WO2012145909A1 - 基于图像色度的彩色数字图像篡改检测方法 - Google Patents

基于图像色度的彩色数字图像篡改检测方法 Download PDF

Info

Publication number
WO2012145909A1
WO2012145909A1 PCT/CN2011/073444 CN2011073444W WO2012145909A1 WO 2012145909 A1 WO2012145909 A1 WO 2012145909A1 CN 2011073444 W CN2011073444 W CN 2011073444W WO 2012145909 A1 WO2012145909 A1 WO 2012145909A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
color space
edge image
calculating
training
Prior art date
Application number
PCT/CN2011/073444
Other languages
English (en)
French (fr)
Inventor
谭铁牛
董晶
王伟
Original Assignee
中国科学院自动化研究所
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 中国科学院自动化研究所 filed Critical 中国科学院自动化研究所
Priority to PCT/CN2011/073444 priority Critical patent/WO2012145909A1/zh
Priority to CN2011800044127A priority patent/CN102959588A/zh
Publication of WO2012145909A1 publication Critical patent/WO2012145909A1/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
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00005Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for relating to image data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00General purpose image data processing
    • G06T1/0021Image watermarking
    • G06T1/0028Adaptive watermarking, e.g. Human Visual System [HVS]-based watermarking
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/457Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by analysing connectivity, e.g. edge linking, connected component analysis or slices
    • 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
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00026Methods therefor
    • H04N1/00037Detecting, i.e. determining the occurrence of a predetermined state
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2201/00General purpose image data processing
    • G06T2201/005Image watermarking
    • G06T2201/0201Image watermarking whereby only tamper or origin are detected and no embedding takes place

