公开(公告)号：US09336556B2

公开(公告)日：2016-05-10

申请号：US14150792

申请日：2014-01-09

Applicant: CANON KABUSHIKI KAISHA

Inventor： Ayuta Kawazu

IPC: G06K9/00 ,  G06T1/00 ,  G06T3/40

CPC classification number: G06T1/0028 ,  G06T3/4015 ,  G06T2201/0201

Abstract: Pattern information of a color filter array and algorithm information of color interpolation processing used in photography of input image data are obtained. Comparative image data is generated from the input image data using the pattern information and algorithm information. The comparative image data and input image data are compared to detect alteration of the input image data.

Abstract translation: 获得滤色器阵列的图案信息和用于输入图像数据的拍摄中使用的颜色插值处理的算法信息。 使用图案信息和算法信息从输入图像数据生成比较图像数据。 将比较图像数据和输入图像数据进行比较，以检测输入图像数据的变化。

公开(公告)号：US20130223731A1

公开(公告)日：2013-08-29

申请号：US13755832

申请日：2013-01-31

Applicant: CANON KABUSHIKI KAISHA

Inventor： Ayuta Kawazu

IPC: G06K9/46

CPC classification number: G06K9/46 ,  G06K2009/00583 ,  G06T1/0028 ,  G06T2201/0201

Abstract: A CFA pattern is extracted from captured image data for each first unit region. A first altered region is detected from disturbance of the periodicity of the CFA pattern, and the first altered region is an image region in which copying has been performed from image data different from the captured image data to the captured image data. The feature amount of the captured image data is extracted for each second unit region different in size from the first unit region. The feature amounts are compared for each second unit region to detect a second altered region, and the second altered region is an image region in which copying has been performed from the captured image data to the captured image data. Information concerning the first and second altered regions are output as alteration detection results in the captured image data.

Abstract translation: 从每个第一单位区域的捕获图像数据中提取CFA图案。 从CFA图案的周期性的干扰中检测第一改变区域，并且第一改变区域是其中已经从与捕获的图像数据不同的图像数据执行复制到图像数据的图像区域。 对于与第一单位区域不同的大小的第二单位区域，提取拍摄图像数据的特征量。 对于每个第二单位区域比较特征量以检测第二改变区域，并且第二改变区域是其中已经从捕获的图像数据执行复制到捕获的图像数据的图像区域。 关于第一和第二改变区域的信息作为捕获图像数据中的改变检测结果被输出。

公开(公告)号：US20130039588A1

公开(公告)日：2013-02-14

申请号：US13559985

申请日：2012-07-27

Applicant: Ji LI ,  Xiaowei Yang

Inventor： Ji LI ,  Xiaowei Yang

IPC: G06K9/46

CPC classification number: G06T1/0028 ,  G06T2201/0081 ,  G06T2201/0201

Abstract: An image processing method and apparatus for tamper proofing are proposed. The method includes: acquiring a first robust feature representation of a first set of robust feature points for an original image and a second robust feature representation of a second set of robust feature points for an image to be detected, respectively; matching the first robust feature representation with the second robust feature representation so as to acquire mismatching feature points; and determining whether the image to be detected has been tampered with relative to the original image based on a distribution characteristic of the mismatching feature points. With the embodiments of the invention, the distribution characteristic of the mismatching feature points is analyzed based on the robust feature representations of the original image and the image to be detected, so that conventional operations on the image can be distinguished effectively from tampering in the image with sufficient robustness.

Abstract translation: 提出了一种用于防篡改的图像处理方法和装置。 该方法包括：分别获取原始图像的第一组鲁棒特征点的第一鲁棒特征表示和待检测图像的第二组鲁棒特征点的第二鲁棒特征表示; 将第一鲁棒特征表示与第二鲁棒特征表示相匹配，以获得不匹配的特征点; 以及基于所述不匹配特征点的分布特性来确定所述待检测图像是否相对于所述原始图像被篡改。 利用本发明的实施例，基于原始图像和要检测的图像的鲁棒特征分析不匹配特征点的分布特性，使得可以有效地区分图像上的常规操作以防止图像中的篡改 具有足够的鲁棒性。

公开(公告)号：US08340347B2

公开(公告)日：2012-12-25

申请号：US12624293

申请日：2009-11-23

Applicant: Junichi Hayashi ,  Yoshiharu Imamoto ,  Nobuhiro Tagashira ,  Kazuya Kishi ,  Yasuhiro Nakamoto ,  Kazuomi Oishi

Inventor： Junichi Hayashi ,  Yoshiharu Imamoto ,  Nobuhiro Tagashira ,  Kazuya Kishi ,  Yasuhiro Nakamoto ,  Kazuomi Oishi

IPC: G06K9/00

CPC classification number: G06T1/0028 ,  G06T1/0064 ,  G06T2201/0201

Abstract: This invention provides a technique of preventing determination of image alteration when digital image data has undergone, e.g., rotation without any substantial change in contents. To do this, an area separation processing unit separates image data into areas. For each of the separated areas, an area feature value calculator calculates an area feature value independent of the coordinate information of the image. An area order sorter sorts the separated areas in accordance with the calculated area feature values. A validation data generation processing unit generates validation data based on the sort result.

