利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

18 条记录 (耗时 0.033 秒)

    LONG-TERM MEMORY IN A VIDEO ANALYSIS SYSTEM
    2.
    发明申请
    LONG-TERM MEMORY IN A VIDEO ANALYSIS SYSTEM 有权
    视频分析系统中的长期记忆

    公开(公告)号：US20100063949A1

    公开(公告)日：2010-03-11

    申请号：US12208551

    申请日：2008-09-11

    申请人： JOHN ERIC EATON ,  Wesley Kenneth Cobb ,  Ming-Jung Seow ,  David Samuel Friedlander ,  Gang Xu

    发明人： JOHN ERIC EATON ,  Wesley Kenneth Cobb ,  Ming-Jung Seow ,  David Samuel Friedlander ,  Gang Xu

    IPC分类号： G06F15/18 ,  G06N3/08 ,  G06K9/00

    CPC分类号： G06N99/005 ,  G06K9/00335 ,  G06K9/00771 ,  G06K9/6296

    摘要： A long-term memory used to store and retrieve information learned while a video analysis system observes a stream of video frames is disclosed. The long-term memory provides a memory with a capacity that grows in size gracefully, as events are observed over time. Additionally, the long-term memory may encode events, represented by sub-graphs of a neural network. Further, rather than predefining a number of patterns recognized and manipulated by the long-term memory, embodiments of the invention provide a long-term memory where the size of a feature dimension (used to determine the similarity between different observed events) may grow dynamically as necessary, depending on the actual events observed in a sequence of video frames.

    摘要翻译： 公开了一种用于存储和检索在视频分析系统观察视频帧流时学习的信息的长期存储器。 长期记忆提供了一个能力随着时间的推移而随着时间的推移而增长的记忆。 另外，长期记忆可以编码由神经网络的子图表示的事件。 此外，本发明的实施例不是预先确定由长期记忆识别和操纵的多个模式，而是提供长期存储器，其中特征维度（用于确定不同观察到的事件之间的相似性）的大小可以动态地增长 根据需要，取决于视频帧序列中观察到的实际事件。

    Intra-trajectory anomaly detection using adaptive voting experts in a video surveillance system
    3.
    发明授权
    Intra-trajectory anomaly detection using adaptive voting experts in a video surveillance system 有权
    视频监控系统中使用自适应投票专家的轨迹内异常检测

    公开(公告)号：US08379085B2

    公开(公告)日：2013-02-19

    申请号：US12543307

    申请日：2009-08-18

    申请人： Wesley Kenneth Cobb ,  David Samuel Friedlander ,  Kishor Adinath Saitwal

    发明人： Wesley Kenneth Cobb ,  David Samuel Friedlander ,  Kishor Adinath Saitwal

    IPC分类号： H04N7/18 ,  H04N5/225 ,  G06K9/00 ,  G06K9/62

    CPC分类号： G06K9/00771 ,  G06K9/6251 ,  G06K9/726

    摘要： A sequence layer in a machine-learning engine configured to learn from the observations of a computer vision engine. In one embodiment, the machine-learning engine uses the voting experts to segment adaptive resonance theory (ART) network label sequences for different objects observed in a scene. The sequence layer may be configured to observe the ART label sequences and incrementally build, update, and trim, and reorganize an ngram trie for those label sequences. The sequence layer computes the entropies for the nodes in the ngram trie and determines a sliding window length and vote count parameters. Once determined, the sequence layer may segment newly observed sequences to estimate the primitive events observed in the scene as well as issue alerts for inter-sequence and intra-sequence anomalies.

    摘要翻译： 机器学习引擎中的序列层，被配置为从计算机视觉引擎的观察中学习。 在一个实施例中，机器学习引擎使用投票专家来分割在场景中观察到的不同对象的自适应共振理论（ART）网络标签序列。 序列层可以被配置为观察ART标签序列并且逐渐地构建，更新和修整和重组那些标记序列的ngram特里。 序列层计算ngram trie中节点的熵，并确定滑动窗口长度和投票计数参数。 一旦确定，序列层可以划分新观察到的序列以估计在场景中观察到的原始事件以及发出序列间和序列内异常的警报。

    Long-term memory in a video analysis system
    5.
    发明授权
    Long-term memory in a video analysis system 有权
    视频分析系统的长期记忆