Definitions

  • the present invention relates to the field of pattern recognition, and more particularly to a tamper detection method for color digital images. Background technique
  • An object of the present invention is to provide a color digital image tampering detection method based on image chromaticity, which can realize accurate and efficient automatic detection of color image tampering.
  • a color digital image tampering detection method based on image chromaticity includes a training process and a classification process, and the training process includes:
  • the feature marked with the category information is input into the SVM classifier for training, and the classifier model is obtained;
  • the classification process includes:
  • the classifier model loaded into the training process is classified to obtain a decision result.
  • the method of the present invention can be used for tamper detection of multimedia data such as images. It can be used for content collection, identification and other related products such as Internet multimedia, news report pictures, judicial photo evidence, etc. Since the present invention does not need to embed the watermark information into the image in advance, it is only necessary to analyze the image itself and the detection feature extraction is simple, fast, and efficient, so that it can be effectively applied in a content forensic environment such as large-scale data communication and multimedia transmission.
  • DRAWINGS 1 is a flow chart of a method for detecting color image tampering based on image chromaticity.
  • FIG. 2 is an image to be detected used in an embodiment of the present invention; the image is a tamper image.
  • the zebras in the picture are from other pictures.
  • (a) is an image to be tested, and the image is a tamper image.
  • (b), (c), and (d) are edge images of the Y, Cb, and Cr components after the color image is converted to the YCbCr color space. detailed description
  • the technical solution adopted by the present invention includes the following steps:
  • Training step S1 First, color space conversion of the color image of the marked category information (tampering class or real class) in the training library, taking the chrominance component thereof, and then calculating the Markov steady state distribution of the edge image as a feature vector. Secondly, classifier training is performed on the extracted feature sets to obtain model parameters of the classifier.
  • Sorting step S2 performing color space conversion on the arbitrarily input color image, taking the chrominance component thereof, calculating the Markov steady state distribution of the edge image as a feature, and inputting the extracted feature vector to the parameter setting obtained in step S1 training.
  • the category information (tampering class or real class) of the input image is output.
  • the training step S1 process is as follows:
  • Step S11 Perform color space conversion on the color image in the training set, and convert it from the RGB color space to the YCbCr color space through a transformation matrix, where Y refers to a luminance component, Cb refers to a blue chrominance component, and Cr refers to a red chrominance component. Extract its chrominance component (Cb or Cr).
  • T a certain threshold
  • Step S13 Calculating the steady-state distribution of the Markov chain of the edge image obtained in S12, and the steady-state distribution refers to the distribution when the Markov chain reaches a stationary state, which can be used to describe the Markov chain.
  • the probability distribution in this stationary state is taken as the image features ⁇ p b P2, P3, P4, P5, P6, P7, p 8 , p 9 ⁇ , which are 9-dimensional.
  • Step S14 input the feature marked with the category information into the SVM classifier for training, and obtain a classifier model.
  • the sorting step S2 is as follows: Step S21: Perform color space conversion on the currently input color image, convert it from RGB color space to YCbCr color space, and extract its chrominance component (Cb or Cr).
  • Step S22 Calculate the edge image of the chrominance component obtained in step S21, and cut off the gradation value greater than ⁇ to obtain a new edge image.
  • Step S23 Calculate the steady-state distribution of the Markov chain of the edge image obtained in S22 and take it as a feature.
  • Step S24 The classifier model obtained in step S14 is loaded, and the feature input in step S23 is input to the SVM classifier to classify, and the current image is determined as a result of the tampering image, and the tamper image detection is completed.
  • the color space conversion refers to converting a color image from an RGB color space to a YCbCr color space, as in equation (1).
  • the edge image refers to an image obtained by convoluting the chrominance component of the color image with the template M, and then taking the absolute value thereof, as in the equation (2).
  • the SVM classifier of the present invention uses a radial basis function as a kernel function to find an optimal classifier model parameter by traversing the search.
  • feature extraction is first performed on an image in a training library, and then a classifier training is performed to obtain a classifier model; the trained classifier model is used for image blind tamper detection,
  • the result of the binarization test a real image or a tampering image.
  • the image file to be processed is first determined to be a color image format and then entered into the flow. Secondly, the color image is color-space converted (RGB to YCbCr) and its Cr component (or Cb component) is taken and its edge image is calculated and truncated. Then, based on the present invention, the transition probability matrix of the edge image is calculated and The steady-state distribution will be used as a feature. Finally, the feature is input into a pre-trained classifier for detection, giving a result of whether the image is a tamper image.
  • RGB to YCbCr color-space converted
  • Cr component or Cb component
  • the obtained feature vector is input into the classifier model of the trained parameter, and the classifier outputs the category information of the picture as the tamper image.
  • the classifier model file is obtained by performing feature extraction on image samples of known tags (real images or tampering images) and then performing classifier training.
  • the SVM is a classifier that seeks to maximize the separation of samples of different tags in the feature space by looking for a classification interface. Parameter adjustment mainly prevents over-learning (ie, the detection rate is high on the training samples and the detection rate is extremely low on the test sample library).