Abstract translation: 本发明提供了当数字图像数据经历例如旋转而没有任何实质内容改变时防止图像变化的确定的技术。 为此，区域分离处理单元将图像数据分离成区域。 对于每个分离区域，区域特征值计算器计算与图像的坐标信息无关的区域特征值。 区域排序排序器根据计算的区域特征值对分离的区域进行排序。 验证数据生成处理单元基于分类结果生成验证数据。

公开(公告)号：US20120148089A1

公开(公告)日：2012-06-14

申请号：US13027675

申请日：2011-02-15

Applicant: Aditya Vashistha ,  Rajarathnam Nallusamy ,  Sanjoy Paul

Inventor： Aditya Vashistha ,  Rajarathnam Nallusamy ,  Sanjoy Paul

IPC: G06K9/00

CPC classification number: G09C5/00 ,  G06T1/005 ,  G06T1/0085 ,  G06T2201/0201 ,  H04L2209/608

Abstract: A method for watermarking of video content is provided. An averaged scene image is computed for each scene of video content by performing averaging of frames present in each scene of video content. For each averaged scene image a set of random numbers are generated using a secret key to identify pixels at random locations of the averaged scene image. The secret key is associated with a watermark pattern corresponding to each averaged scene image. The identified pixels in each averaged scene image are mapped to each pixel of corresponding watermark pattern to obtain respective mapped pixels. Using respective mapped pixels, values of verification information are fetched and assigned using predetermined rules. The values of verification information are arranged to obtain first visual cryptographic share of watermark pattern for each averaged scene image.

Abstract translation: 提供了一种用于水印视频内容的方法。 通过对视频内容的每个场景中存在的帧进行平均，对视频内容的每个场景计算平均场景图像。 对于每个平均场景图像，使用秘密密钥来生成一组随机数，以识别平均场景图像的随机位置处的像素。 秘密密钥与对应于每个平均场景图像的水印图案相关联。 将每个平均场景图像中的识别像素映射到相应水印图案的每个像素，以获得相应的映射像素。 使用相应的映射像素，使用预定规则来获取和分配验证信息的值。 布置验证信息的值以获得每个平均场景图像的水印图案的第一可视加密份额。

公开(公告)号：US20120076422A1

公开(公告)日：2012-03-29

申请号：US12892148

申请日：2010-09-28

Applicant: Lei Yang ,  Jun Tian ,  Dapeng Oliver Wu ,  Hong Heather Yu

Inventor： Lei Yang ,  Jun Tian ,  Dapeng Oliver Wu ,  Hong Heather Yu

IPC: G06K9/46

CPC classification number: G06K9/6211 ,  G06F17/30256 ,  G06T1/0028 ,  G06T2201/0052 ,  G06T2201/0201 ,  H04N21/44008 ,  H04N21/4627 ,  H04N21/8352 ,  H04N21/8355

Abstract: In accordance with an embodiment, a method of authenticating images includes electronically receiving an anchor image and a query image, performing a feature point extraction of an anchor image, and performing a feature point extraction of a query image. The method also includes clustering feature points of the anchor image and feature points of the query image, where clustering includes determining matching feature points, determining outlier feature points, and excluding outlier feature points. Whether the anchor image is similar to the query image is determined based on a distance between the feature points of the anchor image and the feature points of the query image. If the anchor image is similar to the query image, possible tampered areas of the query image based on the outlier feature points are identified.

Abstract translation: 根据实施例，认证图像的方法包括电子地接收锚图像和查询图像，执行锚图像的特征点提取，以及执行查询图像的特征点提取。 该方法还包括锚图像的聚类特征点和查询图像的特征点，其中聚类包括确定匹配特征点，确定离群点特征点，并排除异常值特征点。 基于锚图像的特征点与查询图像的特征点之间的距离来确定锚图像是否类似于查询图像。 如果锚图像与查询图像相似，则识别基于异常特征点的查询图像的可能的篡改区域。

公开(公告)号：US07840007B2

公开(公告)日：2010-11-23

申请号：US10497236

申请日：2002-12-03

Applicant: Naoto Takeda ,  Masamichi Taniguchi ,  Kunio Taniguchi ,  Kiyotaka Yamamoto

Inventor： Naoto Takeda ,  Masamichi Taniguchi ,  Kunio Taniguchi ,  Kiyotaka Yamamoto

IPC: H04N7/167

CPC classification number: G06T1/0021 ,  G06T2201/0201

Abstract: A digital data false alteration detection program causes a computer to execute (a) a step (S1) of dividing digital data into a plurality of smaller block data, (b) a step (S2) of extracting noise inherent to a digital data acquisition device for each of the small block data, (c) a step (S3) of calculating correlation of the noise between adjacent small block data, and (d) a step (S4) of detecting small block data having noise correlation lower than a level predetermined for the surrounding small block data, as falsely altered data.

