is configured For: comparing the magnitude relationship between the first low frequency coefficient and the second low frequency coefficient at the same embedded position on the first low frequency coefficient matrix and the second low frequency coefficient matrix, And the obtained relationship between the size of the two is matched with the corresponding relationship between the preset embedded value in the storage module and the size relationship of the two low-frequency coefficients, and the embedded value at each embedded position is obtained, thereby obtaining an embedded digital watermark image.
  • the watermark extraction processing module is further configured to: after obtaining the embedded digital watermark image, reordering and/or decrypting the watermark.
  • the method and system for image digital watermark embedding and extracting proposed by the invention have good watermark concealability, and can resist histogram equalization, image brightness adjustment, contrast adjustment, salt and pepper noise, multiplicative noise, Gaussian noise, low-pass filtering,
  • JPEG Joint Photographic Experts Group
  • FIG. 1 is a flow chart of an image digital watermark embedding method according to an embodiment of the present invention
  • FIG. 2 is a block diagram of an image digital watermark embedding system according to an embodiment of the present invention.
  • FIG. 3 is a flowchart of an image digital watermark extraction method according to an embodiment of the present invention.
  • FIG. 4 is a block diagram of an image digital watermark extraction system in accordance with an embodiment of the present invention.
  • an image digital watermark embedding method includes the following steps: Step S101: performing DWT on the original carrier digital image X to obtain two intermediate frequency coefficient matrices LH1 and HL1;
  • the low frequency coefficient matrix LL1 After DWT is performed on the original carrier digital image X, the low frequency coefficient matrix LL1, the horizontal intermediate frequency coefficient matrix LH1, the vertical intermediate frequency coefficient matrix HL1, and the high frequency coefficient matrix HH1 are selected.
  • the intermediate frequency coefficient matrices LH1 and HL1 perform the following operations.
  • Step S102 performing DCT on the horizontal intermediate frequency coefficient matrix LH1 to obtain a matrix LH1-DCT, and then performing DWT on the matrix LH1-DCT, and selecting a low-frequency coefficient matrix LH1 DCT LL2 obtained as a first low-frequency coefficient matrix;
  • the low-frequency coefficient matrix LH1-DCT-LL2 After DWT is performed on the matrix LH1-DCT, the low-frequency coefficient matrix LH1-DCT-LL2, the intermediate-frequency coefficient matrix LH1-DCT-LH2 and LH1-DCT-HL2, and the high-frequency coefficient matrix LH1 DCT HH2 are obtained, and the low-frequency coefficient matrix is selected.
  • LHl - DCT - LL2 as the first low frequency coefficient matrix
  • Step S103 performing DWT on the vertical intermediate frequency coefficient matrix HL1, and selecting the low frequency coefficient matrix HL1_LL2 obtained after the transformation as the second low frequency coefficient matrix;
  • the low frequency coefficient matrix HL1 - LL2 After DWT is performed on the matrix HL1, the low frequency coefficient matrix HL1 - LL2, the intermediate frequency coefficient matrix HL1 - LH2 and HL1 - HL2, and the high frequency coefficient matrix HL1 - HH2 are obtained, and the low frequency coefficient matrix HL1 - LL2 is selected as the second low frequency coefficient.
  • Step S104 at the same embedding position (x, y) on the first low-frequency coefficient matrix LH1 - DCT - LL2 and the second low-frequency coefficient matrix HL1_LL2, according to the watermark value to be embedded at the embedding position (x, y), The size of the two low frequency coefficients is adjusted such that the relationship between the adjusted two low frequency coefficients is in accordance with the correspondence between the preset embedded value and the magnitude relationship of the two low frequency coefficients.
  • the watermark to be embedded is a meaningful binary image, that is, the value embedded in each embedding position (x, y) is 0 or 1;
  • the correspondence between the preset embedded value and the magnitude relationship of the two low frequency coefficients may be: If an embedded position (i, j) is to be embedded with 0, the first low frequency coefficient LHl_DCT_LL2 corresponding to the position (i, j) is corresponding.
  • the first low frequency coefficient LH1 corresponding to the position (i, j) is DCT—LL2(i, j) > second low-frequency coefficient HL1—LL2(i, j), where l ⁇ i ⁇ M, Kj ⁇ N, M is 1/4 of the height of the original carrier digital image X, N is the original
  • the carrier digital image has a width 1/4 of the width.
  • the corresponding relationship may also be set to other.
  • the present invention is not limited thereto, as long as different embedding values are corresponding to different first low-frequency coefficients LH1 - DCT - LL2 (i, j) and second low-frequency coefficients HL1.
  • LH1 DCT LL2 ( x, y ) and HL1_LL2 ( x, y ) needs to be adjusted, it can be adjusted according to certain rules, for example, by exchanging the values of these two coefficients, or if the two data are equal. , you can make fine adjustments.
  • Steps S101 to S104 complete the embedding of the watermark.
  • the embedded image appears in plaintext. If you want to embed a hidden image, if it is intercepted by a third party, it cannot be kept secret.
  • the digital watermark image may be scrambled and/or encrypted and then embedded in the W, which will effectively prevent information leakage.
  • the step S102 is to perform DCT on the vertical intermediate frequency coefficient matrix HL1 to obtain a matrix HL1 - DCT, and then perform DWT transformation on the matrix HL1 - DCT, and select a low frequency coefficient matrix HL1 obtained after the transformation - DCT_LL2 is used as the first low-frequency coefficient matrix.
  • the horizontal intermediate frequency coefficient matrix LH1 is DWT
  • the transformed low-frequency coefficient matrix LH1_LL2 is selected as the second low-frequency coefficient matrix, and the remaining steps are unchanged.
  • the embodiment of the present invention further provides a watermark embedding system, as shown in FIG. 2, comprising a matrix transformation processing module 21, a storage module 22, and a watermark embedding processing module 23, wherein:
  • the matrix transformation processing module 21 is configured to perform DWT on the original carrier digital image to obtain two intermediate frequency coefficient matrices; perform DCT on one of the intermediate frequency coefficient matrices, then perform DWT, and then select the obtained low frequency coefficient matrix as the first low frequency. Coefficient matrix; DWT is performed on another intermediate frequency coefficient matrix, and then the obtained low frequency coefficient matrix is selected as the second low frequency coefficient matrix;
