公开(公告)号：EP3310044A1

公开(公告)日：2018-04-18

申请号：EP17178658.5

申请日：2017-06-29

申请人： Synology Incorporated

发明人： Hsu, Szu-Lu ,  Chiu, Yu-Hsiang ,  Lee, Szu-Hsien

IPC分类号： H04N7/18 ,  G06K9/00 ,  G07C9/00

CPC分类号： G06K9/00718 ,  G06K9/00771 ,  G06K9/00778 ,  G06K9/623 ,  G06K2009/00738 ,  G06T7/70 ,  G06T11/60 ,  G06T2207/10016 ,  G06T2207/20221 ,  G06T2207/30196 ,  G06T2207/30232 ,  G06T2207/30242 ,  G07C9/00 ,  H04N7/181

摘要： A method and an associated apparatus (105, 105-1) for performing cooperative counting are provided, where the method is applicable to a counter system (100, 100-1), and the counter system (100, 100-1) includes a plurality of cameras ({20-1, 20-2, ..., 20-N}, Cam(A), Cam(B), Cam(C), Cam(D)). The method includes: setting a plurality of points on an electronic map (300MAP, 1000MAP) as a plurality of predetermined points according to user inputs; determining at least one rule related to the predetermined points according to rule information, where the rule information is stored in the counter system (100, 100-1); respectively performing video object detection upon a plurality of images captured by the cameras ({20-1, 20-2, ..., 20-N}, Cam(A), Cam(B), Cam(C), Cam(D)) to generate detection results respectively corresponding to the cameras ({20-1, 20-2, ..., 20-N}, Cam(A), Cam(B), Cam(C), Cam(D)); and merging the detection results respectively corresponding to the cameras ({20-1, 20-2, ..., 20-N}, Cam(A), Cam(B), Cam(C), Cam(D)), to count events complying with the at least one rule.

摘要翻译： 提供了一种用于执行协作计数的方法和相关设备（105,105-1），其中该方法适用于计数器系统（100,100-1），并且计数器系统（100,100-1）包括 （{20-1,20-2，...，20-N}，凸轮（A），凸轮（B），凸轮（C），凸轮（D））。 该方法包括：根据用户输入将电子地图上的多个点（300MAP，1000MAP）设置为多个预定点; 根据规则信息确定与预定点相关的至少一个规则，其中规则信息存储在计数器系统（100,100-1）中; 分别对摄像机（{20-1,20-2，...，20-N}，Cam（A），Cam（B），Cam（C），Cam（ D））以产生分别对应于相机（{20-1,20-2，...，20-N}，Cam（A），Cam（B），Cam（C），Cam（D） ）; （{20-1,20-2，...，20-N}，Cam（A），Cam（B），Cam（C），Cam（D））的检测结果合并， 以计数符合至少一个规则的事件。

公开(公告)号：EP3297277A1

公开(公告)日：2018-03-21

申请号：EP17179461.3

申请日：2017-07-04

申请人： Accenture Global Solutions Limited

发明人： VERDEJO, Dominique ,  ZAMAN, Md Faisal ,  LECUE, Freddy

IPC分类号： H04N7/18

CPC分类号： G06K9/00711 ,  G06K9/00577 ,  G06K9/00771 ,  G06K9/4604 ,  G06K2009/00738 ,  H04L51/04 ,  H04N7/188 ,  H04W4/14

摘要： A device may receive, from another device, an image that depicts one or more events occurring at a location. The device may detect an event of the one or more events depicted in the image. The device may determine a first score that indicates a likelihood that the event is an anomalous event. The first score may be based on first data received from the other device. The device may determine a second score that indicates a similarity between the event and another event. The second score may be based on second data received from the other device. The second data may comprise data from a sensor, a wearable device, a mobile device, and/or a handheld device. The device may perform an action based on the first score or the second score.

摘要翻译： 设备可以从另一设备接收描绘在一个位置处发生的一个或多个事件的图像。 该设备可以检测图像中描绘的一个或多个事件的事件。 设备可以确定指示事件是异常事件的可能性的第一分数。 第一分数可以基于从另一个设备接收到的第一数据。 设备可以确定指示事件与另一事件之间的相似性的第二分数。 第二分数可以基于从另一设备接收的第二数据。 第二数据可以包括来自传感器，可穿戴设备，移动设备和/或手持设备的数据。 设备可以基于第一分数或第二分数来执行动作。

