公开(公告)号：US20100085475A1

公开(公告)日：2010-04-08

申请号：US12286798

申请日：2008-10-02

申请人： Shekhar Madnani ,  Shailesh Kumar

发明人： Shekhar Madnani ,  Shailesh Kumar

IPC分类号： H04N7/01

CPC分类号： H04N17/00 ,  H04N5/142 ,  H04N5/145

摘要： A method and system for calculating an interlace artifact in image data are disclosed. A motion picture of the image data comprises a series of frames, captured at a predefined interval of time. During processing of the motion picture, the frames are divided into fields, each field comprising one or more pixels. A difference between the pixels of the fields is calculated. Thereafter, edges of the pixels are calculated in the fields. The method and system then identify the focused area in the fields. To calculate the interlace artifact in the motion picture, the displacement of the focused area is calculated by using motion vectors. The artifacts are calculated as a ratio of a number of pixels based on motion vector calculation.

摘要翻译： 公开了一种用于计算图像数据中的隔行伪影的方法和系统。 图像数据的运动图像包括以预定的时间间隔捕获的一系列帧。 在运动图像的处理期间，帧被分成场，每个场包括一个或多个像素。 计算场的像素之间的差异。 此后，在场中计算像素的边缘。 然后，该方法和系统确定了该领域的重点领域。 为了计算运动图像中的交错伪影，通过使用运动矢量来计算聚焦区域的位移。 伪像被计算为基于运动矢量计算的像素数的比率。

公开(公告)号：US20190385569A1

公开(公告)日：2019-12-19

申请号：US15908835

申请日：2018-03-01

申请人： ALOK KUMAR ,  BHUPENDER KUMAR ,  SHEKHAR MADNANI

发明人： ALOK KUMAR ,  BHUPENDER KUMAR ,  SHEKHAR MADNANI

IPC分类号： G09G5/10 ,  G09G3/36

摘要： A system and a method for adjusting luminance flashes in a video stream are disclosed. The method includes identifying a sequence of frames, having the luminance flashes, from the video stream. The sequence of frames is extended at both sides (start and end) based on a predefined threshold. The method further comprises dividing an extended sequence of frames into at least three segments. Further, the method includes determining a correction factor and a correction constant for each of the at least three segments. Thereafter, the method includes modifying luminance values of pixels of each of the at least three segments based on the correction factor and the correction constant, thereby adjusting the luminance flashes in the video stream.

公开(公告)号：US20130259392A1

公开(公告)日：2013-10-03

申请号：US13549507

申请日：2012-07-15

申请人： Praney Mahajan ,  Shekhar Madnani

发明人： Praney Mahajan ,  Shekhar Madnani

IPC分类号： G06K9/36

CPC分类号： H04N7/0115

摘要： A system and method for the identification and analysis of cadence pattern is disclosed. The method uses previous and current fields to generate the difference between the field values. The difference of these values along with the field relations is passed to the state machine to generate the state of the top and bottom fields. Based on the top and bottom state the cadence signature is generated and by using the Fourier analysis the principle frequency of repeated cadence pattern signature sequence is identified. Each of the cadence signatures present in the cadence pattern signature sequence is decoded to calculate the pull-down value of the cadence pattern. The pull down value then gives the actual cadence pattern.

摘要翻译： 公开了一种用于识别和分析节奏模式的系统和方法。 该方法使用先前和当前字段来生成字段值之间的差异。 这些值与字段关系的差异被传递到状态机以产生顶部和底部字段的状态。 基于顶部和底部状态，生成节奏签名，并通过傅里叶分析，确定重复节奏模式签名序列的主要频率。 存在于节奏模式签名序列中的每个节奏签名被解码以计算节奏模式的下拉值。 然后，下拉值给出实际的节奏模式。

公开(公告)号：US08049817B2

公开(公告)日：2011-11-01

申请号：US12286798

申请日：2008-10-02

申请人： Shekhar Madnani ,  Shailesh Kumar

发明人： Shekhar Madnani ,  Shailesh Kumar

IPC分类号： H04N7/01

CPC分类号： H04N17/00 ,  H04N5/142 ,  H04N5/145

摘要： A method and system for calculating an interlace artifact in image data are disclosed. A motion picture of the image data comprises a series of frames, captured at a predefined interval of time. During processing of the motion picture, the frames are divided into fields, each field comprising one or more pixels. A difference between the pixels of the fields is calculated. Thereafter, edges of the pixels are calculated in the fields. The method and system then identify the focused area in the fields. To calculate the interlace artifact in the motion picture, the displacement of the focused area is calculated by using motion vectors. The artifacts are calculated as a ratio of a number of pixels based on motion vector calculation.

摘要翻译： 公开了一种用于计算图像数据中的隔行伪影的方法和系统。 图像数据的运动图像包括以预定的时间间隔捕获的一系列帧。 在运动图像的处理期间，帧被分成场，每个场包括一个或多个像素。 计算场的像素之间的差异。 此后，在场中计算像素的边缘。 然后，该方法和系统确定了该领域的重点领域。 为了计算运动图像中的交错伪影，通过使用运动矢量来计算聚焦区域的位移。 伪像被计算为基于运动矢量计算的像素数的比率。

公开(公告)号：US08432976B2

公开(公告)日：2013-04-30

申请号：US12930640

申请日：2011-01-12

申请人： Shekhar Madnani ,  Raman Kumar Sonkhla

发明人： Shekhar Madnani ,  Raman Kumar Sonkhla

IPC分类号： H04N7/12

CPC分类号： H04N5/44 ,  H04N5/144 ,  H04N7/0115

摘要： The invention provides a method, system and computer program product for detecting field order of a video sequence. The method includes processing a top and bottom field picture of a first frame and a second frame, in which the first and second frames are consecutive frames of a video sequence. Difference frames are obtained by calculating the difference of the top and bottom field pictures of the first and second frames. Thereafter, field order is detected using the difference frames.