  • the storage module 22 is configured to store a preset embedded value and a correspondence relationship between the first low frequency coefficient and the second low frequency coefficient obtained by the matrix transformation processing module 21; the watermark embedding process Module 23, for using the first low frequency coefficient matrix and the second low frequency coefficient moment Adjusting the size of the first low frequency coefficient and the second low frequency coefficient according to the watermark value to be embedded at the embedded position, so that the size relationship of the adjusted two low frequency coefficients conforms to the storage module 22 Corresponding relationship between the preset embedded value and the size relationship of the two low frequency coefficients.
  • the watermark embedding processing module 23 is further configured to scramble and/or encrypt the watermark before adjusting the size of the two low frequency coefficients according to the watermark value to be embedded at the embedding position.
  • FIG. 3 shows an image digital watermark extraction method according to an embodiment of the present invention.
  • the method for extracting the watermark embedded by the method described in FIG. 1 includes the following steps:
  • Step S301 performing DWT on the watermarked image Y to obtain two intermediate frequency coefficient matrices LH1 Y and HL1_Y;
  • low frequency coefficient matrix LL1 - Y low frequency coefficient matrix
  • horizontal intermediate frequency coefficient matrix LH1 - ⁇ horizontal intermediate frequency coefficient matrix
  • HL1 - ⁇ vertical intermediate frequency coefficient matrix
  • high frequency coefficient matrix HH1 - ⁇ select 2 intermediate frequency coefficient matrix LHl — ⁇ and HL1 — ⁇
  • Step S302 performing DCT on the horizontal intermediate frequency coefficient matrix LH1 - , to obtain a matrix LH1 DCT Y, and then performing DWT on the matrix LH1 - DCT - ,, and selecting the low frequency coefficient matrix LH1 - DCT - LL2 - ⁇ obtained as the first low frequency Coefficient matrix
  • Step S303 performing DWT on the vertical intermediate frequency coefficient matrix HL1_ ⁇ , and selecting the low frequency coefficient matrix HL1_LL2_ ⁇ obtained after the transformation as the second low frequency coefficient matrix;
  • the low-frequency coefficient matrix HL1—LL2— ⁇ After DWT is performed on the matrix HL1— ⁇ , the low-frequency coefficient matrix HL1—LL2— ⁇ , the intermediate frequency coefficient matrix HL1—LH2— ⁇ and HL1—HL2— ⁇ , and the high-frequency coefficient matrix HL1— ⁇ 2— ⁇ are obtained, and the low frequency is selected.
  • a coefficient matrix HL1 - LL2 - ⁇ as a second low frequency coefficient matrix
  • steps S301 to S303 are the same as the processing methods of steps S101 to S103.
  • the object is different.
  • the processing object is the original carrier digital image X.
  • the processing object is the image in which the watermark is embedded.
  • step S102 is to perform DCT on the vertical intermediate frequency coefficient matrix HL1 to obtain a matrix HL1-DCT, and then perform DWT on the matrix HL1-DCT, and obtain the transformed result.
  • the low frequency coefficient matrix HL1_DCT_LL2 is used as the first low frequency coefficient matrix
  • the step S103 is to perform DWT on the horizontal intermediate frequency coefficient matrix LH1
  • the low frequency coefficient matrix LH1_LL2 obtained after the transformation is selected as the second low frequency coefficient matrix, and the remaining steps are unchanged. Accordingly, when the watermark is extracted, it is performed in the same manner.
  • Step S304 Comparing the sizes of the two low frequency coefficients at the same embedding position (x, y) on the first low frequency coefficient matrix LH1 - DCT - LL2 - Y and the second low frequency coefficient matrix HL1 - LL2 - Y, and comparing The size relationship is matched to the correspondence between the preset embedded value and the size relationship of the two low frequency coefficients, and the embedded value at each embedded position (x, y) is obtained, thereby obtaining the embedded digital watermark image W.
  • steps S301 to S304 are equivalent to performing reverse transformation on steps S101 to S104, and the embedded digital watermark image can be conveniently extracted based on the correspondence relationship between the embedded value of the same preset and the size relationship of the two low frequency coefficients obtained by the processing. W.
  • the reordering and/or decryption process is performed after the embedded digital watermark image is obtained.
  • the embodiment of the present invention further provides a system, as shown in FIG. 4, comprising a matrix transformation processing module 41, a storage module 42, and a watermark extraction processing module 43, wherein:
  • the matrix transformation processing module 41 is configured to perform DWT on the image embedded with the watermark to obtain two intermediate frequency coefficient matrices; perform DCT on one of the intermediate frequency coefficient matrices, then perform DWT, and then select the obtained low frequency coefficient matrix as the first a low frequency coefficient matrix; performing D WT on another intermediate frequency coefficient matrix, and then selecting the obtained low frequency coefficient matrix as the second low frequency coefficient matrix;
  • the storage module 42 is configured to store a preset embedded value and a correspondence relationship between the first low frequency coefficient and the second low frequency coefficient obtained by the matrix transformation processing module 41;
  • the watermark extraction processing module 43 is configured to compare the size relationship between the first low frequency coefficient and the second low frequency coefficient at the same embedded position on the first low frequency coefficient matrix and the second low frequency coefficient matrix, and obtain the obtained two The size relationship is matched to the correspondence between the preset embedded value in the storage module 42 and the size relationship of the two low frequency coefficients, and the embedded value at each embedded position is obtained, thereby obtaining an embedded digital watermark image.
  • the watermark extraction processing module 43 is further configured to reorder and/or decrypt the watermark after obtaining the embedded digital watermark image.
  • the method and system for image digital watermark embedding and extracting proposed by the invention have good watermark concealability, and can resist histogram equalization, image brightness adjustment, contrast adjustment, salt and pepper noise, multiplicative noise, Gaussian noise, low-pass filtering, JPEG compression of various attacks, watermark robust, can extract digital watermarks without the original carrier image, can be widely used in the field of copyright protection and information security.

