专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US08965106B1 Detecting image inconsistencies 有权
标题翻译：检测图像不一致
公开(公告)号：US08965106B1
公开(公告)日：2015-02-24
申请号：US13472804
申请日：2012-05-16
申请人： Hany Farid , Eric Kee , James O'Brien
发明人： Hany Farid , Eric Kee , James O'Brien
IPC分类号： G06K9/00
CPC分类号： G06T7/0002 , G06K9/00664 , G06K9/4661
摘要： Detecting a consistency of image components is disclosed. A first constraint associated with a possible convergence area of a first component of an image is received. A second constraint associated with a possible convergence area of a second component of the image is received. The first constraint and the second constraint are used to provide a result associated with whether the first component and the second component are consistent.
摘要翻译：公开了检测图像分量的一致性。接收与图像的第一分量的可能会聚区域相关联的第一约束。接收与图像的第二分量的可能会聚区域相关联的第二约束。第一约束和第二约束用于提供与第一分量和第二分量是否一致的结果。

2. 发明申请

US20100309343A1 DEVICE AND METHOD FOR DETECTING DUST ON AN IMAGE SENSOR 有权
标题翻译：用于检测图像传感器上的尘埃的装置和方法
公开(公告)号：US20100309343A1
公开(公告)日：2010-12-09
申请号：US12864743
申请日：2008-10-29
申请人： Hany Farid
发明人： Hany Farid
IPC分类号： H04N5/217 , G06K9/40
CPC分类号： G06K9/4661 , H04N5/217 , H04N5/2171 , H04N2201/0084
摘要： A system and method for detecting dust (18) on an image sensor (20) from a single captured image (14) of an actual scene (12) includes a control system (22) that evaluates at least one of a hue channel (466), a value channel (470), and a saturation channel (468) of the captured image (14) to determine if there is dust (18) on the image sensor (20). For example, the control system (22) can evaluate the hue channel (466) and the value channel (470) of the captured image (14) to determine if there is dust (18) on the image sensor (20). With information from the hue channel (466) and the value channel (470), the control system (22) can compute a computed probability (572) of dust (18) for a plurality of pixels (362) of the captured image (14).
摘要翻译：一种用于从实际场景（12）的单个拍摄图像（14）检测图像传感器（20）上的灰尘（18）的系统和方法包括：控制系统（22），其评估色调通道（466 ），所述捕获图像（14）的值通道（470）和饱和通道（468），以确定所述图像传感器（20）上是否存在尘埃（18）。例如，控制系统（22）可以评估捕获图像（14）的色相通道（466）和值通道（470），以确定图像传感器（20）上是否有灰尘（18）。通过来自色调通道（466）和值通道（470）的信息，控制系统（22）可以计算所捕获图像（14）的多个像素（362）的灰尘（18）的计算概率（572））。

3. 发明申请

US20110019909A1 DEVICE AND METHOD FOR DETECTING WHETHER AN IMAGE IS BLURRED 有权
标题翻译：用于检测图像的设备和方法
公开(公告)号：US20110019909A1
公开(公告)日：2011-01-27
申请号：US12933857
申请日：2009-06-22
申请人： Hany Farid , Li Hong
发明人： Hany Farid , Li Hong
IPC分类号： G06K9/62
CPC分类号： G06K9/6269 , G06K9/6234 , H04N5/23248
摘要： The present invention is directed to a method for detecting or predicting (302, 602) whether a test image is blurred. In one embodiment, the method includes extracting a training statistical signature (366) that is based on a plurality of data features (362, 364) from a training image set (14, 16), the training image set (14, 16) including a sharp image (14) and a blurry image (16); training a classifier (368) to discriminate between the sharp image (14) and the blurry image (16) based on the training statistical signature; and applying (302, 602) the trained classifier to a test image that is not included in the training image set (14, 16) to predict whether the test image is sharp (18) or blurry (20). The step of extracting can include measuring one or more statistical moments (576, 776) for various levels (L0-L5), estimating a covariance (577, 777) between adjacent levels (L0-L5), and/or extracting various metadata features (364, 664) from the images (14, 16). The step of training (300, 600) can include training a non-linear support vector machine (300) or a linear discriminant analysis (600) on the training statistical signature of the training image set (14, 16).
摘要翻译：本发明涉及一种用于检测或预测（302,602）测试图像是否模糊的方法。在一个实施例中，该方法包括从训练图像集（14,16）提取基于多个数据特征（362,364）的训练统计签名（366），训练图像集（14,16）包括清晰图像（14）和模糊图像（16）; 训练分类器（368）以基于训练统计签名来区分锐利图像（14）和模糊图像（16）; 以及将所训练的分类器应用（302,602）到不包括在训练图像集（14,16）中的测试图像，以预测测试图像是否锐利（18）或模糊（20）。提取步骤可以包括测量各种级别（L0-L5）的一个或多个统计矩（576,776），估计相邻级别（L0-L5）之间的协方差（577,777）和/或提取各种元数据特征（14,36,64）。训练步骤（300,600）可以包括在训练图像集（14,16）的训练统计签名上训练非线性支持向量机（300）或线性判别分析（600）。

