公开(公告)号：US09100717B2

公开(公告)日：2015-08-04

申请号：US14049919

申请日：2013-10-09

申请人： Interra Systems, Inc.

发明人： Ramandeep Singh Sandhu ,  Kulbhushan ,  Manik Gupta

IPC分类号： H04H20/14 ,  H04N21/80 ,  G06F9/38

CPC分类号： H04N21/80 ,  G06F9/3851 ,  H04N21/44008 ,  H04N21/4425 ,  H04N21/8455 ,  H04N21/8456

摘要： The disclosed embodiments relates to methods and systems for performing a content verification in a video content. The method includes identification of one or more random access points in the video content. Thereafter, one or more work units are created based on the identified one or more random access points. The method further includes processing of the one or more work units by one or more processing units simultaneously, and determining one or more errors in each of the one or more work units while the decoding. The one or more determined errors in each of processed one or more work units are then merged with each other. A verification report is generated based on at least the merging.

摘要翻译： 所公开的实施例涉及用于在视频内容中执行内容验证的方法和系统。 该方法包括识别视频内容中的一个或多个随机接入点。 此后，基于所识别的一个或多个随机接入点来创建一个或多个工作单元。 该方法还包括同时处理一个或多个处理单元的一个或多个工作单元，并且在解码时确定一个或多个工作单元中的每一个中的一个或多个错误。 然后，处理的一个或多个工作单元中的每一个中的一个或多个确定的错误彼此合并。 至少基于合并生成验证报告。

公开(公告)号：US20150078654A1

公开(公告)日：2015-03-19

申请号：US14027123

申请日：2013-09-13

申请人： Interra Systems, Inc.

发明人： Santanu Chaudhury ,  Brejesh Lall ,  Rohit Mungre ,  Yogesh Singh ,  Shekhar Madnani

IPC分类号： G06K9/62 ,  G06T7/00 ,  G06K9/46

CPC分类号： G06K9/00765 ,  G06K9/00744 ,  G06K9/036 ,  G06K9/4604 ,  G06T7/0002 ,  G06T2207/10016 ,  G06T2207/30168

摘要： System and method for identifying erroneous videos and assessing video quality is provided. Feature vectors are generated corresponding to a plurality of frames associated with the one or more videos. The feature vectors are subsequently subjected to anomaly detection to obtain first and second normalized path lengths and normalized anomaly measures. The first and second normalized path lengths and normalized anomaly measures are provided to a regression model to identify the erroneous video.

摘要翻译： 提供了识别错误视频和评估视频质量的系统和方法。 对应于与一个或多个视频相关联的多个帧产生特征向量。 特征向量随后进行异常检测，以获得第一和第二归一化路径长度和归一化异常测量。 将第一和第二标准化路径长度和归一化异常测量值提供给回归模型以识别错误视频。

公开(公告)号：US08111929B2

公开(公告)日：2012-02-07

申请号：US12072977

申请日：2008-02-29

申请人： Praveen Kumar Tiwari ,  Abhishek Kumar Tiwary ,  Anuj Verma ,  Manik Gupta ,  Prabhanjana Kumar Nallani

发明人： Praveen Kumar Tiwari ,  Abhishek Kumar Tiwary ,  Anuj Verma ,  Manik Gupta ,  Prabhanjana Kumar Nallani

IPC分类号： G06K9/36

CPC分类号： H04N19/86 ,  G06T7/0002 ,  G06T2207/10016 ,  G06T2207/20021 ,  G06T2207/30168

摘要： A method and system for detecting and quantifying blockiness in a video file is disclosed. The video file is a file that has been decompressed by using standard DCT algorithms. The method includes segmenting each frame of the video file into multiple blocks. The method also involves comparing the intensity gradients of each block with one or more threshold values. The one or more threshold values represent predefined intensity variation characteristics. Further, the method includes determining the intensity variation parameters of each block, based on the comparison. Thereafter, a blockiness index is calculated for each block, after which a blockiness value is calculated for each frame. Finally, a blockiness level is assigned to each frame, based on its blockiness value. The blockiness level is a comparative measure of the blockiness of a frame that represents the blockiness content in the frame.

摘要翻译： 公开了一种用于检测和量化视频文件中的块效应的方法和系统。 视频文件是使用标准DCT算法解压缩的文件。 该方法包括将视频文件的每个帧分割成多个块。 该方法还涉及将每个块的强度梯度与一个或多个阈值进行比较。 一个或多个阈值表示预定的强度变化特征。 此外，该方法包括基于该比较来确定每个块的强度变化参数。 此后，针对每个块计算块符号索引，之后为每个帧计算块效应值。 最后，基于其阻塞值，为每个帧分配一个阻塞级别。 阻塞水平是表示帧中的块状内容的帧的块性的比较度量。