Abstract

本发明提出了一种图像数字水印嵌入和提取的方法及系统,该水印嵌入方法包括: 对原载体数字图像进行DWT,获 得2个中频系数矩阵; 对其中一个中频系数矩阵先进行DCT,再进行DWT,取得第一低频系数矩阵;对另一个中频系数矩阵进行DWT变换,取得第二低频系数矩阵;在第一低频系数矩阵和第二低频系数矩阵上各相同的嵌入位置处,根据欲嵌入的水印数值,调整第一低频系数和第二低频系数的大小,使得调整后的两低频系数的大小关系符合预设的嵌入的数值与所述两低频系数的大小关系的对应关系。在提取水印时只需按照水印嵌入方法进行逆向变换即可,水印提取无需原载体数字图像。

Description

一种图像数字水印嵌入和提取的方法及系统
技术领域
本发明涉及计算机信息安全、 版权保护领域, 尤其涉及一种图像数字水 印嵌入和提取的方法及系统。
背景技术
数字水印 (Digital Watermark)技术是指使用信号处理技术在数字化的多媒 体数据中嵌入隐蔽的标记, 这种标记通常是不可见的, 只有通过专用的检测 器或阅读器才能提取。 嵌入数字媒体中的信息必须具有以下 2个基本特性才 能称为数字水印: 易被察觉。
(b)、 鲁棒性: 鲁棒性是指含有水印的数字媒体在经历多种无意或有意的 信号处理过程后, 数字水印仍能保持完整性或仍能被准确鉴别。 可能的信号 处理过程包括信道噪声、 滤波、数 /模转换、模 /数转换、 重釆样、 剪切、位移、 尺度变化以及有损压缩编码等。
在当前各类的水印嵌入和提取技术中, 变换域水印以其可以嵌入大量数 据而不会降低水印的隐蔽性的优势成为比较重要的一种水印技术。 这类技术 一般基于常用的图像变换, 基于局部或是全部的变换, 这些变换包括离散余 弦变换 (Discrete Cosine Transform, DCT)、 离散小波变换 (Discrete Wavelet Transform , DWT)、 离散傅氏变换 (Discrete Fourier Transform, DFT)、 奇异 值分解 (Singular value decomposition , SVD)、 离散哈达马变换 (Discrete Hadamard Transform, DHT)等。 其中基于分块的 DCT是最常用的变换之一, 现在所釆用的静止图像压缩标准 JPEG也是基于分块 DCT的。
最早的基于分块 DCT 的一种数字水印技术方案是由一个密钥随机地选 择图像的一些分块, 在频域系数的中频系数上稍稍改变一个三元组以隐藏二 进制序列信息。 选择在中频分量编码是因为在高频编码易于被各种信号处理 方法所破坏, 而在低频编码则由于人的视觉对低频分量很敏感, 对低频分量 的改变易于被察觉。 该数字水印技术对有损压缩和低通滤波是稳健的。
另一种 DCT数字水印技术是首先把图像分成 8x8的不互相重叠块,在经 过分块 DCT后, 得到由 DCT系数组成的频率块, 然后随机选取一些频率块, 将水印信号嵌入到由密钥控制选择的一些 DCT系数中。该方法是通过对选定 的 DCT系数进行微小变换以满足特定的关系, 以此来表示一个比特的信息。 在水印信息提取时, 则选取相同的 DCT系数, 并根据系数之间的关系抽取比 特信息。
除了上述有代表性的变换域方法外, 还有一些基于变换域的数字水印技 术, 它们当中有相当一部分都是上述方法的改进及发展, 这其中有代表性的 方法是 I.Podichuk和 Zeng Wen-jun提出的方法。 他们的方法^^于静止图像 的 DCT变换或小波变换,研究视觉模型模块返回数字水印应加载在何处及每 处可承受的 J D(Just Noticeable Difference恰好可察觉差另' 的量值 (加载数字 水印的强度上限), 这种水印方法是自适应的。
当前大多基于变换域的数字水印都是私有水印, 需要原始数字媒体才能 检测出数字水印, 私有水印只能作为版权保护来使用, 大大限制了数字水印 技术的应用范围。并且现有数字水印技术只使用单一的变换进行水印的嵌入, 这使得水印的抗攻击力大打折扣。 发明内容
本发明要解决的技术问题是提出一种图像数字水印嵌入和提取的方法及 系统, 用于数字图像的版权保护, 也可以将需要隐秘的数据隐藏起来, 能广 泛使用于版权保护、 信息保密领域。
为解决上述技术问题, 本发明提出一种图像数字水印嵌入方法, 包括: 对原载体数字图像进行 DWT, 获得 2个中频系数矩阵;
对其中一个中频系数矩阵先进行 DCT, 再进行 DWT, 然后选取得到的 低频系数矩阵作为第一低频系数矩阵;
对另一个中频系数矩阵进行 DWT,然后选取得到的低频系数矩阵作为第 二低频系数矩阵;
在第一低频系数矩阵和第二低频系数矩阵上各相同的嵌入位置处, 根据 所述嵌入位置处欲嵌入的水印数值, 调整第一低频系数和第二低频系数的大 小, 使得调整后的两低频系数的大小关系符合预设的嵌入的数值与所述两低 频系数的大小关系的对应关系。
上述图像数字水印嵌入方法还可具有以下特点:
对原载体数字图像进行 DWT后, 获得的 2个中频系数矩阵为水平中频 系数矩阵(LH1 )和垂直中频系数矩阵(HL1 ) 。
上述图像数字水印嵌入方法还可具有以下特点:
在所述根据所述嵌入位置处欲嵌入的水印数值调整第一低频系数和第二 低频系数的大小的步骤之前, 上述方法还包括: 对水印进行置乱和 /或加密处 理。
为解决上述技术问题, 本发明还提出一种图像数字水印嵌入系统, 包括 矩阵变换处理模块、 存储模块和水印嵌入处理模块, 其中:
所述矩阵变换处理模块设置为: 对原载体数字图像进行 DWT, 获得 2个 中频系数矩阵; 对其中一个中频系数矩阵先进行 DCT, 再进行 DWT, 然后 选取得到的低频系数矩阵作为第一低频系数矩阵; 对另一个中频系数矩阵进 行 D WT, 然后选取得到的低频系数矩阵作为第二低频系数矩阵;
所述存储模块设置为: 存储一预设的嵌入的数值与经所述矩阵变换处理 模块变换后得到的第一低频系数与第二低频系数的大小关系的对应关系; 所述水印嵌入处理模块设置为: 在第一低频系数矩阵和第二低频系数矩 阵上各相同的嵌入位置处, 根据所述嵌入位置处欲嵌入的水印数值, 调整第 一低频系数和第二低频系数的大小, 使得调整后的两低频系数的大小关系符 合所述存储模块中预设的嵌入的数值与所述两低频系数的大小关系的对应关 系。
上述图像数字水印嵌入系统还可具有以下特点:
所述水印嵌入处理模块还设置为: 在根据该嵌入位置处欲嵌入的水印数 值调整第一低频系数与第二低频系数的大小之前, 对水印进行置乱和 /或加密 处理。
为解决上述技术问题, 本发明还提出一种图像数字水印提取方法, 包括: 对嵌入了水印的图像进行 D WT , 获得 2个中频系数矩阵;
对其中一个中频系数矩阵先进行 DCT, 再进行 DWT, 然后选取得到的 低频系数矩阵作为第一低频系数矩阵;
对另一个中频系数矩阵进行 DWT,然后选取得到的低频系数矩阵作为第 二低频系数矩阵;