4. 发明授权

US08654217B2 Device and method for detecting dust on an image sensor 有权
标题翻译：用于检测图像传感器上的灰尘的装置和方法
公开(公告)号：US08654217B2
公开(公告)日：2014-02-18
申请号：US12864743
申请日：2008-10-29
申请人： Hany Farid
发明人： Hany Farid
IPC分类号： H04N5/217
CPC分类号： G06K9/4661 , H04N5/217 , H04N5/2171 , H04N2201/0084
摘要： A system and method for detecting dust (18) on an image sensor (20) from a single captured image (14) of an actual scene (12) includes a control system (22) that evaluates at least one of a hue channel (466), a value channel (470), and a saturation channel (468) of the captured image (14) to determine if there is dust (18) on the image sensor (20). For example, the control system (22) can evaluate the hue channel (466) and the value channel (470) of the captured image (14) to determine if there is dust (18) on the image sensor (20). With information from the hue channel (466) and the value channel (470), the control system (22) can compute a computed probability (572) of dust (18) for a plurality of pixels (362) of the captured image (14).
摘要翻译：一种用于从实际场景（12）的单个拍摄图像（14）检测图像传感器（20）上的灰尘（18）的系统和方法包括：控制系统（22），其评估色调通道（466 ），所述捕获图像（14）的值通道（470）和饱和通道（468），以确定所述图像传感器（20）上是否存在尘埃（18）。例如，控制系统（22）可以评估捕获图像（14）的色相通道（466）和值通道（470），以确定图像传感器（20）上是否有灰尘（18）。通过来自色调通道（466）和值通道（470）的信息，控制系统（22）可以计算所捕获图像（14）的多个像素（362）的灰尘（18）的计算概率（572））。

5. 发明授权

US08538140B2 Device and method for detecting whether an image is blurred 有权
标题翻译：用于检测图像是否模糊的装置和方法
公开(公告)号：US08538140B2
公开(公告)日：2013-09-17
申请号：US12933857
申请日：2009-06-22
申请人： Hany Farid , Li Hong
发明人： Hany Farid , Li Hong
IPC分类号： G06K9/00
CPC分类号： G06K9/6269 , G06K9/6234 , H04N5/23248
摘要： The present invention is directed to a method for detecting or predicting (302, 602) whether a test image is blurred. In one embodiment, the method includes extracting a training statistical signature (366) that is based on a plurality of data features (362, 364) from a training image set (14, 16), the training image set (14, 16) including a sharp image (14) and a blurry image (16); training a classifier (368) to discriminate between the sharp image (14) and the blurry image (16) based on the training statistical signature; and applying (302, 602) the trained classifier to a test image that is not included in the training image set (14, 16) to predict whether the test image is sharp (18) or blurry (20). The step of extracting can include measuring one or more statistical moments (576, 776) for various levels (L0-L5), estimating a covariance (577, 777) between adjacent levels (L0-L5), and/or extracting various metadata features (364, 664) from the images (14, 16). The step of training (300, 600) can include training a non-linear support vector machine (300) or a linear discriminant analysis (600) on the training statistical signature of the training image set (14, 16).
摘要翻译：本发明涉及一种用于检测或预测（302,602）测试图像是否模糊的方法。在一个实施例中，该方法包括从训练图像集（14,16）提取基于多个数据特征（362,364）的训练统计签名（366），训练图像集（14,16）包括清晰图像（14）和模糊图像（16）; 训练分类器（368）以基于训练统计签名来区分锐利图像（14）和模糊图像（16）; 以及将所训练的分类器应用（302,602）到不包括在训练图像集（14,16）中的测试图像，以预测测试图像是否锐利（18）或模糊（20）。提取步骤可以包括测量各种级别（L0-L5）的一个或多个统计矩（576,776），估计相邻级别（L0-L5）之间的协方差（577,777）和/或提取各种元数据特征（14,36,64）。训练步骤（300,600）可以包括在训练图像集（14,16）的训练统计签名上训练非线性支持向量机（300）或线性判别分析（600）。

6. 发明授权

US5703677A Single lens range imaging method and apparatus 失效
标题翻译：单镜头范围成像方法和装置
公开(公告)号：US5703677A
公开(公告)日：1997-12-30
申请号：US557527
申请日：1995-11-14
申请人： Eero Simoncelli , Hany Farid
发明人： Eero Simoncelli , Hany Farid
IPC分类号： G02B7/28 , G02B26/02 , G01C3/08
CPC分类号： G02B7/28 , G02B26/02
摘要： A single lens range sensor comprises a camera 27 having a pair of plano-convex lenses 28a and 28b, an optical mask 30, an imaging sensor 32, a digitizer 34 and a computer 36. The optical mask 30 may be implemented with a liquid crystal array having spatially varying opacity which can be switched (e.g., by the computer 36) between masks derived from a differentiable mask function and its derivative. Alternatively, the mask 30 may be composed of two printed masks which are mechanically switched to obtain two images based on the differentiable mask function and its derivative. The imaging sensor 32 may be implemented with a CCD array. The invention avoids the correspondence problems of the two camera and moving camera approaches, and is simple to calibrate.
摘要翻译：单个透镜范围传感器包括具有一对平凸透镜28a和28b的相机27，光学掩模30，成像传感器32，数字转换器34和计算机36.光学掩模30可以用液晶阵列具有空间变化的不透明度，其可以在从可微分掩模函数导出的掩模与其导数之间切换（例如，由计算机36）。或者，掩模30可以由两个印刷的掩模组成，其被机械地切换以基于可微分掩模函数及其导数获得两个图像。成像传感器32可以用CCD阵列来实现。本发明避免了两台摄像机和移动摄像机的对应关系，简单易行。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式