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Description

基于图像色度的彩色数字图像篡改检测方法 技术领域
本发明涉及模式识别领域, 特别涉及彩色数字图像的篡改检测方 法。 背景技术
随着近年来计算机和多媒体技术的飞速发展, 特别是大量图像处理 软件 (例如 Photoshop) 的出现, 使得数字图像的篡改、 伪造己不再是 一件难事。 大量篡改的图像出现在新闻报道, 摄影比赛, 甚至是法律证 据中。 这使得社会对图像篡改检测技术的需求十分迫切。 数字图像篡改 检测(Digital Image Tampering Detection)的目的是通过分析图像数据进 而发现篡改行为的存在, 甚至能够定位到篡改区域。 通过图像篡改检测 可以发现不实的新闻图片报道、伪造的图像证据等,对于网络信息安全、 公共安全, 司法取证等等具有重大意义。
图像处理软件的蓬勃发展使图像篡改操作越来越容易, 数字图像处 理的各种技术交替使用使得图像篡改以假乱真的手段越来越高明, 但同 时也使篡改图像越来越难被人主观检测出来, 这就需要通过计算机算法 辅助进行篡改图像鉴别。 目前数字图像篡改检测算法分为两大类: 主动 方式和被动方式。 主动方式需要在真实图像生成过程中 (拍照时) 或发 布之前, 在其中嵌入水印信息。 在验证图像真实性时, 只需从图像中提 取水印信息, 如果水印信息被破坏, 则说明图像有可能被篡改。 然而这 种方式在实际应用中不是很方便, 我们不能让所有的相机都添加水印嵌 入模块。 这使得被动取证方式得到很大的关注。 被动方式是直接从图像 本身收集证据, 认为篡改操作破坏了图像的一致性 [A.C. Popescu and H. Farid, " Exposing digital forgeries in color filter array interpolated images, " IEEE Transactions on Signal Processing, vol. 53, no. 10, pp. 3948-3959, 2005.],或者使图像中的相机固有噪声改变 或消失 [ J. Lukas, J, Fridrich, M. Goljan, "Detecting digital image forgeries using sensor pattern noise, " Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series., vol. 6072, pp. 362-372, 2006], 或者使图像的高阶统计量发生改变 [Y.Q. Shi, W. Chen, and C. Chen,
I " A natural image model approach to splicing detection, " Proceedings of the 9ih workshop on Multimedia & security, pp.51-62, 2007] o 发明内容
本发明的目的是提供一种基于图像色度的彩色数字图像篡改检测 方法, 能够实现准确高效的彩色图像篡改自动检测。
为实现上述目的, 一种基于图像色度的彩色数字图像篡改检测方 法, 包括训练过程和分类过程, 所述训练过程包括:
对训练库中已标记类别信息的彩色图像从 RGB 色彩空间转换到 YCbCr色彩空间, 并提取其色度分量;
计算色度分量的边缘图像, 并对大于阈值的灰度值进行截断, 得到 新的边缘图像;
对新的边缘图像计算马尔可夫链的稳态分布;
将标记好类别信息的特征输入到 SVM分类器中进行训练, 得到分 类器模型;
所述分类过程包括:
对任意输入的彩色图像从 RGB色彩空间转换到 YCbCr色彩空间, 并 提取其色度分量;
计算色度分量的边缘图像, 并对大于阈值的灰度值进行截断, 得到 新的边缘图像;
对新的边缘图像计算马尔可夫链的稳态分布作为特征;
将所述特征载入训练过程中的分类器模型进行分类, 得到判决结 果。
本发明的方法可以用于图像等多媒体数据的篡改检测。可运用于互 联网多媒体、 新闻报道图片、 司法图片证据等的内容取证、 鉴别等相应 产品。 由于本发明不需要事先向图像中嵌入水印信息, 只需分析图像本 身而且其检测特征提取简单、 快速、 高效, 故可在大规模数据通信、 多 媒体传输等内容取证环境下能够得到有效的应用。 附图说明 图 1 本发明基于图像色度的彩色图像篡改检测方法的流程图 图 2 本发明实施例中使用的待检测图像; 该图像为篡改图像。 图 中斑马来自其他图片。图中(a)为待测图像,该图像为篡改图像。 (b)、 (c) 、 ( d) 分别为将该彩色图像转换到 YCbCr色彩空间后, Y、 Cb、 Cr分量的边缘图像。 具体实施方式
如图 1所示, 本发明所采用的技术方案包括以下步骤:
训练步骤 S1 :首先,对训练库中已标记类别信息(篡改类或真实类) 的彩色图像进行色彩空间转换, 取其色度分量, 然后计算其边缘图像的 马尔可夫稳态分布, 作为特征向量。 其次, 对提取的特征集合进行分类 器训练, 得到分类器的模型参数。
分类步骤 S2:对任意输入的彩色图像进行色彩空间转换,取其色度 分量, 计算其边缘图像的马尔可夫稳态分布作为特征, 将提取的特征向 量输入到步骤 S1训练得到的参数设定的分类器模型中, 输出输入图像 的类别信息 (篡改类或真实类) 。
所述训练步骤 S1过程如下:
步骤 S11 : 对训练集中彩色图像进行色彩空间转化, 通过变换矩阵 将其从 RGB色彩空间转换到 YCbCr色彩空间, 其中 Y是指亮度分量, Cb指蓝色色度分量, Cr指红色色度分量。提取其色度分量 (Cb或 Cr)。
步骤 S12: 对步骤 S11中得到的色度分量, 计算其边缘图像, 并对 其大于特定阈值 T (如 T=8) 的灰度值进行截断 (大于 8的数都用 8代 替) , 得到新的边缘图像。
步骤 S13:对 S12中得到的边缘图像计算其马尔可夫链的稳态分布, 稳态分布是指马尔可夫链达到平稳状态时的分布, 可用其描述马尔可夫 链。 将这一平稳状态时的概率分布作为图像特征 {pb P2, P3, P4, P5, P6, P7, p8, p9}, 共 9维。
步骤 S14: 将标记好类别信息的特征输入到 SVM分类器中进行训 练, 得到分类器模型。
所述分类步骤 S2如下: 步骤 S21: 对当前输入的彩色图像进行色彩空间转化, 将其从 RGB 色彩空间转换到 YCbCr色彩空间, 提取其色度分量 (Cb或 Cr) 。
步骤 S22: 对步骤 S21中得到的色度分量, 计算其边缘图像, 并对 其大于 Τ的灰度值进行截断, 得到新的边缘图像。
步骤 S23: 对 S22中得到的边缘图像计算其马尔可夫链的稳态分布 并将其作为特征。
步骤 S24: 载入步骤 S14中所得到的分类器模型, 将步骤 S23中得 到的特征输入 SVM分类器进行分类, 得到当前图像是否为篡改图像的 判定结果, 完成篡改图像检测。
所述的色彩空间转换是指将彩色图像从 RGB 色彩空间转换到 YCbCr色彩空间, 如式 (1) 。
(1)
Figure imgf000006_0002
边缘图像是指对彩色图像的色度分量用模板 M进行卷积操作, 然 后取其绝对值所得到的图像, 如式 (2) 。
E = \M®I\ (2) 其中, M为一 3X3的矩阵, 可取:
Figure imgf000006_0001
计算得到边缘图像后, 利用公式 (3) 对其进行截断处理 (通常 T=8) e(i,j)<T
(3)
T e(i,j)≥T 所述稳态分布 ( τ) 表示为:
π = πΡ
其中 是由公式 (2) 得到的边缘图像的转移概率矩阵, 其计算公式为: 式中^ H为边缘图像的长和宽, 为冲击函数, 等号成立时取 1, 不成 立时为 0。
本发明的 SVM分类器采用径向基函数作为核函数, 通过遍历搜索 的方式找到最优的分类器模型参数。
根据本发明的实施例, 在彩色图像篡改检测中, 首先对训练库中的 图像进行特征提取, 然后进行分类器训练得到分类器模型; 将训练好的 分类器模型用于图像盲篡改检测, 给出二值化的检测结果: 真实图像或 篡改图像。
如图 1所示, 首先对待处理的图像文件, 确定为彩色图像格式后进 入本流程。 其次对彩色图像进行色彩空间转换 (RGB转换到 YCbCr) 并取其 Cr分量(或 Cb分量)并计算其边缘图像并将其截断, 然后, 基 于本发明, 计算边缘图像的转移概率矩阵并将其稳态分布将作为特征, 最后, 将特征输入到预先训练好分类器中去进行检测, 给出图像是否为 篡改图像的结果。
实施例
以图 2中的图像为例进行说明。
首先,对该图像进行色彩空间转换(RGB空间转换到 YCbCr空间)。 然后, 计算 Cr的边缘图像 (图 2d) 并将其用 T=8截断。 基于本发 明, 求该边缘图像的转移概率矩阵尸。
其次,计算稳态分布 Γ ,得到一个 9维的特征向量 τ={ρΐ 5 p2, p.,, ρ4, Ρ5,
Ρ6, Ρ7, Ρ8, Ρ9}。
最后, 将得到的特征向量输入到已训练好参数的分类器模型中, 得 到分类器输出该图片的类别信息为篡改图像。
分类器模型文件是通过对已知标签(真实图像或篡改图像) 的图像 样本, 进行特征提取, 然后进行分类器训练得到的。 SVM 是一种分类 器, 它主要是通过寻找一个分类界面, 最大程度的将不同标签的样本在 特征空间中分开。 参数调节主要防止过学习情况的发生(即在训练样本 上检测率很高, 而在测试样本库上检测率极低) 。
以上所述, 仅为本发明中的具体实施方式, 但本发明的保护范围并 不局限于此。任何熟悉该技术的本领域技术人员在本发明所揭露的技术 范围内, 可以进行变换或替换。 所进行的变换和替换都应涵盖在本发明 的包含范围之内。 因此, 本发明的保护范围应该以权利要求书的保护范 围为准。