在第一低频系数矩阵和第二低频系数矩阵上各相同的嵌入位置处, 比较 第一低频系数与第二低频系数的大小关系, 并将得到的两者的大小关系去匹 配预设的嵌入的数值与该两低频系数的大小关系的对应关系, 得到各嵌入位 置处的嵌入数值, 从而获得嵌入的数字水印图像。
上述图像数字水印提取方法还可具有以下特点:
对嵌入了水印的图像进行 DWT后, 获得的 2个中频系数矩阵为水平中 频系数矩阵(LH1_Y )和垂直中频系数矩阵(HL1_Y ) 。
上述图像数字水印提取方法还可包括:
如果嵌入水印时, 对水印进行了置乱和 /或加密处理, 则在获得嵌入的数 字水印图像后, 进行重排序和 /或解密处理。
为解决上述技术问题, 本发明还提出一种图像数字水印提取系统, 包括 矩阵变换处理模块、 存储模块和水印提取处理模块, 其中:
所述矩阵变换处理模块设置为: 对嵌入了水印的图像进行 DWT, 获得 2 个中频系数矩阵; 对其中一个中频系数矩阵先进行 DCT, 再进行 DWT, 然 后选取得到的低频系数矩阵作为第一低频系数矩阵; 对另一个中频系数矩阵 进行 D WT, 然后选取得到的低频系数矩阵作为第二低频系数矩阵;
所述存储模块设置为: 存储一预设的嵌入的数值与经所述矩阵变换处理 模块变换后得到的第一低频系数与第二低频系数的大小关系的对应关系; 所述水印提取处理模块设置为: 在第一低频系数矩阵和第二低频系数矩 阵上各相同的嵌入位置处, 比较第一低频系数与第二低频系数的大小关系, 并将得到的两者的大小关系去匹配所述存储模块中预设的嵌入的数值与该两 低频系数的大小关系的对应关系, 得到各嵌入位置处的嵌入数值, 从而获得 嵌入的数字水印图像。
上述图像数字水印提取系统还可具有以下特点:
所述水印提取处理模块还设置为: 在获得嵌入的数字水印图像后, 对水 印进行重排序和 /或解密处理。
本发明提出的一种图像数字水印嵌入和提取的方法及系统, 水印隐蔽性 好, 可以抗直方图均衡化、 图像亮度调整、 对比度调整、 椒盐噪声、 乘性噪 声、 高斯噪声、 低通滤波、 联合图像专家组(JPEG )压缩各种攻击, 水印鲁 棒性强, 可以在没有原始载体图像的情况下也能提取出数字水印, 能广泛使 用于版权保护、 信息保密领域。
附图概述
图 1是本发明实施例一种图像数字水印嵌入方法流程图;
图 2是本发明实施例一种图像数字水印嵌入系统方框图;
图 3是本发明实施例一种图像数字水印提取方法流程图;
图 4是本发明实施例一种图像数字水印提取系统方框图。
本发明的较佳实施方式
在本发明实施例中, 假设原载体数字图像为 X, 数字水印图像为\^。 下 面将结合附图详细说明本发明实施方案。
参考图 1 , 该图示出了本发明实施例一种图像数字水印嵌入方法, 包括 如下步骤: 步骤 S 101:对原载体数字图像 X进行 DWT,得到 2个中频系数矩阵 LH1 和 HL1 ;
对原载体数字图像 X进行 DWT后将得到: 低频系数矩阵 LL1、 水平中 频系数矩阵 LH1、 垂直中频系数矩阵 HL1 , 以及高频系数矩阵 HH1 , 选取 2 个中频系数矩阵 LHl和 HL1进行下述操作。
步骤 S102: 对水平中频系数矩阵 LH1进行 DCT, 得到矩阵 LHl— DCT, 然后对矩阵 LHl— DCT 进行 DWT , 选取变换后得到的低频系数矩阵 LH1 DCT LL2作为第一低频系数矩阵;
对矩阵 LHl— DCT进行 DWT后, 将得到低频系数矩阵 LHl— DCT— LL2、 中频系数矩阵 LHl— DCT— LH2 和 LHl— DCT— HL2 , 以及高频系数矩阵 LH1 DCT HH2 ,选取其中的低频系数矩阵 LHl— DCT— LL2作为第一低频系数 矩阵;
步骤 S103: 对垂直中频系数矩阵 HL1进行 DWT, 选取变换后得到的低 频系数矩阵 HL1— LL2作为第二低频系数矩阵;
对矩阵 HL1进行 DWT后, 将得到低频系数矩阵 HL1— LL2、 中频系数矩 阵 HL1— LH2和 HL1— HL2, 以及高频系数矩阵 HL1— HH2, 选取其中的低频系 数矩阵 HL1— LL2作为第二低频系数矩阵;
步骤 S104: 在第一低频系数矩阵 LHl— DCT— LL2 和第二低频系数矩阵 HL1_LL2上各相同的嵌入位置(x,y )处, 根据该嵌入位置(x,y )处欲嵌入的 水印数值, 调整两个低频系数的大小, 使得调整后的两低频系数之间的大小 关系符合预设的嵌入的数值与该两低频系数的大小关系的对应关系。
欲嵌入的水印为一有意义的二值图像, 即在各嵌入位置(x,y )上嵌入的 数值为 0或者 1 ;
预设的嵌入的数值与两低频系数的大小关系的对应关系可以是: 如果在 某一嵌入位置(i , j)欲嵌入 0 , 则对应该位置(i , j)上的第一低频系数 LHl_DCT_LL2(i, j) <第二低频系数 HL1— LL2(i, j); 如果在该嵌入位置 (i, j) 欲嵌入 1 , 则对应该位置 (i, j)上的第一低频系数 LHl—DCT—LL2(i, j) >第二低 频系数 HL1— LL2(i, j), 其中, l<i<M, Kj<N, M为原载体数字图像 X高度 的 1/4, N为原载体数字图像 X宽度的 1/4。
当然, 该对应关系也可以设置为其他, 本发明并不以此为限制, 只要令 不同的嵌入数值对应于不同的第一低频系数 LHl— DCT— LL2(i, j)与第二低频 系数 HL1— LL2(i, j)的大小关系, 根据该对应关系, 在获知嵌入数值时可明确 得到其对应的第一低频系数 LH1— DCT— LL2(i , j)与第二低频系数 HLl— LL2(i , j)的大小关系, 反之, 在获知第一低频系数 LH1— DCT— LL2(i , j)与第二低频系 数 HLl— LL2(i , j)的大小关系时, 也可明确得到其对应的嵌入数值即可。
LH1 DCT LL2 ( x,y )与 HL1_LL2 ( x,y ) 的大小关系若需要调整, 可按 照一定的规则进行调整, 例如可以通过交换这两个系数的取值来实现, 或者 如果两个数据相等, 则可进行微调。
步骤 S101至步骤 S104完成了水印的嵌入, 通过对经上述变换后得到的 两个低频系数进行微调,在水印的隐蔽性和鲁棒性之间找到了较好的平衡点。
在上述方法中, 嵌入的图像是以明文的形式出现, 如果想嵌入隐秘图像 的时候, 如果被第三方截获, 不能保密。 针对这种情况, 在本发明实施例中 可以将数字水印图像为 W进行置乱和 /或加密处理后再嵌入, 将有效防止信 息泄露。
在另一实施例中,也可以是, 上述步骤 S102为对垂直中频系数矩阵 HL1 进行 DCT, 得到矩阵 HLl— DCT, 然后对矩阵 HLl— DCT进行 DWT变换, 选 取变换后得到的低频系数矩阵 HLl— DCT— LL2作为第一低频系数矩阵, 上述 步骤 S103为对水平中频系数矩阵 LH1进行 DWT, 选取变换后得到的低频系 数矩阵 LH1— LL2作为第二低频系数矩阵, 其余步骤不变。
