公开(公告)号：US07961908B2

公开(公告)日：2011-06-14

申请号：US12023877

申请日：2008-01-31

申请人： Meir Tzur ,  Ben-Zion Shaick ,  Itsik Dvir ,  Victor Pinto

发明人： Meir Tzur ,  Ben-Zion Shaick ,  Itsik Dvir ,  Victor Pinto

IPC分类号： G06K9/00 ,  G06K9/46 ,  G06K9/36

CPC分类号： G06K9/60 ,  G06K9/3208 ,  G06K9/62 ,  G06K9/6282 ,  G06T3/60 ,  H04N5/23219 ,  H04N5/23229 ,  H04N2101/00

摘要： The likelihood of a particular type of object, such as a human face, being present within a digital image, and its location in that image, are determined by comparing the image data within defined windows across the image in sequence with two or more sets of data representing features of the particular type of object. The evaluation of each set of features after the first is preferably performed only on data of those windows that pass the evaluation with respect to the first set of features, thereby quickly narrowing potential target windows that contain at least some portion of the object. Correlation scores are preferably calculated by the use of non-linear interpolation techniques in order to obtain a more refined score. Evaluation of the individual windows also preferably includes maintaining separate feature set data for various positions of the object around one axis and rotating the feature set data with respect to the image data for the individual windows about another axis.

摘要翻译： 通过将图像中定义的窗口内的图像数据与两个或更多个组合的顺序进行比较来确定特定类型的对象（例如人脸）存在于数字图像内的可能性及其在该图像中的位置 表示特定类型对象的特征的数据。 优选地，对第一组之后的每组特征的评估仅对通过相对于第一组特征的评估的那些窗口的数据执行，从而快速缩小包含对象的至少一部分的潜在目标窗口。 相关分数优选通过使用非线性内插技术来计算，以获得更精细的分数。 各个窗口的评估还优选地包括为围绕一个轴的对象的各种位置维护单独的特征集数据，并且相对于关于另一个轴的各个窗口的图像数据旋转特征集数据。

公开(公告)号：US20060017814A1

公开(公告)日：2006-01-26

申请号：US10897186

申请日：2004-07-21

申请人： Victor Pinto ,  Itsik Dvir

发明人： Victor Pinto ,  Itsik Dvir

IPC分类号： H04N5/228

CPC分类号： H04N5/2327 ,  H04N5/145 ,  H04N5/23248 ,  H04N5/23254 ,  H04N5/23258 ,  H04N5/23274 ,  H04N19/527

摘要： Video data of successive image frames are processed in a digital camera or other video image data acquisition device immediately after acquiring the video data, or during post-processing of the video data at some time after acquiring the video data, to reduce the effects of unintended motion (jitter) of the hand-held devices by stabilizing the images. A processing circuit used to calculate an estimate of motion between components of successive image frames as part of a MPEG-4 or other compression algorithm is also used to estimate motion upon which the video data are altered to stabilize the images. The individual images may be pre-stabilized by using the results of stabilizing a prior image in order to reduce the amount of processing necessary to stabilize a current image.

摘要翻译： 连续图像帧的视频数据在获取视频数据之后立即在数字照相机或其他视频图像数据采集装置中进行处理，或者在获取视频数据之后的某个时间对视频数据进行后处理，以减少意外的影响 通过稳定图像来实现手持设备的运动（抖动）。 用于计算作为MPEG-4或其他压缩算法的一部分的连续图像帧的分量之间的运动估计的处理电路也用于估计视频数据被改变以使图像稳定的运动。 可以通过使用稳定现有图像的结果来预先确定各个图像，以便减少稳定当前图像所需的处理量。