Claims

权 利 要 求
1. 一种基于图像色度的彩色数字图像篡改检测方法,包括训练过程 和分类过程, 所述训练过程包括:
δ 对训练库中已标记类别信息的彩色图像从 RGB 色彩空间转换到
YCbCr色彩空间, 并提取其色度分量;
计算色度分量的边缘图像, 并对大于阈值的灰度值进行截断, 得到 新的边缘图像;
对新的边缘图像计算马尔可夫链的稳态分布;
0 将标记好类别信息的特征输入到 SVM分类器中进行训练, 得到分 类器模型;
所述分类过程包括- 对任意输入的彩色图像从 RGB色彩空间转换到 YCbCr色彩空间, 并 提取其色度分量;
5 计算色度分量的边缘图像, 并对大于阈值的灰度值进行截断, 得到 新的边缘图像;
对新的边缘图像计算马尔可夫链的稳态分布作为特征;
将所述特征载入训练过程中的分类器模型进行分类, 得到判决结 果。
0 2. 根据权利要求 1所述的方法,其特征在于所述已标记类别信息包 括篡改信息或真实信息。
3.根据权利要求 1 所述的方法, 其特征在于按下式将彩色图像从 RGB色彩空间转换到 YCbCr色彩空间:
Figure imgf000009_0001
5 4.根据权利要求 1所述的方法, 其特征在于所述色度分量包括蓝色 色度分量和红色色度分量。
5.根据权利要求 1所述的方法, 其特征在于按下式对阈值的灰度值 进行截断处理-
Figure imgf000010_0001
6.根据权利要求 1所述的方法, 其特征在于所述阈值为 8.
7.根据权利要求 6所述的方法,其特征在于大于 8的数都用 8代替。
8.根据权利要求 1所述的方法, 其特征在于所述分类器模型采用 向基函数作为核函数, 以便找到最优的分类器模型参数。
PCT/CN2011/073444 2011-04-28 2011-04-28 基于图像色度的彩色数字图像篡改检测方法 WO2012145909A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2011/073444 WO2012145909A1 (zh) 2011-04-28 2011-04-28 基于图像色度的彩色数字图像篡改检测方法
CN2011800044127A CN102959588A (zh) 2011-04-28 2011-04-28 基于图像色度的彩色数字图像篡改检测方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2011/073444 WO2012145909A1 (zh) 2011-04-28 2011-04-28 基于图像色度的彩色数字图像篡改检测方法

Publications (1)

Publication Number Publication Date
WO2012145909A1 true WO2012145909A1 (zh) 2012-11-01

Family

ID=47071563

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/073444 WO2012145909A1 (zh) 2011-04-28 2011-04-28 基于图像色度的彩色数字图像篡改检测方法

Country Status (2)

