公开(公告)号：US20120163670A1

公开(公告)日：2012-06-28

申请号：US13413549

申请日：2012-03-06

申请人： John Eric EATON ,  Wesley Kenneth COBB ,  Dennis Gene URECH ,  Bobby Ernest BLYTHE ,  David Samuel FRIEDLANDER ,  Rajkiran Kumar GOTTUMUKKAL ,  Lon William RISINGER ,  Kishor Adinath SAITWAL ,  Ming-Jung SEOW ,  David Marvin SOLUM ,  Gang XU ,  Tao YANG

发明人： John Eric EATON ,  Wesley Kenneth COBB ,  Dennis Gene URECH ,  Bobby Ernest BLYTHE ,  David Samuel FRIEDLANDER ,  Rajkiran Kumar GOTTUMUKKAL ,  Lon William RISINGER ,  Kishor Adinath SAITWAL ,  Ming-Jung SEOW ,  David Marvin SOLUM ,  Gang XU ,  Tao YANG

IPC分类号： G06K9/00

CPC分类号： G06K9/00771 ,  G08B13/19608 ,  G08B13/19613

摘要： Embodiments of the present invention provide a method and a system for analyzing and learning behavior based on an acquired stream of video frames. Objects depicted in the stream are determined based on an analysis of the video frames. Each object may have a corresponding search model used to track an object's motion frame-to-frame. Classes of the objects are determined and semantic representations of the objects are generated. The semantic representations are used to determine objects' behaviors and to learn about behaviors occurring in an environment depicted by the acquired video streams. This way, the system learns rapidly and in real-time normal and abnormal behaviors for any environment by analyzing movements or activities or absence of such in the environment and identifies and predicts abnormal and suspicious behavior based on what has been learned.

摘要翻译： 本发明的实施例提供了一种基于所获取的视频帧流来分析和学习行为的方法和系统。 基于对视频帧的分析来确定流中描绘的对象。 每个对象可以具有用于跟踪对象的帧到帧的运动的对应的搜索模型。 确定对象的类，并生成对象的语义表示。 语义表示用于确定对象的行为，并了解在所获取的视频流中描绘的环境中发生的行为。 通过这种方式，系统通过分析环境中的动作或活动或不存在这些环境，快速，实时地对任何环境的正常和异常行为进行学习，并根据所学到的知识和预测异常和可疑行为。

公开(公告)号：US20120275649A1

公开(公告)日：2012-11-01

申请号：US13545950

申请日：2012-07-10

申请人： Wesley Kenneth COBB ,  Ming-Jung SEOW ,  Tao YANG

发明人： Wesley Kenneth COBB ,  Ming-Jung SEOW ,  Tao YANG

IPC分类号： G06K9/62

CPC分类号： G06K9/00771 ,  G06T7/215 ,  G06T7/254 ,  G06T2207/10016 ,  G06T2207/30196 ,  G06T2207/30236 ,  G06T2207/30241 ,  G08B13/19608

摘要： Techniques are disclosed for detecting foreground objects in a scene captured by a surveillance system and tracking the detected foreground objects from frame to frame in real time. A motion flow field is used to validate foreground objects(s) that are extracted from the background model of a scene. Spurious foreground objects are filtered before the foreground objects are provided to the tracking stage. The motion flow field is also used by the tracking stage to improve the performance of the tracking as needed for real time surveillance applications.

摘要翻译： 公开了用于检测由监视系统捕获的场景中的前景物体并且实时地从帧到帧跟踪检测到的前景物体的技术。 运动流场用于验证从场景的背景模型中提取的前景对象。 在将前景对象提供给跟踪阶段之前，会对前景对象进行过滤。 跟踪阶段还使用运动流场来提高实时监控应用所需的跟踪性能。

公开(公告)号：US20120224746A1

公开(公告)日：2012-09-06

申请号：US13472214

申请日：2012-05-15

申请人： Wesley Kenneth COBB ,  David FRIEDLANDER ,  Kishor Adinath SAITWAL ,  Ming-Jung SEOW ,  Gang XU

发明人： Wesley Kenneth COBB ,  David FRIEDLANDER ,  Kishor Adinath SAITWAL ,  Ming-Jung SEOW ,  Gang XU

IPC分类号： G06K9/62

CPC分类号： G06K9/00771

摘要： Techniques are disclosed for a video surveillance system to learn to recognize complex behaviors by analyzing pixel data using alternating layers of clustering and sequencing. A combination of a self organizing map (SOM) and an adaptive resonance theory (ART) network may be used to identify a variety of different anomalous inputs at each cluster layer. As progressively higher layers of the cortex model component represent progressively higher levels of abstraction, anomalies occurring in the higher levels of the cortex model represent observations of behavioral anomalies corresponding to progressively complex patterns of behavior.