为了实现上述水印嵌入方法,本发明实施例还提供了一种水印嵌入系统, 如图 2所示, 包括矩阵变换处理模块 21、 存储模块 22、 水印嵌入处理模块 23 , 其中:
所述矩阵变换处理模块 21 , 用以对原载体数字图像进行 DWT, 获得 2 个中频系数矩阵; 对其中一个中频系数矩阵先进行 DCT, 再进行 DWT, 然 后选取得到的低频系数矩阵作为第一低频系数矩阵; 对另一个中频系数矩阵 进行 DWT, 然后选取得到的低频系数矩阵作为第二低频系数矩阵;
所述存储模块 22, 用以存储一预设的嵌入的数值与经所述矩阵变换处理 模块 21变换后得到的第一低频系数与第二低频系数的大小关系的对应关系; 所述水印嵌入处理模块 23 , 用以在第一低频系数矩阵和第二低频系数矩 阵上各相同的嵌入位置处, 根据所述嵌入位置处欲嵌入的水印数值, 调整第 一低频系数和第二低频系数的大小, 使得调整后的两低频系数的大小关系符 合所述存储模块 22 中预设的嵌入的数值与所述两低频系数的大小关系的对 应关系。
进一步地, 所述水印嵌入处理模块 23 , 还用以在根据该嵌入位置处欲嵌 入的水印数值调整两个低频系数的大小之前, 对水印进行置乱和 /或加密处 理。
下面参考图 3 , 该图示出了本发明实施例一种图像数字水印提取方法, 对釆用图 1所述的方法嵌入的水印进行提取, 包括如下步骤:
步骤 S301 : 对嵌入水印的图像 Y进行 DWT, 得到 2个中频系数矩阵 LH1 Y和 HL1— Y;
对嵌入水印的图像 Y进行 DWT后将得到: 低频系数矩阵 LL1— Y、 水平 中频系数矩阵 LHl— Υ、垂直中频系数矩阵 HL1— Υ,以及高频系数矩阵 HH1— Υ, 选取 2中频系数矩阵 LHl— Υ和 HL1— Υ进行下述操作。
步骤 S302 : 对水平中频系数矩阵 LHl— Υ 进行 DCT , 得到矩阵 LH1 DCT Y, 然后对矩阵 LHl— DCT— Υ进行 DWT, 选取变换后得到的低频 系数矩阵 LHl— DCT— LL2— Υ作为第一低频系数矩阵;
对矩阵 LHl— DCT— Υ 进行 DWT 后, 将得到低频系数矩阵 LHl— DCT— LL2— Υ、 中频系数矩阵 LHl— DCT— LH2— Υ和 LHl— DCT— HL2— Υ, 以及高频系数矩阵 LHl— DCT— ΗΗ2— Υ , 选取其中的低频系数矩阵 LH1 DCT LL2 Y作为第一低频系数矩阵;
步骤 S303: 对垂直中频系数矩阵 HL1— Υ进行 DWT, 选取变换后得到的 低频系数矩阵 HL1— LL2— Υ作为第二低频系数矩阵;
对矩阵 HL1— Υ进行 DWT后, 将得到低频系数矩阵 HL1— LL2— Υ、 中频 系数矩阵 HL1— LH2— Υ和 HL1— HL2— Υ, 以及高频系数矩阵 HL1— ΗΗ2— Υ, 选 取其中的低频系数矩阵 HL1— LL2— Υ作为第二低频系数矩阵;
上述步骤 S301〜步骤 S303与步骤 S101〜步骤 S103的处理方法相同, 处 理对象不同, 在图 1嵌入水印过程中, 处理对象为原载体数字图像 X, 在图 3 提取水印过程中, 处理对象为嵌入了水印的图像丫。
如果在水印嵌入时, 釆用的是上述的另一种方法, 即步骤 S102为对垂直 中频系数矩阵 HL1进行 DCT, 得到矩阵 HLl— DCT, 然后对矩阵 HLl— DCT 进行 DWT,选取变换后得到的低频系数矩阵 HLl— DCT— LL2作为第一低频系 数矩阵, 步骤 S103为对水平中频系数矩阵 LH1进行 DWT, 选取变换后得到 的低频系数矩阵 LH1— LL2作为第二低频系数矩阵,其余步骤不变,则相应地, 在水印提取时, 按照相同的方法执行。 步骤 S304:在第一低频系数矩阵 LH1— DCT— LL2— Y和第二低频系数矩阵 HLl— LL2— Y上各相同的嵌入位置( x,y )处比较两个低频系数的大小, 并将比 较出的大小关系去匹配预设的嵌入的数值与该两低频系数的大小关系的对应 关系, 得到各嵌入位置(x,y )处的嵌入数值, 从而获得嵌入的数字水印图像 W。
上述步骤 S301〜步骤 S304相当于对步骤 S101〜步骤 S104进行逆向变换, 基于同一预设的嵌入的数值与处理得到的 2个低频系数的大小关系的对应关 系可以方便地提取出嵌入的数字水印图像 W。
如果嵌入水印时, 对水印进行了置乱和 /或加密处理, 则在获得嵌入的数 字水印图像后, 还进行重排序和 /或解密处理。
为了实现上述水印嵌入方法, 本发明实施例还提供了一种系统, 如图 4 所示, 包括矩阵变换处理模块 41、 存储模块 42、 水印提取处理模块 43 , 其 中:
所述矩阵变换处理模块 41 , 用以对嵌入了水印的图像进行 DWT, 获得 2 个中频系数矩阵; 对其中一个中频系数矩阵先进行 DCT, 再进行 DWT, 然 后选取得到的低频系数矩阵作为第一低频系数矩阵; 对另一个中频系数矩阵 进行 D WT, 然后选取得到的低频系数矩阵作为第二低频系数矩阵;
所述存储模块 42, 用以存储一预设的嵌入的数值与经所述矩阵变换处理 模块 41变换后得到的第一低频系数与第二低频系数的大小关系的对应关系; 所述水印提取处理模块 43 , 用以在第一低频系数矩阵和第二低频系数矩 阵上各相同的嵌入位置处, 比较第一低频系数与第二低频系数的大小关系, 并将得到的两者的大小关系去匹配所述存储模块 42 中预设的嵌入的数值与 该两低频系数的大小关系的对应关系, 得到各嵌入位置处的嵌入数值, 从而 获得嵌入的数字水印图像。
进一步地, 所述水印提取处理模块 43 , 还用以在获得嵌入的数字水印图 像后, 对水印进行重排序和 /或解密处理。
本发明还可有其他多种实施例,在不背离本发明精神及其实质的情况下 , 改变和变形都应属于本发明所附的权利要求的保护范围。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序 来指令相关硬件完成, 所述程序可以存储于计算机可读存储介质中, 如只读 存储器、 磁盘或光盘等。 可选地, 上述实施例的全部或部分步骤也可以使用 一个或多个集成电路来实现。 相应地, 上述实施例中的各模块 /单元可以釆用 硬件的形式实现, 也可以釆用软件功能模块的形式实现。 本发明不限制于任 何特定形式的硬件和软件的结合。
工业实用性
本发明提出的一种图像数字水印嵌入和提取的方法及系统, 水印隐蔽性 好, 可以抗直方图均衡化、 图像亮度调整、 对比度调整、 椒盐噪声、 乘性噪 声、 高斯噪声、 低通滤波、 JPEG压缩各种攻击, 水印鲁棒性强, 可以在没有 原始载体图像的情况下也能提取出数字水印, 能广泛使用于版权保护、 信息 保密领域。

