公开(公告)号：US09626849B2

公开(公告)日：2017-04-18

申请号：US14738816

申请日：2015-06-12

Applicant: GOOGLE INC.

Inventor： Michael Dixon ,  Dongeek Shin ,  George Alban Heitz, III ,  Srivatsan Varadharajan

IPC: H04N7/18 ,  G08B13/196 ,  H04N5/14 ,  G06K9/00

CPC classification number: G08B13/19606 ,  G06K9/00771 ,  G06K9/2036 ,  G08B13/19604 ,  G08B13/19608 ,  G08B13/1961 ,  H04N5/144 ,  H04N7/181

Abstract: A process reduces false positive security alerts. The process is performed at a computing device having one or more processors, and memory storing one or more programs configured for execution by the one or more processors. The process computes a depth map for a scene monitored by a video camera using a plurality of IR images captured by the video camera and uses the depth map to identify a first region within the scene having historically above average false positive detected motion events. In some instances, the first region is a ceiling, a window, or a television. The process monitors a video stream provided by the video camera to identify motion events, excluding the first region, and generates a motion alert when there is detected motion in the scene outside of the first region and the detected motion satisfies threshold criteria.

公开(公告)号：US09449229B1

公开(公告)日：2016-09-20

申请号：US14738034

申请日：2015-06-12

Applicant: GOOGLE INC.

Inventor： Jason N. Laska ,  Wei Hua ,  Rizwan Ahmed Chaudhry ,  Srivatsan Varadharajan ,  George Alban Heitz, III

IPC: G06K9/00 ,  G08B13/196 ,  H04N7/18 ,  G06K9/62 ,  G06T7/20 ,  H04N5/14

CPC classification number: G06K9/00718 ,  G06F3/0482 ,  G06F3/04847 ,  G06F3/04855 ,  G06F3/04886 ,  G06K9/00765 ,  G06K9/00771 ,  G06K9/6267 ,  G06K2009/00738 ,  G06T7/20 ,  G08B13/19604 ,  G08B13/19606 ,  G08B13/19608 ,  G08B13/19669 ,  G08B13/19684 ,  G11B27/3027 ,  G11B27/34 ,  H04N5/144 ,  H04N5/77 ,  H04N7/18 ,  H04N9/8205 ,  H04N21/4334 ,  H04N21/4335 ,  H04N21/44008 ,  H04N21/8133 ,  H04W4/80

Abstract: The various embodiments described herein include methods, devices, and systems for categorizing motion event candidates. In one aspect, a method includes receiving and processing video frames that include a motion event candidate. The processing includes: (a) obtaining background factors corresponding to a background in at least a subset of the video frames; (b) utilizing the background factors to identify one or more motion entities; (c) for each motion entity, obtaining one or more representative motion vectors based on a motion track of the respective motion entity; (d) identifying one or more features in at least a subset of the video frames; and (e) aggregating the background factors, the representative motion vectors, and the features to generate motion features. The method further includes sending the motion features to an event categorizer, where the event categorizer assigns a motion event category to the motion event candidate based on the received motion features.

Abstract translation: 本文描述的各种实施例包括用于对运动事件候选进行分类的方法，装置和系统。 一方面，一种方法包括接收和处理包括运动事件候选的视频帧。 该处理包括：（a）获得与视频帧的至少一个子集中的背景对应的背景因子; （b）利用背景因素来识别一个或多个运动实体; （c）对于每个运动实体，基于相应运动实体的运动轨迹获得一个或多个代表运动矢量; （d）识别视频帧的至少一个子集中的一个或多个特征; 和（e）聚集背景因子，代表运动矢量和特征以产生运动特征。 该方法还包括将运动特征发送到事件分类器，其中事件分类器基于所接收的运动特征向运动事件候选者分配运动事件类别。