摘要翻译： 公开了用于视频监控系统的技术，以通过使用交替的聚类和排序层分析像素数据来学习识别复杂行为。 可以使用自组织图（SOM）和自适应共振理论（ART）网络的组合来识别每个簇层处的各种不同的异常输入。 随着逐渐更高层的皮层模型组件代表逐渐提高的抽象水平，在较高级别的皮质模型中发生的异常表示对应于逐渐复杂的行为模式的行为异常的观察。

公开(公告)号：US20170295070A1

公开(公告)日：2017-10-12

申请号：US15481320

申请日：2017-04-06

申请人： Tao YANG ,  Ming-Jung SEOW

发明人： Tao YANG ,  Ming-Jung SEOW

IPC分类号： H04L12/24 ,  G06N99/00 ,  H04L29/08

CPC分类号： H04L41/142 ,  G06N20/00 ,  H04L41/12 ,  H04L41/16 ,  H04L43/04 ,  H04L67/10

摘要： Techniques are disclosed for processing data collected from network components for analysis by a machine learning engine of a Cognitive AI System. A network data processing driver receives a stream of data from a data collector which obtains data from one or more network data sources. The driver normalizes the stream of data to one or more feature values each corresponding to the network data sources and generates a sample vector from the feature values. The sample vector is formatted to be analyzed by the machine learning engine.

公开(公告)号：US20170293606A1

公开(公告)日：2017-10-12

申请号：US15481298

申请日：2017-04-06

申请人： Gang XU ,  Tao YANG ,  Ming-Jung SEOW

发明人： Gang XU ,  Tao YANG ,  Ming-Jung SEOW

IPC分类号： G06F17/27 ,  G06N3/08 ,  G06N3/04

CPC分类号： G06F17/2735 ,  G06F17/271 ,  G06F17/277 ,  G06F17/2785 ,  G06N3/04 ,  G06N3/0409 ,  G06N3/0454 ,  G06N3/08 ,  G06N3/088 ,  G06N5/04 ,  G06N5/048 ,  G06N20/00

摘要： Techniques are disclosed to optimize feature selection in generating betas for a feature dictionary of a neuro-linguistic Cognitive AI System. A machine learning engine receives a sample vector of input data to be analyzed by the neuro-linguistic Cognitive AI System. The neuro-linguistic Cognitive AI System is configured to generate multiple feature words for each of a plurality of sensors. The machine learning engine identifies a sensor specified in the sample vector and selects optimization parameters for generating feature words based on the identified sensor.

公开(公告)号：US20170293609A1

公开(公告)日：2017-10-12

申请号：US15481302

申请日：2017-04-06

申请人： Tao YANG ,  Gang XU ,  Wesley Kenneth COBB ,  Omni AI, Inc.

发明人： Ming-Jung SEOW ,  Tao YANG ,  Gang XU ,  Wesley Kenneth COBB

IPC分类号： G06F17/27 ,  G06N99/00 ,  G06N5/04

CPC分类号： G06F17/2735 ,  G06F17/271 ,  G06F17/277 ,  G06F17/2785 ,  G06N3/04 ,  G06N3/0409 ,  G06N3/0454 ,  G06N3/08 ,  G06N3/088 ,  G06N5/04 ,  G06N5/048 ,  G06N20/00

摘要： Techniques are disclosed for analyzing and learning behaviors based on acquired sensor data. A neuro-linguistic cognitive engine performs learning and analysis on linguistic content (e.g., identified alpha symbols, betas, and gammas) obtained by a linguistic model that clusters observations to generate the linguistic content. The neuro-linguistic cognitive engine compares new data to learned patterns stored in short and longer-term memories and determines whether to issue special event notifications indicating anomalous behavior. In one embodiment, condition(s) may be generated for new data and checked against inference nodes of an inference network. Inference nodes matching the condition(s) are executed to, e.g., compare the new data with the learned patterns, with output from the inference nodes being used to generate additional condition(s) that are again matched to inference nodes which may be executed.