公开(公告)号：US20130034300A1

公开(公告)日：2013-02-07

申请号：US13540440

申请日：2012-07-02

申请人： Gal Ben-David ,  Victor Pinto ,  Itsik Dvir ,  Yonatan Simson

发明人： Gal Ben-David ,  Victor Pinto ,  Itsik Dvir ,  Yonatan Simson

IPC分类号： G06K9/36

CPC分类号： H04N19/625 ,  H04N19/126 ,  H04N19/14 ,  H04N19/17 ,  H04N19/174 ,  H04N19/18 ,  H04N19/186

摘要： Embodiments are directed towards compressing an image to substantially a preset file size using statistical information obtained from a single subset of the image and an initial compression of the single subset. A representative subset portion of the image is selected based in part on a clustering analysis of the image. The representative subset is then compressed, in one embodiment, twice, in order to obtain statistics useable for the entire image. A scale factor is then determined that may be used in the quantization and for creating a Bit Rate Control (BRC) curve that represents an amount of accumulated bits per Minimal Codec Unit (MCU). During the compression process, the BRC curve is used to prevent accumulating bits from over shooting a final preset file size target.

摘要翻译： 实施例旨在使用从图像的单个子集获得的统计信息和单个子集的初始压缩将图像压缩到基本上预设的文件大小。 部分地基于图像的聚类分析来选择图像的代表性子集部分。 然后，在一个实施例中，代表性子集被压缩两次，以获得可用于整个图像的统计信息。 然后确定可用于量化和用于创建表示每个最小编解码单元（MCU）的累加比特量的比特率控制（BRC）曲线的比例因子。 在压缩过程中，BRC曲线用于防止累积位过度拍摄最终预设的文件大小目标。

公开(公告)号：US08289401B2

公开(公告)日：2012-10-16

申请号：US12767753

申请日：2010-04-26

申请人： Victor Pinto ,  Itsik Dvir

发明人： Victor Pinto ,  Itsik Dvir

IPC分类号： H04N5/228

CPC分类号： H04N5/2327 ,  H04N5/145 ,  H04N5/23248 ,  H04N5/23254 ,  H04N5/23258 ,  H04N5/23274 ,  H04N19/527

摘要： Video data that includes a plurality of images acquired over time are processed in an apparatus immediately after acquiring the video data, or during some later post-processing of the video data, to reduce the effects of unintended motion in the acquired images. Coarse and fine compensation for motion is performed on video frames that correspond to the acquired images. A coarse compensation value is employed to compress each video frame that is both coarse and fine motion compensated.

摘要翻译： 包括随时间采集的多个图像的视频数据在获取视频数据之后或在视频数据的稍后后处理期间的设备中被处理，以减少所获取的图像中的非预期运动的影响。 在对应于所获取的图像的视频帧上执行用于运动的粗略和精细补偿。 采用粗略的补偿值来压缩粗略和微细运动补偿的每个视频帧。

公开(公告)号：US20080292206A1

公开(公告)日：2008-11-27

申请号：US11752817

申请日：2007-05-23

申请人： Itsik Dvir

发明人： Itsik Dvir

IPC分类号： G06K9/40

CPC分类号： G06T5/009

摘要： The described methods and corresponding apparatus are for use in highlight recovery in a set of image data. The image data is inverted, treated by a digital lighting process, the subsequently inverted again. This allows the set of data image to be treated for dynamic range compression of detail in both the dark regions and the light regions by the same algorithm. In other words, the process applied to the dark regions can be performed by a correction algorithm and the process applied to the light regions performed by the same algorithm, but with the image data negated before and after the correction algorithm is applied.

摘要翻译： 所描述的方法和相应的装置用于一组图像数据中的高光恢复。 图像数据被反转，通过数字照明处理，随后再次反转。 这允许通过相同的算法来处理在暗区域和光区域中的细节的动态范围压缩的数据图像集合。 换句话说，施加到暗区域的处理可以通过校正算法执行，并且应用于通过相同算法执行的光区域的处理，但是在应用校正算法之前和之后使图像数据被否定。

公开(公告)号：US08503815B2

公开(公告)日：2013-08-06

申请号：US11752817

申请日：2007-05-23

申请人： Itsik Dvir

发明人： Itsik Dvir

IPC分类号： G06K9/40

CPC分类号： G06T5/009

摘要： The described methods and corresponding apparatus are for use in highlight recovery in a set of image data. The image data is inverted, treated by a digital lighting process, the subsequently inverted again. This allows the set of data image to be treated for dynamic range compression of detail in both the dark regions and the light regions by the same algorithm. In other words, the process applied to the dark regions can be performed by a correction algorithm and the process applied to the light regions performed by the same algorithm, but with the image data negated before and after the correction algorithm is applied.