Country Link
CN (1) CN102959588A (zh)
WO (1) WO2012145909A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103679626A (zh) * 2013-11-28 2014-03-26 中国科学院自动化研究所 一种基于dct系数的图像篡改定位方法
CN106355180A (zh) * 2016-09-07 2017-01-25 武汉安可威视科技有限公司 一种基于颜色与边缘特征相结合的车牌定位方法
CN108108772A (zh) * 2018-01-06 2018-06-01 天津大学 一种基于配电线路航拍图像的绝缘子污闪状态检测方法
CN108230227A (zh) * 2018-02-06 2018-06-29 中国银行股份有限公司 一种图像篡改的识别方法、装置及电子设备
CN108470141A (zh) * 2018-01-27 2018-08-31 天津大学 一种基于统计特征与机器学习的配电线路中绝缘子识别方法
CN109416747A (zh) * 2016-09-30 2019-03-01 富士通株式会社 目标颜色识别方法、装置以及计算机系统

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105654089A (zh) * 2014-08-20 2016-06-08 江南大学 基于Markov过程和Gabor滤波的图像重采样检测
CN105678308A (zh) * 2016-01-12 2016-06-15 中国科学院自动化研究所 一种基于光照方向不一致性的图像拼接检测方法
CN110363207A (zh) * 2018-04-09 2019-10-22 阿里巴巴集团控股有限公司 图像检测和分类器模型的生成方法、装置及设备
CN109558872B (zh) * 2018-11-22 2022-02-11 四川大学 一种车辆颜色识别方法
CN111784708B (zh) * 2020-07-03 2021-03-12 上海骏聿数码科技有限公司 图像防篡改检验方法和装置

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002158865A (ja) * 2000-11-22 2002-05-31 Seiko Epson Corp 画像情報処理方法及び画像情報処理装置
US20080037823A1 (en) * 2006-06-30 2008-02-14 New Jersey Institute Of Technology Method and apparatus for image splicing/tampering detection using moments of wavelet characteristic functions and statistics of 2-d phase congruency arrays
US20080193031A1 (en) * 2007-02-09 2008-08-14 New Jersey Institute Of Technology Method and apparatus for a natural image model based approach to image/splicing/tampering detection
CN101366061A (zh) * 2005-12-16 2009-02-11 Ncr公司 检测改进质量的伪造介质
CN101464952A (zh) * 2007-12-19 2009-06-24 中国科学院自动化研究所 基于轮廓的异常行为识别方法
CN101493939A (zh) * 2009-02-27 2009-07-29 西北工业大学 基于小波域同态滤波的检测伪造图像的方法
CN101493927A (zh) * 2009-02-27 2009-07-29 西北工业大学 基于边缘方向特征的图像可信度检测方法
CN101859385A (zh) * 2010-06-29 2010-10-13 上海大学 基于图像的局部模糊篡改盲检测方法
US20110002504A1 (en) * 2006-05-05 2011-01-06 New Jersey Institute Of Technology System and/or method for image tamper detection
US20110064315A1 (en) * 2009-09-15 2011-03-17 Texas Instruments Incorporated Method and apparatus for image capturing tampering detection

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101777129B (zh) * 2009-11-25 2012-05-23 中国科学院自动化研究所 一种基于特征检测的图像匹配方法

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002158865A (ja) * 2000-11-22 2002-05-31 Seiko Epson Corp 画像情報処理方法及び画像情報処理装置
CN101366061A (zh) * 2005-12-16 2009-02-11 Ncr公司 检测改进质量的伪造介质
US20110002504A1 (en) * 2006-05-05 2011-01-06 New Jersey Institute Of Technology System and/or method for image tamper detection
US20080037823A1 (en) * 2006-06-30 2008-02-14 New Jersey Institute Of Technology Method and apparatus for image splicing/tampering detection using moments of wavelet characteristic functions and statistics of 2-d phase congruency arrays
US20080193031A1 (en) * 2007-02-09 2008-08-14 New Jersey Institute Of Technology Method and apparatus for a natural image model based approach to image/splicing/tampering detection
CN101464952A (zh) * 2007-12-19 2009-06-24 中国科学院自动化研究所 基于轮廓的异常行为识别方法
CN101493939A (zh) * 2009-02-27 2009-07-29 西北工业大学 基于小波域同态滤波的检测伪造图像的方法
CN101493927A (zh) * 2009-02-27 2009-07-29 西北工业大学 基于边缘方向特征的图像可信度检测方法
US20110064315A1 (en) * 2009-09-15 2011-03-17 Texas Instruments Incorporated Method and apparatus for image capturing tampering detection
CN101859385A (zh) * 2010-06-29 2010-10-13 上海大学 基于图像的局部模糊篡改盲检测方法