Claims

权 利 要 求 书
1、 一种图像数字水印嵌入方法, 其包括:
对原载体数字图像进行离散小波变换(DWT ) , 获得 2个中频系数矩阵; 对其中一个中频系数矩阵先进行离散余弦变换(DCT ) , 再进行 DWT, 然后选取得到的低频系数矩阵作为第一低频系数矩阵;
对另一个中频系数矩阵进行 DWT,然后选取得到的低频系数矩阵作为第 二低频系数矩阵;
在第一低频系数矩阵和第二低频系数矩阵上各相同的嵌入位置处, 根据 所述嵌入位置处欲嵌入的水印数值, 调整第一低频系数和第二低频系数的大 小, 使得调整后的两低频系数的大小关系符合预设的嵌入的数值与所述两低 频系数的大小关系的对应关系。
2、 如权利要求 1所述的图像数字水印嵌入方法, 其中,
对原载体数字图像进行 DWT后, 获得的所述 2个中频系数矩阵为水平 中频系数矩阵(LH1 )和垂直中频系数矩阵(HL1 ) 。
3、 如权利要求 1所述的图像数字水印嵌入方法, 其中,
在所述根据所述嵌入位置处欲嵌入的水印数值调整第一低频系数和第二 低频系数的大小的步骤之前, 所述方法还包括: 对所述水印进行置乱和 /或加 密处理。
4、 一种图像数字水印嵌入系统,其包括矩阵变换处理模块、存储模块和 水印嵌入处理模块, 其中:
所述矩阵变换处理模块设置为: 对原载体数字图像进行离散小波变换 ( DWT ) , 获得 2个中频系数矩阵; 对其中一个中频系数矩阵先进行离散余 弦变换(DCT ) , 再进行 DWT, 然后选取得到的低频系数矩阵作为第一低频 系数矩阵; 对另一个中频系数矩阵进行 DWT, 然后选取得到的低频系数矩阵 作为第二低频系数矩阵;
所述存储模块设置为: 存储一预设的嵌入的数值与经所述矩阵变换处理 模块变换后得到的第一低频系数与第二低频系数的大小关系的对应关系; 所述水印嵌入处理模块设置为: 在第一低频系数矩阵和第二低频系数矩 阵上各相同的嵌入位置处, 根据所述嵌入位置处欲嵌入的水印数值, 调整第 一低频系数和第二低频系数的大小, 使得调整后的两低频系数的大小关系符 合所述存储模块中预设的嵌入的数值与所述两低频系数的大小关系的对应关 系。
5、 如权利要求 4所述的图像数字水印嵌入系统, 其中:
所述水印嵌入处理模块还设置为: 在根据该嵌入位置处欲嵌入的水印数 值调整第一低频系数与第二低频系数的大小之前, 对水印进行置乱和 /或加密 处理。
6、 一种图像数字水印提取方法, 其包括:
对嵌入了水印的图像进行离散小波变换(DWT ) , 获得 2个中频系数矩 阵;
对其中一个中频系数矩阵先进行离散余弦变换(DCT ) , 再进行 DWT, 然后选取得到的低频系数矩阵作为第一低频系数矩阵;
对另一个中频系数矩阵进行 DWT,然后选取得到的低频系数矩阵作为第 二低频系数矩阵;
在第一低频系数矩阵和第二低频系数矩阵上各相同的嵌入位置处, 比较 第一低频系数与第二低频系数的大小关系, 并将得到的两者的大小关系去匹 配预设的嵌入的数值与该两低频系数的大小关系的对应关系, 得到各嵌入位 置处的嵌入数值, 从而获得嵌入的数字水印图像。
7、 如权利要求 6所述的图像数字水印提取方法, 其中:
对嵌入了水印的图像进行 DWT后, 获得的所述 2个中频系数矩阵为水 平中频系数矩阵(LH1— Y )和垂直中频系数矩阵(HL1— Y ) 。
8、 如权利要求 7所述的图像数字水印提取方法, 该方法还包括: 如果嵌入水印时, 对水印进行了置乱和 /或加密处理, 则在获得嵌入的数 字水印图像后, 进行重排序和 /或解密处理。
9、 一种图像数字水印提取系统,其包括矩阵变换处理模块、存储模块和 水印提取处理模块, 其中:
所述矩阵变换处理模块设置为: 对嵌入了水印的图像进行离散小波变换 ( DWT ) , 获得 2个中频系数矩阵; 对其中一个中频系数矩阵先进行离散余 弦变换(DCT ) , 再进行 DWT, 然后选取得到的低频系数矩阵作为第一低频 系数矩阵; 对另一个中频系数矩阵进行 DWT, 然后选取得到的低频系数矩 阵作为第二低频系数矩阵; 所述存储模块设置为: 存储一预设的嵌入的数值与经所述矩阵变换处理 模块变换后得到的第一低频系数与第二低频系数的大小关系的对应关系; 所述水印提取处理模块设置为: 在第一低频系数矩阵和第二低频系数矩 阵上各相同的嵌入位置处, 比较第一低频系数与第二低频系数的大小关系, 并将得到的两者的大小关系去匹配所述存储模块中预设的嵌入的数值与该两 低频系数的大小关系的对应关系, 得到各嵌入位置处的嵌入数值, 从而获得 嵌入的数字水印图像。
10、 如权利要求 9所述的图像数字水印提取系统, 其中:
所述水印提取处理模块还设置为: 在获得嵌入的数字水印图像后, 对水 印进行重排序和 /或解密处理。
PCT/CN2010/074143 2009-09-03 2010-06-21 一种图像数字水印嵌入和提取的方法及系统 WO2011026365A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
PL10813287T PL2428930T3 (pl) 2009-09-03 2010-06-21 Sposób i system do osadzania i wyodrębniania cyfrowego znaku wodnego obrazu
EP10813287.9A EP2428930B1 (en) 2009-09-03 2010-06-21 Method and system for embedding and extracting image digital watermark
ES10813287T ES2431595T3 (es) 2009-09-03 2010-06-21 Método y sistema de incorporación y de extracción de una marca de agua digital de imagen
US13/258,003 US8615103B2 (en) 2009-09-03 2010-06-21 Method and system for embedding and extracting image digital watermark