摘要翻译： 所描述的方法和相应的装置用于一组图像数据中的高光恢复。 图像数据被反转，通过数字照明处理，随后再次反转。 这允许通过相同的算法来处理在暗区域和光区域中的细节的动态范围压缩的数据图像集合。 换句话说，施加到暗区域的处理可以通过校正算法执行，并且应用于通过相同算法执行的光区域的处理，但是在应用校正算法之前和之后使图像数据被否定。

公开(公告)号：US07899267B2

公开(公告)日：2011-03-01

申请号：US11752797

申请日：2007-05-23

申请人： Itsik Dvir

发明人： Itsik Dvir

IPC分类号： G06K9/40

CPC分类号： H04N5/243 ,  H04N1/4072 ,  H04N9/68 ,  H04N9/735

摘要： A local method uses a cascade of multiple nonlinear edge preserving filters, and nonlinear pixel point operations, to calculate the pixel gain. Multiple low-pass filters are used, being applied to surrounding neighborhoods of the current pixel of narrow and of wide extent. The number of filter stages may be determined based on the image content. The coefficients used to combine a gray level image extracted from the input image with the ascending scale regions can be automatically extracted from high-pass filtered images of the ascending scale regions. Multiplying each color component of the input image by one or more pixel dependent gain or attenuation factors, using a nonlinear mapping function that can lighten shadow regions as well as darken bright regions, generates the output image.

摘要翻译： 本地方法使用多个非线性边缘保留滤波器的级联和非线性像素点运算来计算像素增益。 使用多个低通滤波器，被应用于窄范围和当前像素的周边邻域。 可以基于图像内容来确定滤波器级的数量。 用于组合从输入图像提取的灰度级图像与上升标度区域的系数可以从上标尺度区域的高通滤波图像中自动提取。 使用可以减轻阴影区域和变暗亮区域的非线性映射函数，将输入图像的每个颜色分量乘以一个或多个像素依赖增益或衰减因子，生成输出图像。

公开(公告)号：US20090161964A1

公开(公告)日：2009-06-25

申请号：US12023877

申请日：2008-01-31

申请人： Meir Tzur ,  Ben-Zion Shaick ,  Itsik Dvir ,  Victor Pinto

发明人： Meir Tzur ,  Ben-Zion Shaick ,  Itsik Dvir ,  Victor Pinto

IPC分类号： G06K9/46

CPC分类号： G06K9/60 ,  G06K9/3208 ,  G06K9/62 ,  G06K9/6282 ,  G06T3/60 ,  H04N5/23219 ,  H04N5/23229 ,  H04N2101/00

摘要： The likelihood of a particular type of object, such as a human face, being present within a digital image, and its location in that image, are determined by comparing the image data within defined windows across the image in sequence with two or more sets of data representing features of the particular type of object. The evaluation of each set of features after the first is preferably performed only on data of those windows that pass the evaluation with respect to the first set of features, thereby quickly narrowing potential target windows that contain at least some portion of the object. Correlation scores are preferably calculated by the use of non-linear interpolation techniques in order to obtain a more refined score. Evaluation of the individual windows also preferably includes maintaining separate feature set data for various positions of the object around one axis and rotating the feature set data with respect to the image data for the individual windows about another axis.