Abstract translation: 数字数据虚假变更检测程序使计算机执行（a）将数字数据分割为多个较小块数据的步骤（S1），（b）提取数字数据采集装置固有的噪声的步骤（S2） 对于每个小块数据，（c）计算相邻小块数据之间的噪声的相关性的步骤（S3），以及（d）检测具有低于预定级别的噪声相关性的小块数据的步骤（S4） 对于周围的小块数据，作为错误更改的数据。

公开(公告)号：US20100091981A1

公开(公告)日：2010-04-15

申请号：US12422677

申请日：2009-04-13

Applicant: Yun-Qing Shi ,  Bin Li

Inventor： Yun-Qing Shi ,  Bin Li

IPC: H04K1/00 ,  G06K9/00

CPC classification number: G06T1/0028 ,  G06T2201/0051 ,  G06T2201/0052 ,  G06T2201/0061 ,  G06T2201/0201

Abstract: Techniques described herein are generally related to steganalysis of suspect media. Steganalysis techniques may include receiving instances of suspect media as input for steganalytic processing. A first set of quantized blocks of data elements may be identified within the media, with this first set of blocks being eligible to be embedded with steganographic data. A second set of quantized blocks of data elements may be identified within the media, with this second set of blocks being ineligible to be embedded with steganographic data. The steganalysis techniques may requantize the first and second blocks. In turn, these techniques may compare statistics resulting from requantizing the first block with statistics resulting from requantizing the second block. The steganalysis techniques may then assess whether the first block of data elements is embedded with steganographic features based on how the statistics of the second blocks compare with the statistics of the first blocks.

Abstract translation: 本文描述的技术通常涉及可疑介质的隐写分析。 隐匿分析技术可以包括接收可疑媒体的实例作为隐写处理的输入。 可以在媒体内识别数据元素的第一组量化块，其中该第一组块有资格嵌入隐写数据。 可以在媒体内识别第二组数据元素的量化块，其中第二组块不符合隐写数据​​的嵌入。 隐写分析技术可以重新调整第一和第二块。 反过来，这些技术可以将从重新量化第一个块得到的统计数据与由再量化第二个块产生的统计数据进行比较。 隐写分析技术可以基于第二块的统计数据如何与第一块的统计量进行比较来评估第一块数据元素是否嵌入隐写特征。

公开(公告)号：US06628824B1

公开(公告)日：2003-09-30

申请号：US09271112

申请日：1999-03-17

Applicant: Ken Belanger

Inventor： Ken Belanger

IPC: G06K900

CPC classification number: G06K9/6212 ,  G06F17/3025 ,  G06K9/4652 ,  G06T7/90 ,  G06T2201/0201 ,  G06T2207/10024

Abstract: A method and apparatus are provided for analyzing, identifying, and comparing images. The method can be used with any visually-displayed medium that is represented in any type of color space. An identified image can be authenticated, registered, marked, compared to another image, or recognized using the method and apparatus according to the present invention. At least one characteristic of an image's color space is selected and determined to generate a unique description of the image. This identification information is then used to compare different identified images to determine if they are identical according to a set of predetermined criteria. The predetermined criteria can be adjusted to permit the identification of images that are identical in part. In the preferred embodiment of the present invention, a software search application, such as a search engine or a spider, is used to locate and retrieve an image to be identified from an electronic network. A notification alarm is triggered when a duplicate image is located. In one embodiment, the present invention is implemented using a computer. One or more software applications, software modules, firmware, and hardware, or any combination thereof, are used to determine the identification information for the selected image characteristics, search for images, provide notification of identical images, and to generate a database of identified images.

Abstract translation: 提供了一种用于分析，识别和比较图像的方法和装置。 该方法可以与在任何类型的颜色空间中表示的任何视觉显示介质一起使用。 识别的图像可以被认证，注册，标记，与另一图像相比较，或者使用根据本发明的方法和装置识别。 选择和确定图像的颜色空间的至少一个特征以产生图像的唯一描述。 然后，该识别信息用于比较不同的识别图像，以根据一组预定标准确定它们是否相同。 可以调整预定标准以允许识别部分相同的图像。 在本发明的优选实施例中，使用诸如搜索引擎或蜘蛛的软件搜索应用来定位和检索要从电子网络识别的图像。 当找到重复的图像时，触发通知报警。 在一个实施例中，使用计算机来实现本发明。 使用一个或多个软件应用程序，软件模块，固件和硬件，或其任何组合来确定所选图像特征的识别信息，搜索图像，提供相同图像的通知，以及生成识别图像的数据库 。