摘要翻译： 本发明提供一种用于检测视频序列的场顺序的方法，系统和计算机程序产品。 该方法包括处理第一帧和第二帧的顶部和底部场图像，其中第一帧和第二帧是视频序列的连续帧。 通过计算第一和第二帧的顶部和底部场图像的差异来获得差分帧。 此后，使用差分帧检测场顺序。

公开(公告)号：US20100092100A1

公开(公告)日：2010-04-15

申请号：US12287830

申请日：2008-10-14

申请人： Shekhar Madnani

发明人： Shekhar Madnani

IPC分类号： G06K9/40

CPC分类号： G06T7/0002 ,  G06T5/003 ,  G06T2207/10016 ,  G06T2207/30168

摘要： A method and system for calculating a blur artifact in a video are disclosed. The video includes a series of frames captured at a predefined interval of time. The frames include one or more pixels. Calculating the blur artifact in the video includes identifying a focused area in a set of frames. Further, edges are detected in each of the frames. Furthermore, the blur artifact is calculated as a ratio of number of blurred pixels and total number edge pixels.

摘要翻译： 公开了一种用于计算视频中的模糊伪影的方法和系统。 视频包括以预定的时间间隔捕获的一系列帧。 帧包括一个或多个像素。 计算视频中的模糊伪像包括识别一组帧中的聚焦区域。 此外，在每个帧中检测边缘。 此外，模糊伪像被计算为模糊像素数和总数边缘像素的比率。

公开(公告)号：US20120176538A1

公开(公告)日：2012-07-12

申请号：US12930640

申请日：2011-01-12

申请人： Shekhar Madnani ,  Raman Kumar Sonkhla

发明人： Shekhar Madnani ,  Raman Kumar Sonkhla

IPC分类号： H04N7/01

CPC分类号： H04N5/44 ,  H04N5/144 ,  H04N7/0115

摘要： The invention provides a method, system and computer program product for detecting field order of a video sequence. The method includes processing a top and bottom field picture of a first frame and a second frame, in which the first and second frames are consecutive frames of a video sequence. Difference frames are obtained by calculating the difference of the top and bottom field pictures of the first and second frames. Thereafter, field order is detected using the difference frames.

摘要翻译： 本发明提供一种用于检测视频序列的场顺序的方法，系统和计算机程序产品。 该方法包括处理第一帧和第二帧的顶部和底部场图像，其中第一帧和第二帧是视频序列的连续帧。 通过计算第一和第二帧的顶部和底部场图像的差异来获得差分帧。 此后，使用差分帧检测场顺序。

公开(公告)号：US08768074B2

公开(公告)日：2014-07-01

申请号：US13549507

申请日：2012-07-15

申请人： Praney Mahajan ,  Shekhar Madnani

发明人： Praney Mahajan ,  Shekhar Madnani

IPC分类号： G06K9/46

CPC分类号： H04N7/0115

摘要： A system and method for the identification and analysis of cadence pattern is disclosed. The method uses previous and current fields to generate the difference between the field values. The difference of these values along with the field relations is passed to the state machine to generate the state of the top and bottom fields. Based on the top and bottom state the cadence signature is generated and by using the Fourier analysis the principle frequency of repeated cadence pattern signature sequence is identified. Each of the cadence signatures present in the cadence pattern signature sequence is decoded to calculate the pull-down value of the cadence pattern. The pull down value then gives the actual cadence pattern.

摘要翻译： 公开了一种用于识别和分析节奏模式的系统和方法。 该方法使用先前和当前字段来生成字段值之间的差异。 这些值与字段关系的差异被传递到状态机以产生顶部和底部字段的状态。 基于顶部和底部状态，生成节奏签名，并通过傅里叶分析，确定重复节奏模式签名序列的主要频率。 存在于节奏模式签名序列中的每个节奏签名被解码以计算节奏模式的下拉值。 然后，下拉值给出实际的节奏模式。

公开(公告)号：US08150191B2

公开(公告)日：2012-04-03

申请号：US12287830

申请日：2008-10-14

申请人： Shekhar Madnani

发明人： Shekhar Madnani

IPC分类号： G06K9/40

CPC分类号： G06T7/0002 ,  G06T5/003 ,  G06T2207/10016 ,  G06T2207/30168

摘要： A method and system for calculating a blur artifact in a video are disclosed. The video includes a series of frames captured at a predefined interval of time. The frames include one or more pixels. Calculating the blur artifact in the video includes identifying a focused area in a set of frames. Further, edges are detected in each of the frames. Furthermore, the blur artifact is calculated as a ratio of number of blurred pixels and total number edge pixels.

摘要翻译： 公开了一种用于计算视频中的模糊伪影的方法和系统。 视频包括以预定的时间间隔捕获的一系列帧。 帧包括一个或多个像素。 计算视频中的模糊伪像包括识别一组帧中的聚焦区域。 此外，在每个帧中检测边缘。 此外，模糊伪像被计算为模糊像素数和总数边缘像素的比率。