摘要翻译： 通过将图像中定义的窗口内的图像数据与两个或更多个组合的顺序进行比较来确定特定类型的对象（例如人脸）存在于数字图像内的可能性及其在该图像中的位置 表示特定类型对象的特征的数据。 优选地，对第一组之后的每组特征的评估仅对通过相对于第一组特征的评估的那些窗口的数据执行，从而快速缩小包含对象的至少一部分的潜在目标窗口。 相关分数优选通过使用非线性内插技术来计算，以获得更精细的分数。 各个窗口的评估还优选地包括为围绕一个轴的对象的各种位置维护单独的特征集数据，并且相对于关于另一个轴的各个窗口的图像数据旋转特征集数据。

公开(公告)号：US08379922B2

公开(公告)日：2013-02-19

申请号：US13099304

申请日：2011-05-02

申请人： Meir Tzur ,  Ben-Zion Shaick ,  Itsik Dvir ,  Victor Pinto

发明人： Meir Tzur ,  Ben-Zion Shaick ,  Itsik Dvir ,  Victor Pinto

IPC分类号： G06K9/00 ,  H04N5/225 ,  G06K9/68

CPC分类号： G06K9/60 ,  G06K9/3208 ,  G06K9/62 ,  G06K9/6282 ,  G06T3/60 ,  H04N5/23219 ,  H04N5/23229 ,  H04N2101/00

摘要： The likelihood of a particular type of object, such as a human face, being present within a digital image, and its location in that image, are determined by comparing the image data within defined windows across the image in sequence with two or more sets of data representing features of the particular type of object. The evaluation of each set of features after the first is preferably performed only on data of those windows that pass the evaluation with respect to the first set of features, thereby quickly narrowing potential target windows that contain at least some portion of the object. Correlation scores are preferably calculated by the use of non-linear interpolation techniques in order to obtain a more refined score. Evaluation of the individual windows also preferably includes maintaining separate feature set data for various positions of the object around one axis and rotating the feature set data with respect to the image data for the individual windows about another axis.

摘要翻译： 通过将图像中定义的窗口内的图像数据与两个或更多个组合的顺序进行比较来确定特定类型的对象（例如人脸）存在于数字图像内的可能性及其在该图像中的位置 表示特定类型对象的特征的数据。 优选地，对第一组之后的每组特征的评估仅对通过相对于第一组特征的评估的那些窗口的数据执行，从而快速缩小包含对象的至少一部分的潜在目标窗口。 相关分数优选通过使用非线性内插技术来计算，以获得更精细的分数。 各个窗口的评估还优选地包括为围绕一个轴的对象的各种位置维护单独的特征集数据，并且相对于关于另一个轴的各个窗口的图像数据旋转特征集数据。

公开(公告)号：US20110235910A1

公开(公告)日：2011-09-29

申请号：US13001631

申请日：2010-06-30

申请人： Omri Soceanu ,  Guy Berdugo ,  Yair Moshe ,  Dmitry Rudoy ,  Itsik Dvir ,  Dan Raudnitz

发明人： Omri Soceanu ,  Guy Berdugo ,  Yair Moshe ,  Dmitry Rudoy ,  Itsik Dvir ,  Dan Raudnitz

IPC分类号： G06K9/46 ,  G06K9/34

CPC分类号： G06K9/00369 ,  G06K9/00295

摘要： Disclosed is a system and method for image processing and image subject matching. A circuit and system may be used for matching/correlating an object/subject or person present (i.e. visible within) within two or more images. An object or person present within a first image or a first series of images (e.g. a video sequence) may be characterized and the characterization information (i.e. one or a set of parameters) relating to the person or object may be stored in a database, random access memory or cache for subsequent comparison to characterization information derived from other images.

摘要翻译： 公开了一种用于图像处理和图像对象匹配的系统和方法。 可以使用电路和系统来匹配/关联在两个或更多个图像内的对象/被摄体或存在的人物（即可见的内部）。 可以对存在于第一图像或第一系列图像（例如，视频序列）中的对象或人物进行表征，并且与个人或对象相关的表征信息（即一个或一组参数）可以存储在数据库中， 随机存取存储器或缓存，用于随后与从其他图像导出的表征信息进行比较。