    公开(公告)号：US08121968B2

    公开(公告)日：2012-02-21

    申请号：US12208551

    申请日：2008-09-11

    申请人： John Eric Eaton ,  Wesley Kenneth Cobb ,  Ming-Jung Seow ,  David Samuel Friedlander ,  Gang Xu

    发明人： John Eric Eaton ,  Wesley Kenneth Cobb ,  Ming-Jung Seow ,  David Samuel Friedlander ,  Gang Xu

    IPC分类号： G06F17/00 ,  G06F15/18 ,  G06N5/02 ,  G06K9/62 ,  G06K9/46 ,  G06K9/66

    CPC分类号： G06N99/005 ,  G06K9/00335 ,  G06K9/00771 ,  G06K9/6296

    摘要： A long-term memory used to store and retrieve information learned while a video analysis system observes a stream of video frames is disclosed. The long-term memory provides a memory with a capacity that grows in size gracefully, as events are observed over time. Additionally, the long-term memory may encode events, represented by sub-graphs of a neural network. Further, rather than predefining a number of patterns recognized and manipulated by the long-term memory, embodiments of the invention provide a long-term memory where the size of a feature dimension (used to determine the similarity between different observed events) may grow dynamically as necessary, depending on the actual events observed in a sequence of video frames.

    摘要翻译： 公开了一种用于存储和检索在视频分析系统观察视频帧流时学习的信息的长期存储器。 长期记忆提供了一个能力随着时间的推移而随着时间的推移而增长的记忆。 另外，长期记忆可以编码由神经网络的子图表示的事件。 此外，本发明的实施例不是预先确定由长期记忆识别和操纵的多个模式，而是提供长期存储器，其中特征维度（用于确定不同观察到的事件之间的相似性）的大小可以动态地增长 根据需要，取决于视频帧序列中观察到的实际事件。

    INTER-TRAJECTORY ANOMALY DETECTION USING ADAPTIVE VOTING EXPERTS IN A VIDEO SURVEILLANCE SYSTEM
    7.
    发明申请
    INTER-TRAJECTORY ANOMALY DETECTION USING ADAPTIVE VOTING EXPERTS IN A VIDEO SURVEILLANCE SYSTEM 有权
    在视频监控系统中使用自适应投票专家进行异地检测

    公开(公告)号：US20110044499A1

    公开(公告)日：2011-02-24

    申请号：US12543318

    申请日：2009-08-18

    申请人： WESLEY KENNETH COBB ,  David Samuel Friedlander ,  Kishor Adinath Saitwal

    发明人： WESLEY KENNETH COBB ,  David Samuel Friedlander ,  Kishor Adinath Saitwal

    IPC分类号： G06K9/00 ,  G08B21/00

    CPC分类号： G06K9/00771 ,  G06K9/6222 ,  G06K9/6296 ,  G06T7/248 ,  G06T2207/10016 ,  G06T2207/30196 ,  G06T2207/30232 ,  G06T2207/30236 ,  G06T2207/30241

    摘要： A sequence layer in a machine-learning engine configured to learn from the observations of a computer vision engine. In one embodiment, the machine-learning engine uses the voting experts to segment adaptive resonance theory (ART) network label sequences for different objects observed in a scene. The sequence layer may be configured to observe the ART label sequences and incrementally build, update, and trim, and reorganize an ngram trie for those label sequences. The sequence layer computes the entropies for the nodes in the ngram trie and determines a sliding window length and vote count parameters. Once determined, the sequence layer may segment newly observed sequences to estimate the primitive events observed in the scene as well as issue alerts for inter-sequence and intra-sequence anomalies.