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2009101712892A CN101699508B (zh) 2009-09-03 2009-09-03 一种图像数字水印嵌入和提取的方法及系统
CN200910171289.2 2009-09-03

Publications (1)

Publication Number Publication Date
WO2011026365A1 true WO2011026365A1 (zh) 2011-03-10

Family

ID=42147966

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/074143 WO2011026365A1 (zh) 2009-09-03 2010-06-21 一种图像数字水印嵌入和提取的方法及系统

Country Status (6)

Country Link
US (1) US8615103B2 (zh)
EP (1) EP2428930B1 (zh)
CN (1) CN101699508B (zh)
ES (1) ES2431595T3 (zh)
PL (1) PL2428930T3 (zh)
WO (1) WO2011026365A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109493270A (zh) * 2018-11-07 2019-03-19 中南大学 一种基于slt-dm的水印图像还原方法
CN109543425A (zh) * 2018-10-26 2019-03-29 浙江师范大学 一种基于张量分解的图像数据隐藏方法
CN111275602A (zh) * 2020-01-16 2020-06-12 深圳市广道高新技术股份有限公司 人脸图像安全保护方法、系统及存储介质
CN111583085A (zh) * 2020-04-16 2020-08-25 天津大学 一种基于变换域的彩色图像数字水印方法
CN114140310A (zh) * 2021-12-07 2022-03-04 合肥工业大学 一种基于dct变换的数字水印硬件电路

Families Citing this family (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101699508B (zh) 2009-09-03 2012-01-11 中兴通讯股份有限公司 一种图像数字水印嵌入和提取的方法及系统
CN102223561B (zh) * 2011-07-01 2013-07-31 宁波大学 一种立体视频图像的盲水印嵌入和提取方法
CN102254298B (zh) * 2011-08-17 2013-06-19 北京新媒传信科技有限公司 数字水印的加解密方法和系统
US9245310B2 (en) * 2013-03-15 2016-01-26 Qumu Corporation Content watermarking
CN103793874B (zh) * 2014-02-14 2017-02-08 中北大学 一种新型保真鲁棒数字水印方法
US9871658B2 (en) * 2015-03-24 2018-01-16 Semiconductor Components Industries, Llc Imaging systems with data encryption and embedding capabalities
CN104794676B (zh) * 2015-04-28 2017-09-29 东南大学 一种保密性强的抗打印扫描数字图像水印方法
CN105072453B (zh) * 2015-07-21 2018-07-24 河海大学 一种面向移动终端的视频水印方法
CN105512999B (zh) * 2015-08-24 2019-08-27 湖南工业大学 一种双变换的彩色图像全息水印方法
CN106056522A (zh) * 2016-05-16 2016-10-26 西安邮电大学 一种基于曲波变换的数字水印系统
CN107633476B (zh) * 2017-09-25 2021-07-27 河南师范大学 一种基于lwt-svd-dct算法的水印嵌入和提取方法
CN108648130B (zh) * 2018-04-04 2022-05-27 宁波大学 一种具有版权保护和篡改定位功能的全盲数字水印方法
CN109086613A (zh) * 2018-07-23 2018-12-25 石家庄国创明慧物联网科技有限公司 一种图形标防伪方法及系统
CN111445374A (zh) * 2018-12-29 2020-07-24 北京奇虎科技有限公司 用于将隐数字水印嵌入图像的水印模板生成方法及装置
CN109685710A (zh) * 2018-12-29 2019-04-26 北京奇虎科技有限公司 一种图像版权隐数字水印嵌入的方法及装置
CN109829846B (zh) * 2019-01-29 2023-07-07 鲁东大学 一种基于二维离散余弦变换的数字图像盲水印方法
CN109903215A (zh) * 2019-04-08 2019-06-18 上海理工大学 基于模运算的小波系数调整水印方法
CN110084733B (zh) * 2019-04-19 2021-02-02 中国科学院自动化研究所 文本图像水印的嵌入方法及系统、提取方法及系统
CN110148100A (zh) * 2019-05-10 2019-08-20 腾讯科技(深圳)有限公司 一种图像变换方法、装置、存储介质及计算机设备
CN112149062A (zh) * 2019-06-26 2020-12-29 中电万维信息技术有限责任公司 基于水印的电子印章生成方法、装置、系统及存储介质
CN110958104B (zh) * 2019-11-15 2021-11-02 电子科技大学 一种基于16qam部分位置嵌入信号水印的通信方法
CN110930288B (zh) * 2019-11-21 2023-09-08 上海鹰瞳医疗科技有限公司 在眼底图像中添加和提取水印的方法及设备
CN111199508B (zh) * 2020-01-03 2022-05-17 鲁东大学 一种融合dct和dht的空域彩色数字图像盲水印方法
CN111325653B (zh) * 2020-02-04 2023-07-25 淮阴师范学院 一种彩色图像数字水印实现方法
CN111681153A (zh) * 2020-06-08 2020-09-18 重庆邮电大学 基于非下采样轮廓波变换的cs-svd的鲁棒数字水印算法
CN112070650B (zh) * 2020-09-15 2023-12-22 中国科学院信息工程研究所 一种针对全景图像的水印嵌入及检测方法
CN115564634B (zh) * 2022-12-05 2023-05-02 杭州海康威视数字技术股份有限公司 视频对抗水印嵌入方法、装置、电子设备及存储介质
CN116563081B (zh) * 2023-07-06 2023-09-26 云南日报报业集团 一种空间域盲图像水印方法
CN116645259B (zh) * 2023-07-18 2023-09-26 云南日报报业集团 一种基于rdwt-hd-svd的水印嵌入提取方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6556689B1 (en) * 1998-05-29 2003-04-29 University Of Delaware Watermarking methods for digital images and videos
US20030095682A1 (en) * 2001-11-20 2003-05-22 Sanghyun Joo Apparatus and method for embedding and extracting digital watermarks based on wavelets
JP2004221925A (ja) * 2003-01-15 2004-08-05 Ricoh Co Ltd 画像処理装置、プログラム及び記憶媒体
CN1658223A (zh) * 2005-03-23 2005-08-24 中山大学 一种基于特征的数字图象认证方法
CN101042769A (zh) * 2007-01-12 2007-09-26 中国人民解放军国防科学技术大学 一种基于小波与dct双重域的主动式数字图像内容鉴别方法
CN101699508A (zh) * 2009-09-03 2010-04-28 中兴通讯股份有限公司 一种图像数字水印嵌入和提取的方法及系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6226387B1 (en) 1996-08-30 2001-05-01 Regents Of The University Of Minnesota Method and apparatus for scene-based video watermarking
US6975733B1 (en) * 1999-09-10 2005-12-13 Markany, Inc. Watermarking of digital images using wavelet and discrete cosine transforms
KR100337954B1 (ko) * 1999-12-16 2002-05-23 최종욱 텍스트 워터마크 삽입 방법 및 장치
WO2002017214A2 (en) 2000-08-24 2002-02-28 Digimarc Corporation Watermarking recursive hashes into frequency domain regions and wavelet based feature modulation watermarks
US7634525B2 (en) * 2005-06-27 2009-12-15 The Aerospace Corporation Haar wavelet transform embedded lossless type IV discrete cosine transform
CN101038771A (zh) * 2006-03-18 2007-09-19 辽宁师范大学 用于音乐作品版权保护的数字水印新方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6556689B1 (en) * 1998-05-29 2003-04-29 University Of Delaware Watermarking methods for digital images and videos
US20030095682A1 (en) * 2001-11-20 2003-05-22 Sanghyun Joo Apparatus and method for embedding and extracting digital watermarks based on wavelets
JP2004221925A (ja) * 2003-01-15 2004-08-05 Ricoh Co Ltd 画像処理装置、プログラム及び記憶媒体
CN1658223A (zh) * 2005-03-23 2005-08-24 中山大学 一种基于特征的数字图象认证方法
CN101042769A (zh) * 2007-01-12 2007-09-26 中国人民解放军国防科学技术大学 一种基于小波与dct双重域的主动式数字图像内容鉴别方法
CN101699508A (zh) * 2009-09-03 2010-04-28 中兴通讯股份有限公司 一种图像数字水印嵌入和提取的方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2428930A4 *