公开(公告)号：EP3263031A1

公开(公告)日：2018-01-03

申请号：EP17173680.4

申请日：2017-05-31

申请人： Fujitsu Limited

发明人： Okada, Yasutaka ,  Murashita, Kimitaka

IPC分类号： A61B5/11 ,  A61B5/00

CPC分类号： G08B21/22 ,  A61B5/1115 ,  A61B5/4806 ,  G06K9/00288 ,  G06K9/00342 ,  G06K9/00691 ,  G06K2009/00738

摘要： A situation identification method includes acquiring a plurality of images; identifying, for each of the plurality of images, a first area including a bed area where a place to sleep appears in an image, and a second area where an area in a predetermined range around the place to sleep appears in the image; detecting a state of a subject to be monitored for each of the plurality of images based on a result of detection of a head area indicating an area of a head of the subject in the first area and a result of detection of a living object in the second area; when the state of the subject changes from a first state to a second state, identifying a situation of the subject based on a combination of the first state and the second state; and outputting information that indicates the identified situation.

摘要翻译： 情况识别方法包括获取多个图像; 针对所述多个图像中的每一个识别第一区域和第二区域，所述第一区域包括睡眠位置出现在床上的区域，所述第二区域在所述图像中出现睡眠位置周围的预定范围内的区域; 基于表示所述第一区域中的所述被摄体的头部的区域的头部区域的检测结果和所述第一区域中的活体的检测结果来检测针对所述多个图像中的每一个的要被监视的对象的状态 第二区; 当所述对象的状态从第一状态改变到第二状态时，基于所述第一状态和所述第二状态的组合来识别所述对象的状况; 并输出指示所识别的情况的信息。

公开(公告)号：EP3241152A1

公开(公告)日：2017-11-08

申请号：EP15876310.2

申请日：2015-12-30

申请人： Morphotrust Usa, LLC

发明人： MARTIN, Brian

IPC分类号： G06K9/00

CPC分类号： G11B27/005 ,  G06K9/00771 ,  G06K2009/00738 ,  G08B13/19608 ,  G08B13/19615 ,  G08B13/19645 ,  H04N5/77 ,  H04N5/783 ,  H04N7/181

摘要： Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving a video feed of a scene that includes an object in at least a portion of the scene. Tracking the object using an object tracking algorithm. Detecting a change in the object from a first frame of the video feed to a second frame of the video feed. Automatically causing an analysis to be performed on a portion of the video feed that includes the object and the change in the object in response to detecting the change. Other implementations of this aspect include corresponding systems, apparatus, and computer programs, configured to perform the actions of the methods, encoded on computer storage devices.

公开(公告)号：EP3200444A4

公开(公告)日：2017-10-25

申请号：EP15803979

申请日：2015-03-02

申请人： ZTE CORP

发明人： CHEN JIANJIANG ,  XU YUANYUAN

IPC分类号： H04N5/232 ,  G06T7/194 ,  G06T11/60 ,  H04N5/272

CPC分类号： G06K9/00744 ,  G06K2009/00738 ,  G06T7/194 ,  G06T11/60 ,  G06T2207/10016 ,  H04N5/232 ,  H04N5/23222 ,  H04N5/23229 ,  H04N5/272

摘要： Disclosed are a method, system and apparatus for processing video shooting. The method includes: shooting a subject in a background sample, the background sample and a target background respectively, generating a first video recording the subject and the background sample, a second video recording the background sample and a third video recording the target background respectively, herein shooting durations of the first video and the second video are less than a shooting duration of the third video; and comparing the first video with the second video and extracting a subject image from the first video to generate a subject frame image sequence of which a background is transparent, and superimposing in turn each subject frame image in the subject image frame sequence onto the third video to generate a video file recording the subject and the target background.

公开(公告)号：EP3186960A1

公开(公告)日：2017-07-05

申请号：EP15825333.6

申请日：2015-07-22

申请人： GoPro, Inc.

发明人： NEWMAN, David A. ,  SILVER, Adam ,  HODULIK, Nick ,  TAYLOR, Jonathan ,  SCHUNCK, Brian G. ,  ADUMILLI, Balineedu Chowdary ,  LUSTIG, Ryan

IPC分类号： H04N9/80

CPC分类号： G06K9/00751 ,  G06K2009/00328 ,  G06K2009/00738 ,  G06K2009/00939 ,  G06K2209/27 ,  G11B27/002 ,  G11B27/031 ,  G11B27/22 ,  G11B27/34 ,  H04N5/232 ,  H04N5/23216 ,  H04N5/23293 ,  H04N5/77 ,  H04N5/772 ,  H04N5/91 ,  H04N9/8205 ,  H04N13/172 ,  H04N13/178 ,  H04N21/2353 ,  H04N21/8549 ,  H04N2201/3228 ,  H04N2201/3229 ,  H04N2201/3256

摘要： Video and corresponding metadata is accessed. Events of interest within the video are identified based on the corresponding metadata, and best scenes are identified based on the identified events of interest. A video summary can be generated including one or more of the identified best scenes. The video summary can be generated using a video summary template with slots corresponding to video clips selected from among sets of candidate video clips. Best scenes can also be identified by receiving an indication of an event of interest within video from a user during the capture of the video. Metadata patterns representing activities identified within video clips can be identified within other videos, which can subsequently be associated with the identified activities.

摘要翻译： 视频和相应的元数据被访问。 基于相应的元数据来识别视频内的感兴趣事件，并基于所识别的感兴趣事件来识别最佳场景。 可以生成包括一个或多个所识别的最佳场景的视频摘要。 可以使用具有与从候选视频剪辑集合中选择的视频剪辑相对应的时隙的视频概要模板来生成视频概要。 通过在捕捉视频期间接收来自用户的视频内的感兴趣事件的指示，也可以识别最佳场景。 表示在视频剪辑内识别的活动的元数据模式可以在其他视频中识别，其随后可以与所识别的活动相关联。

公开(公告)号：EP3121747A1

公开(公告)日：2017-01-25

申请号：EP16175643.2

申请日：2016-06-22

申请人： Konica Minolta, Inc.

发明人： Miyake, Nobuyuki ,  Muraoka, Shintaro ,  Haraguchi, Tsuyoshi ,  Hosoki, Tetsu

IPC分类号： G06F19/00 ,  G06K9/00 ,  G06T7/00

CPC分类号： G11B27/34 ,  A61B6/461 ,  G06F19/00 ,  G06K9/00496 ,  G06K9/00751 ,  G06K9/6212 ,  G06K9/6269 ,  G06K2009/00738 ,  G06K2209/05 ,  G06T7/0012 ,  G06T2207/10116 ,  G06T2207/30061 ,  G16H50/20 ,  H04N9/87

摘要： To provide a console (2, 3) capable of automatically finding the important spot in a dynamic image.
The console includes a display means (24, 34) which plays back a dynamic image formed of a plurality of frame images to display according to operation of an operator; and a learning means (21, 31) which, when detecting that operation of possibly expressing interest is performed by the operator during playback of the dynamic image, performs statistical processing on a pattern of change in time dv(i)/dt of a value v(i) of an index regarding a dynamic state of a site to be examined and/or change in time of the value v(i) of the index in a frame image displayed on the display means (24, 34) at that time to learn appearance frequency of the value v(i) of the index and/or the pattern of the change in time dv(i)/dt of the value v(i) of the index.

摘要翻译： 提供一个能够自动找到动态图像中重要位置的控制台（2,3）。 控制台包括显示装置（24,34），其播放由多个帧图像形成的动态图像，以根据操作者的操作显示; 以及学习装置（21,31），当检测到在动态图像的重放期间由操作者执行可能表示兴趣的操作时，对值的变化模式（dv（i）/ dt）进行统计处理 v（i）关于在该时间显示在显示装置（24,34）上的帧图像中的要被检查的场所的动态状态的指标和/或时间上的变化值v（i） 学习索引的值v（i）的出现频率和/或索引值v（i）的时间变化dv（i）/ dt的模式。

公开(公告)号：EP3098755A1

公开(公告)日：2016-11-30

申请号：EP16171900.0

申请日：2016-05-30

申请人： Accenture Global Services Limited

发明人： BATALLER, Cyrille ,  ASTROM, Anders ,  SCHIOPU, Vitalie ,  KHALAFI, Hakim

IPC分类号： G06K9/00

CPC分类号： G06K9/00765 ,  G06F17/3028 ,  G06K9/00255 ,  G06K9/00295 ,  G06K9/00718 ,  G06K9/00744 ,  G06K9/00771 ,  G06K9/00778 ,  G06K9/6202 ,  G06K9/6263 ,  G06K2009/00738 ,  G06T7/0002 ,  G06T7/13 ,  G06T7/292 ,  G06T7/75 ,  G06T7/80 ,  G06T15/205 ,  G06T19/006 ,  G06T2207/10016 ,  G06T2207/20061 ,  G06T2207/30168 ,  G06T2207/30201 ,  G06T2207/30232 ,  G06T2207/30242 ,  G08B13/19602 ,  G08B13/19608 ,  H04N7/18 ,  H04N7/181

摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying people depicted in images. In one aspect, a process includes receiving an image from a camera. A face of a person is detected in the image. The image is compared to a set of images stored in a local cache. A determination is made whether the face of the person matches a face of a person depicted by at least one image of the set of images. In response to determining that the face of the person does not match a face of a person depicted by at least one image of the set of images, a selection is made of a highest quality image of the face of the person. A server system compares the highest quality image to images from data for the multiple people to identify the person.

摘要翻译： 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于识别图像中描绘的人。 在一个方面，一个过程包括从相机接收图像。 在图像中检测到人脸。 将图像与存储在本地高速缓存中的一组图像进行比较。 确定人的面部是否与由该组图像中的至少一个图像描绘的人的脸匹配。 响应于确定人的面部与图像集合中的至少一个图像所描绘的人的脸部不匹配，选择人的面部的最高质量的图像。 服务器系统将最高质量的图像与来自多个人的数据的图像进行比较，以识别该人。

公开(公告)号：EP3012813A3

公开(公告)日：2016-10-12

申请号：EP15180790.6

申请日：2015-08-12

申请人： The Boeing Company

发明人： AUSTIN, Thomas Robert

IPC分类号： G08B21/04 ,  G06K9/00 ,  G08B29/18 ,  G08B21/10 ,  G08B21/12

CPC分类号： G08B13/19602 ,  G06F3/16 ,  G06K9/00288 ,  G06K9/00348 ,  G06K9/00771 ,  G06K9/00778 ,  G06K9/2018 ,  G06K2009/00738 ,  G06T2207/30108 ,  G08B21/0423 ,  G08B21/0476 ,  G08B21/10 ,  G08B21/12 ,  G08B29/188 ,  H04N5/33 ,  H04N5/772 ,  H04N7/181

摘要： A method and apparatus for detecting an occurrence of an event of interest. The apparatus comprises a surveillance system, a detector, and an analyzer. The surveillance system monitors subjects within an environment to generate monitoring data. The detector detects a number of indicator instances exhibited by at least a portion of the subjects using the monitoring data. Each of the number of indicator instances is an instance of a corresponding event indicator in a set of pre-selected event indicators. The analyzer evaluates the number of indicator instances and additional information to determine whether at least a portion of the number of indicator instances meets an event detection threshold, thereby indicating a detection of an occurrence of an event of interest.

摘要翻译： 一种用于检测感兴趣事件的发生的方法和装置。 该装置包括监视系统，检测器和分析器。 监视系统监视环境中的对象以生成监视数据。 检测器使用监测数据检测至少一部分受试者展示的指示符实例的数量。 多个指示符实例中的每一个是一组预先选择的事件指示符中的相应事件指示符的实例。 分析器评估指示符实例的数量和附加信息，以确定指示符实例的数量的至少一部分是否满足事件检测阈值，从而指示检测到感兴趣的事件的发生。

公开(公告)号：EP3026613A1

公开(公告)日：2016-06-01

申请号：EP15194477.4

申请日：2015-11-13

申请人： Yokogawa Electric Corporation

发明人： Miyata, Yuuji ,  Sakuraba, Yuichi

IPC分类号： G06Q10/06 ,  G06F17/10 ,  G06F11/00

CPC分类号： G06N7/005 ,  G06F11/0706 ,  G06F11/079 ,  G06K9/00 ,  G06K9/00677 ,  G06K9/6278 ,  G06K2009/00738 ,  G06N5/047 ,  G06Q10/063 ,  G06Q10/067

摘要： An associated event group preparation unit calculates a degree of association between the events on the basis of an event matrix, and prepares an associated event group for each event. A cause-and-effect relationship model establishment unit establishes a probabilistic cause-and-effect relationship model by a Bayesian network on the basis of the event matrix, for each associated event group. An improvement candidate pattern receiving unit receives a setting of an improvement candidate pattern in which a condition of an event to be set as an improvement candidate is determined by attributes of the base point event and the associated event and a conditional probability between the base point event and the associated event. A pattern analysis unit extracts a probabilistic cause-and-effect relationship model conforming to any one of the set improvement candidate patterns, from the established probabilistic cause-and-effect relationship models for each event.

摘要翻译： 相关联的事件组准备单元基于事件矩阵来计算事件之间的关联程度，并为每个事件准备相关联的事件组。 因果关系模型建立单元对于每个相关联的事件组，基于事件矩阵，由贝叶斯网络建立概率因果关系模型。 改进候选模式接收单元接收改变候选模式的设置，其中将要设置为改进候选的事件的条件由基点事件和关联事件的属性确定，并且基点事件之​​间的条件概率 和相关事件。 模式分析单元从针对每个事件的已建立的概率因果关系模型中提取符合设定的改进候选模式中的任何一个的概率因果关系模型。