    摘要翻译： 机器学习引擎中的序列层，被配置为从计算机视觉引擎的观察中学习。 在一个实施例中，机器学习引擎使用投票专家来分割在场景中观察到的不同对象的自适应共振理论（ART）网络标签序列。 序列层可以被配置为观察ART标签序列并且逐渐地构建，更新和修整和重组那些标记序列的ngram特里。 序列层计算ngram trie中节点的熵，并确定滑动窗口长度和投票计数参数。 一旦确定，序列层可以划分新观察到的序列以估计在场景中观察到的原始事件以及发出序列间和序列内异常的警报。

    INTRA-TRAJECTORY ANOMALY DETECTION USING ADAPTIVE VOTING EXPERTS IN A VIDEO SURVEILLANCE SYSTEM
    8.
    发明申请
    INTRA-TRAJECTORY ANOMALY DETECTION USING ADAPTIVE VOTING EXPERTS IN A VIDEO SURVEILLANCE SYSTEM 有权
    在视频监控系统中使用自适应投票专家进行异常检测

    公开(公告)号：US20110043626A1

    公开(公告)日：2011-02-24

    申请号：US12543307

    申请日：2009-08-18

    申请人： WESLEY KENNETH COBB ,  David Samuel Friedlander ,  Kishor Adinath Saitwal

    发明人： WESLEY KENNETH COBB ,  David Samuel Friedlander ,  Kishor Adinath Saitwal

    IPC分类号： H04N7/18 ,  G06K9/48

    CPC分类号： G06K9/00771 ,  G06K9/6251 ,  G06K9/726

    摘要： A sequence layer in a machine-learning engine configured to learn from the observations of a computer vision engine. In one embodiment, the machine-learning engine uses the voting experts to segment adaptive resonance theory (ART) network label sequences for different objects observed in a scene. The sequence layer may be configured to observe the ART label sequences and incrementally build, update, and trim, and reorganize an ngram trie for those label sequences. The sequence layer computes the entropies for the nodes in the ngram trie and determines a sliding window length and vote count parameters. Once determined, the sequence layer may segment newly observed sequences to estimate the primitive events observed in the scene as well as issue alerts for inter-sequence and intra-sequence anomalies.

    摘要翻译： 机器学习引擎中的序列层，被配置为从计算机视觉引擎的观察中学习。 在一个实施例中，机器学习引擎使用投票专家来分割在场景中观察到的不同对象的自适应共振理论（ART）网络标签序列。 序列层可以被配置为观察ART标签序列并且逐渐地构建，更新和修整和重组那些标记序列的ngram特里。 序列层计算ngram trie中节点的熵，并确定滑动窗口长度和投票计数参数。 一旦确定，序列层可以划分新观察到的序列以估计在场景中观察到的原始事件以及发出序列间和序列内异常的警报。

    Adaptive voting experts for incremental segmentation of sequences with prediction in a video surveillance system
    10.
    发明授权
    Adaptive voting experts for incremental segmentation of sequences with prediction in a video surveillance system 有权
    自适应投票专家用于在视频监控系统中预测的序列的增量分割

    公开(公告)号：US08295591B2

    公开(公告)日：2012-10-23

    申请号：US12543379

    申请日：2009-08-18

    申请人： Wesley Kenneth Cobb ,  Bobby Ernest Blythe ,  David Samuel Friedlander ,  Kishor Adinath Saitwal ,  Gang Xu

    发明人： Wesley Kenneth Cobb ,  Bobby Ernest Blythe ,  David Samuel Friedlander ,  Kishor Adinath Saitwal ,  Gang Xu

    IPC分类号： G06K9/00

    CPC分类号： G06K9/00771

    摘要： A sequence layer in a machine-learning engine configured to learn from the observations of a computer vision engine. In one embodiment, the machine-learning engine uses the voting experts to segment adaptive resonance theory (ART) network label sequences for different objects observed in a scene. The sequence layer may be configured to observe the ART label sequences and incrementally build, update, and trim, and reorganize an ngram trie for those label sequences. The sequence layer computes the entropies for the nodes in the ngram trie and determines a sliding window length and vote count parameters. Once determined, the sequence layer may segment newly observed sequences to estimate the primitive events observed in the scene as well as issue alerts for inter-sequence and intra-sequence anomalies.

    摘要翻译： 机器学习引擎中的序列层，被配置为从计算机视觉引擎的观察中学习。 在一个实施例中，机器学习引擎使用投票专家来分割在场景中观察到的不同对象的自适应共振理论（ART）网络标签序列。 序列层可以被配置为观察ART标签序列并且逐渐地构建，更新和修整和重组那些标记序列的ngram特里。 序列层计算ngram trie中节点的熵，并确定滑动窗口长度和投票计数参数。 一旦确定，序列层可以划分新观察到的序列以估计在场景中观察到的原始事件以及发出序列间和序列内异常的警报。