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103679626A (zh) * 2013-11-28 2014-03-26 中国科学院自动化研究所 一种基于dct系数的图像篡改定位方法
CN106355180A (zh) * 2016-09-07 2017-01-25 武汉安可威视科技有限公司 一种基于颜色与边缘特征相结合的车牌定位方法
CN109416747A (zh) * 2016-09-30 2019-03-01 富士通株式会社 目标颜色识别方法、装置以及计算机系统
CN109416747B (zh) * 2016-09-30 2022-01-14 富士通株式会社 目标颜色识别方法、装置以及计算机系统
CN108108772A (zh) * 2018-01-06 2018-06-01 天津大学 一种基于配电线路航拍图像的绝缘子污闪状态检测方法
CN108108772B (zh) * 2018-01-06 2021-08-10 天津大学 一种基于配电线路航拍图像的绝缘子污闪状态检测方法
CN108470141A (zh) * 2018-01-27 2018-08-31 天津大学 一种基于统计特征与机器学习的配电线路中绝缘子识别方法
CN108470141B (zh) * 2018-01-27 2021-08-10 天津大学 一种基于统计特征与机器学习的配电线路中绝缘子识别方法
CN108230227A (zh) * 2018-02-06 2018-06-29 中国银行股份有限公司 一种图像篡改的识别方法、装置及电子设备

Also Published As

Publication number Publication date
CN102959588A (zh) 2013-03-06

Similar Documents

Publication Publication Date Title
WO2012145909A1 (zh) 基于图像色度的彩色数字图像篡改检测方法
US11263500B2 (en) Image detection methods and apparatus
Marra et al. On the vulnerability of deep learning to adversarial attacks for camera model identification
Wang et al. Detection and localization of image forgeries using improved mask regional convolutional neural network
Fridrich et al. Detection of copy-move forgery in digital images
CN103914839B (zh) 一种基于隐写分析的图像拼接篡改检测方法及装置
CN106530200B (zh) 一种基于深度学习模型的隐写图像检测方法及系统
Mahmood et al. A survey on block based copy move image forgery detection techniques
Hussain et al. Image forgery detection using multi-resolution Weber local descriptors
Wang et al. LBP-SVD based copy move forgery detection algorithm
CN110457996B (zh) 基于vgg-11卷积神经网络的视频运动对象篡改取证方法
Hussain et al. Copy-move image forgery detection using multi-resolution weber descriptors
Dixit et al. Copy-move forgery detection exploiting statistical image features
Kang et al. Color Image Steganalysis Based on Residuals of Channel Differences.
Hakimi et al. Image-splicing forgery detection based on improved lbp and k-nearest neighbors algorithm
CN1527585A (zh) 一种数字信息安全传输和应用方法
Li et al. Distinguishing computer graphics from photographic images using a multiresolution approach based on local binary patterns
Parveen et al. Classification and evaluation of digital forensic tools
CN111027573A (zh) 基于盲取证技术的图像真伪鉴别方法
CN104077738A (zh) 基于局部直方图特征的彩色图像水印方法
Ahmed et al. Image splicing detection and localisation using efficientnet and modified u-net architecture
CN106845540B (zh) 一种图像重采样操作插值类型识别方法
Kumar et al. Towards recent developments in the field of digital image forgery detection
Rezaei et al. Stegrt1: a dataset for evaluating steganalysis systems in real-world scenarios
Chen et al. Identification of image global processing operator chain based on feature decoupling

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201180004412.7

Country of ref document: CN

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

Ref document number: 11864310

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11864310

Country of ref document: EP

Kind code of ref document: A1