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109543425A (zh) * 2018-10-26 2019-03-29 浙江师范大学 一种基于张量分解的图像数据隐藏方法
CN109543425B (zh) * 2018-10-26 2023-03-10 浙江师范大学 一种基于张量分解的图像数据隐藏方法
CN109493270A (zh) * 2018-11-07 2019-03-19 中南大学 一种基于slt-dm的水印图像还原方法
CN109493270B (zh) * 2018-11-07 2023-03-24 中南大学 一种基于slt-dm的水印图像还原方法
CN111275602A (zh) * 2020-01-16 2020-06-12 深圳市广道高新技术股份有限公司 人脸图像安全保护方法、系统及存储介质
CN111275602B (zh) * 2020-01-16 2023-06-30 深圳市广道数字技术股份有限公司 人脸图像安全保护方法、系统及存储介质
CN111583085A (zh) * 2020-04-16 2020-08-25 天津大学 一种基于变换域的彩色图像数字水印方法
CN111583085B (zh) * 2020-04-16 2023-07-04 天津大学 一种基于变换域的彩色图像数字水印方法
CN114140310A (zh) * 2021-12-07 2022-03-04 合肥工业大学 一种基于dct变换的数字水印硬件电路
CN114140310B (zh) * 2021-12-07 2024-02-13 合肥工业大学 一种基于dct变换的数字水印硬件电路

Also Published As

Publication number Publication date
US8615103B2 (en) 2013-12-24
CN101699508B (zh) 2012-01-11
PL2428930T3 (pl) 2013-12-31
EP2428930B1 (en) 2013-08-14
EP2428930A1 (en) 2012-03-14
CN101699508A (zh) 2010-04-28
EP2428930A4 (en) 2012-10-10
ES2431595T3 (es) 2013-11-27
US20120163652A1 (en) 2012-06-28

Similar Documents

Publication Publication Date Title
WO2011026365A1 (zh) 一种图像数字水印嵌入和提取的方法及系统
Totla et al. Comparative analysis of watermarking in digital images using DCT & DWT
Meenpal et al. Digital watermarking technique using dual tree complex wavelet transform
Luo et al. Fast and robust watermarking of JPEG files
Yang et al. A semi-fragile reversible data hiding by coefficient-bias algorithm
Rajput et al. A novel technique for RGB invisible watermarking based on 2-DWT-DCT algorithm
Singh et al. Spatial and frequency domain for grey level digital images
Kumar et al. Review of video watermarking techniques
Negrat et al. Variable length encoding in multiple frequency domain steganography
Guo et al. Digital image watermarking for joint ownership verification without a trusted dealer
Shahid et al. Digital video watermarking: Issues and challenges
Rana et al. Digital Watermarking Image Using Discrete Wavelet Transform and Discrete Cosine Transform with Noise Identification
Sarma et al. An Entropy based Video Watermarking Scheme
Coria et al. An access control video watermarking method that is robust to geometric distortions
Lee et al. Reversible data embedding for tamper-proof watermarks
Goswami et al. Coloured and Gray Scale Image Steganography using Block Level DWT DCT Transformation
Elbaşi Robust mpeg watermarking in dwt four bands
Prajapati Transform based digital image watermarking techniques for image authentication
Chaudhary et al. Digital video watermarking scheme using wavelets with MATLAB
Ghoshal et al. Controlled data hiding technique for color image authentication in frequency domain (CDHTCIAFD)
AU2021101400A4 (en) An improved cyber security system with digital watermarking using combined transformation approach
Shekhawat et al. A robust watermarking technique based on biorthogonal wavelet transform
Thanh et al. Extended DCT domain for improving the quality of watermarked image
HOVANČÁK et al. Digital image watermarking in different color models
Abusham et al. Fusion of Watermarking and Steganography for Protecting Image Ownership

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10813287

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 13258003

Country of ref document: US

WWE Wipo information: entry into national phase

Ref document number: 2010813287

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE