WO2018112722A1 - 一种视频巡检方法及其系统 - Google Patents
一种视频巡检方法及其系统 Download PDFInfo
- Publication number
- WO2018112722A1 WO2018112722A1 PCT/CN2016/110938 CN2016110938W WO2018112722A1 WO 2018112722 A1 WO2018112722 A1 WO 2018112722A1 CN 2016110938 W CN2016110938 W CN 2016110938W WO 2018112722 A1 WO2018112722 A1 WO 2018112722A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- camera
- patrol
- video
- intelligent analysis
- list
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Definitions
- the invention relates to the technical field of intelligent analysis technology and video monitoring system, in particular to a video inspection method and system thereof.
- Video patrol is an important component of the video surveillance system technology. It stipulates a set of camera lists, and in turn, plays real-time video in a specified video window according to the specified patrol interval, and cooperates with the manual guard to the video screen. The exception is handled.
- Intelligent analysis is a computer image visual analysis technology that analyzes and tracks the objects appearing in the camera scene by separating the background and the target in the scene.
- the user can preset different alarm rules in different camera scenes, and once the target appears in the scene, the behavior of violating the predefined rules is formed, and a corresponding intelligent analysis event is formed.
- the main intelligent analysis events perimeter prevention, population statistics, item reservation, item theft, crowd gathering, crowd dispersal, sputum, traffic incident detection, fireworks detection, crowd density, fights, etc.
- the video patrol is created by manually selecting the camera to be patrolled, setting the patrol interval, and then specifying the video window. Finally, setting the start condition of the video patrol. When the start condition is met, the video is automatically started. Inspection mission.
- the creation method has the following problems:
- the camera of the inspection is manually selected and has certain subjectivity. For example, a camera with a sparsely populated video screen is selected, and inspection of these cameras wastes part of the system resources.
- the inspection interval cannot be automatically adjusted according to the density of people in the video screen.
- the inspection intervals of all cameras are the same. As described in the problem 1, the inspection time interval between the dense and the sparse camera is the same, and part of the system resources is wasted.
- the main purpose of the present invention is to provide a video inspection method and system thereof, which aims to make full use of system resources and better realize video inspection.
- the present invention provides a video inspection method, wherein the method includes the steps of:
- the number of intelligent analysis events corresponding to the camera is obtained, and the inspection camera list is given accordingly.
- the method further includes the steps of: generating a corresponding camera tag for the camera; the camera tag includes the number of intelligent analysis events, the browsing click rate, and the location At least one of them.
- the method further includes the step of: according to the patrol camera list
- the camera label corresponding to the camera in the camera automatically sets different patrol scenes for selection; the patrol scene includes:
- the camera in the camera tag that browses the click rate exceeding the second preset value constitutes a second patrol scene
- All the cameras in the patrol camera list constitute a fourth patrol scene.
- the method further includes the following steps: acquiring the location The historical intelligent analysis event corresponding to the camera in the patrol camera list is described, and the patrol time interval of each round of the camera is dynamically changed accordingly.
- the obtaining the historical intelligent analysis event corresponding to the camera in the patrol camera list, and dynamically changing the patrol time interval of the camera for each round of patrol according to the method includes:
- N cn a total number of intelligent analysis events corresponding to the camera C n in a certain period of the previous T n days, forming an array with the time period [T 1 , T 2 , . . . T n ] the number of arrays corresponding to a list of intelligent analysis of events [N c1, N c2, ... N cn]; determining the maximum value and the minimum value N cmax N cmin.
- the video inspection task switches the camera C n , and the camera C n will play the T cnow duration.
- the present invention further provides a video inspection system, including:
- a configuration unit configured to receive a configuration of an intelligent analysis task for the camera
- An intelligent analysis unit configured to acquire the real-time video stream of the camera, perform video analysis according to the intelligent analysis task, and detect an intelligent analysis event;
- the list output unit is configured to acquire the number of intelligent analysis events corresponding to the camera and receive a list of the patrol cameras according to the data patrol task.
- the method further includes a label generating unit, configured to generate a corresponding camera label for the camera; the camera label includes at least one of a number of intelligent analysis events, a browsing click rate, and a location.
- a label generating unit configured to generate a corresponding camera label for the camera; the camera label includes at least one of a number of intelligent analysis events, a browsing click rate, and a location.
- the method further includes a patrol scene generating unit, configured to be used according to the patrol camera list
- the camera label corresponding to the camera automatically sets different patrol scenes for selection;
- the patrol scene includes:
- the camera in the camera tag that browses the click rate exceeding the second preset value constitutes a second patrol scene
- All the cameras in the patrol camera list constitute a fourth patrol scene.
- the patrol time calculation unit is configured to acquire a historical intelligent analysis event corresponding to the camera in the patrol camera list, and dynamically change the patrol time interval of each round of the camera according to the patrol time interval.
- the patrol time calculation unit is specifically configured to:
- N cn a total number of intelligent analysis events corresponding to the camera C n in a certain period of the previous T n days, forming an array with the time period [T 1 , T 2 , . . . T n ] the number of arrays corresponding to a list of intelligent analysis of events [N c1, Nc 2, ... N cn]; determining the maximum value and the minimum value N cmax N cmin.
- the video inspection task switches the camera C n , and the camera C n will play the T cnow duration.
- the invention provides a video inspection method and system thereof, and analyzes events through historical intelligence Statistics, when the new video inspection task is realized, the camera list is automatically recommended, which reduces the difference and irrationality of the video inspection configuration due to the uneven artificial experience, and improves the reasonable applicability of the video monitoring system resources.
- FIG. 1 is a schematic structural diagram of hardware of a video inspection system for implementing various embodiments of the present invention
- FIG. 2 is a schematic flowchart of a video inspection method according to a first embodiment of the present invention
- FIG. 3 is a schematic flowchart of a video inspection method according to a second embodiment of the present invention.
- FIG. 4 is a schematic flowchart of a video inspection method according to a third embodiment of the present invention.
- FIG. 5 is a schematic flowchart diagram of a video inspection method according to a fourth embodiment of the present invention.
- FIG. 6 is a schematic structural diagram of a video inspection system according to a sixth embodiment of the present invention.
- FIG. 7 is a schematic structural diagram of a video inspection system according to a seventh embodiment of the present invention.
- FIG. 8 is a schematic structural diagram of a video inspection system according to an eighth embodiment of the present invention.
- FIG. 9 is a schematic structural diagram of a video inspection system according to a ninth embodiment of the present invention.
- the hardware structure of the video inspection system provided by the present invention includes a video client 10, a central server 20, an intelligent analysis server 30, a database 40, a forwarding server 50, and a camera 70.
- the central server 20 is the core of the video inspection system and is used for processing various service functions such as device management and intelligent analysis event data storage.
- the intelligent analysis server 30 can perform intelligent analysis on the real-time video of the camera according to certain intelligent analysis rules, and report the intelligent analysis event to the central server 20, and the central server 20 reports the video to the video client 10.
- the video client 10 is a video patrol system service client, and has a business function of viewing real-time video of the camera, receiving intelligent analysis events, and video inspection.
- the forwarding server 50 is a multimedia forwarding server that is responsible for distributing the video stream of the front-end camera.
- the database 40 can store business data of the video surveillance system, such as device information, intelligent analysis events, and the like.
- a first embodiment of the present invention provides a video inspection method, including the following steps:
- the user can complete the configuration of the intelligent analysis task of the camera through the video client according to the scene illuminated by the camera and the intelligent analysis rule; the configured intelligent analysis task can be saved to the database through the central server;
- S12 Acquire the real-time video stream of the camera, perform video analysis according to the intelligent analysis task, and detect an intelligent analysis event;
- the intelligent analysis server reads the intelligent analysis task from the central server
- the real-time video stream of the camera is obtained by the forwarding server, and the video analysis is performed.
- the report is reported to the central server, and the central server saves the intelligent analysis event to The database notifies the video client at the same time.
- the above intelligent analysis events may include: perimeter prevention, population statistics, item reservation, item theft, crowd gathering, crowd dispersal, sputum, traffic incident detection, pyrotechnic detection, crowd density, fight, etc.;
- the video client when the video client receives the video patrol task configured by the user, the video client can pass the medium
- the heart server obtains and sorts the number of intelligent analysis events of all cameras from the database. According to the sorting result, a camera with a relatively high frequency of intelligent analysis events is added to the video inspection queue, and the camera that is not online is excluded.
- a second embodiment of the present invention provides a video inspection method, including the following steps:
- the user can complete the configuration of the intelligent analysis task of the camera through the video client according to the scene illuminated by the camera and the intelligent analysis rule; the configured intelligent analysis task can be saved to the database through the central server;
- the camera tag includes at least one of a number of intelligent analysis events, a browsing click rate, and a location;
- the intelligent analysis server reads the intelligent analysis task from the central server
- the real-time video stream of the camera is obtained by the forwarding server, and the video analysis is performed.
- the report is reported to the central server, and the central server saves the intelligent analysis event to The database notifies the video client at the same time.
- the above intelligent analysis events may include: perimeter prevention, population statistics, item reservation, item theft, crowd gathering, crowd dispersal, sputum, traffic incident detection, pyrotechnic detection, crowd density, fight, etc.;
- the number of intelligent analysis events of all the cameras can be obtained from the database through the central server and sorted, and according to the sorting result, a camera with a high frequency of intelligent analysis events will appear. Join the video inspection queue and exclude cameras that are not online.
- a third embodiment of the present invention provides a video inspection method, including the following steps:
- the user can complete the configuration of the intelligent analysis task of the camera through the video client according to the scene illuminated by the camera and the intelligent analysis rule; the configured intelligent analysis task can be saved to the database through the central server;
- the camera tag includes at least one of a number of intelligent analysis events, a browsing click rate, and a location;
- the intelligent analysis server reads the intelligent analysis task from the central server
- the real-time video stream of the camera is obtained by the forwarding server, and the video analysis is performed.
- the report is reported to the central server, and the central server saves the intelligent analysis event to The database notifies the video client at the same time.
- the above intelligent analysis events may include: perimeter prevention, population statistics, item reservation, item theft, crowd gathering, crowd dispersal, sputum, traffic incident detection, pyrotechnic detection, crowd density, fight, etc.;
- the number of intelligent analysis events of all the cameras can be obtained from the database through the central server and sorted, and according to the sorting result, a camera with a high frequency of intelligent analysis events will appear. Join the video inspection queue and exclude cameras that are not online.
- the patrol scene includes:
- the camera in the camera tag that browses the click rate exceeding the second preset value constitutes a second patrol scene
- All the cameras in the patrol camera list constitute a fourth patrol scene
- the at least one camera in the patrol camera list may also constitute a fifth patrol scene
- the user can select one of the above-mentioned inspection scenarios to execute.
- a fourth embodiment of the present invention provides a video inspection method, including the following steps:
- the user can complete the configuration of the intelligent analysis task of the camera through the video client according to the scene illuminated by the camera and the intelligent analysis rule; the configured intelligent analysis task can be saved to the database through the central server;
- the intelligent analysis server reads the intelligent analysis task from the central server
- the real-time video stream of the camera is obtained by the forwarding server, and the video analysis is performed.
- the report is reported to the central server, and the central server saves the intelligent analysis event to The database notifies the video client at the same time.
- the above intelligent analysis events may include: perimeter prevention, population statistics, item reservation, item theft, crowd gathering, crowd dispersal, sputum, traffic incident detection, pyrotechnic detection, crowd density, fight, etc.;
- the number of intelligent analysis events of all the cameras can be obtained from the database through the central server and sorted, and according to the sorting result, a camera with a high frequency of intelligent analysis events will appear. Join the video inspection queue while arranging Except for cameras that are not online.
- the method further includes the steps of: generating a corresponding camera tag for the camera; the camera tag includes at least one of a number of intelligent analysis events, a browsing click rate, and a location;
- the method further includes the steps of: automatically setting different patrol scenes for selection according to the camera label corresponding to the camera in the patrol camera list; the patrol scene includes: the camera label The camera that the number of the smart analysis events exceeds the first preset value constitutes a first patrol scene; the camera in the camera tag that browses the click rate exceeds the second preset value constitutes a second patrol scene; the camera label is The camera in the middle location is a third patrol scene; all the cameras in the patrol camera list constitute a fourth patrol scene; and at least one camera may be customized in the patrol camera list.
- the fifth patrol scenario the user can select one of the patrol scenarios to perform.
- step S44 only the historical intelligent analysis event corresponding to the camera in the selected patrol scene is acquired, and the patrol time
- a fifth embodiment of the present invention provides a video patrol method, which includes the same steps as S41 to S44 mentioned in the fourth embodiment, and is not described herein.
- step S44 includes:
- N cn a total number of intelligent analysis events corresponding to the camera C n in a certain period of the previous T n days, forming an array with the time period [T 1 , T 2 , . . . T n ] the number of arrays corresponding to a list of intelligent analysis of events [N c1, N c2, ... N cn]; determining the maximum value and the minimum value N cmax N cmin.
- the video inspection task switches the camera C n , and the camera C n will play the T cnow duration.
- the method further includes the steps of: generating a corresponding camera tag for the camera; the camera tag includes at least one of a number of intelligent analysis events, a browsing click rate, and a location;
- the method further includes the steps of: automatically setting different patrol scenes for selection according to the camera label corresponding to the camera in the patrol camera list; the patrol scene includes: the camera label The camera that the number of the smart analysis events exceeds the first preset value constitutes a first patrol scene; the camera in the camera tag that browses the click rate exceeds the second preset value constitutes a second patrol scene; the camera label is The camera in the middle location is a third patrol scene; all the cameras in the patrol camera list constitute a fourth patrol scene; and at least one camera may be customized in the patrol camera list.
- the fifth patrol scenario the user can select one of the patrol scenarios to perform.
- step S44 only the historical intelligent analysis event corresponding to the camera in the selected patrol scene is acquired, and the patrol time
- the video patrol method in the embodiment of the present invention is described above.
- the video patrol system in the embodiment of the present invention is described below.
- a sixth embodiment of the present invention provides a video patrol system, including a configuration unit 11, an intelligent analysis unit 21, and a list output unit 12.
- the configuration unit 11 is configured to receive a configuration of an intelligent analysis task for the camera; in a specific implementation, the configuration unit 11 may be configured in the video client. The user can follow the scene illuminated by the camera And the intelligent analysis rule completes the configuration of the intelligent analysis task of the camera through the video client; the configured intelligent analysis task can be saved to the database through the central server;
- the intelligent analysis unit 21 is configured to acquire the camera real-time video stream and perform video analysis according to the intelligent analysis task to detect an intelligent analysis event; more specifically, the intelligent analysis unit 21 is a unit of the intelligent analysis server, and the intelligent analysis server is from the central server. After reading the intelligent analysis task, the real-time video stream of the camera is obtained by the forwarding server, and the video analysis is performed. When the intelligent analysis event is detected, the event is reported to the central server, and the central server saves the intelligent analysis event to the database and notifies the video client.
- the above intelligent analysis events may include: perimeter prevention, population statistics, item reservation, item theft, crowd gathering, crowd dispersal, sputum, traffic incident detection, pyrotechnic detection, crowd density, fight, etc.;
- the list output unit 12 is configured to acquire the number of intelligent analysis events corresponding to the camera and receive a list of the patrol cameras according to the video patrol task.
- the list output unit 12 can be set on the video client. More specifically, when the video client receives the video patrol task configured by the user, the number of intelligent analysis events of all the cameras can be obtained from the database through the central server and sorted according to the order. As a result, a camera with a high frequency of intelligent analysis events will be added to the video inspection queue, and the camera that is not online will be excluded.
- a seventh embodiment of the present invention provides a video inspection system, including a configuration unit 11, an intelligent analysis unit 21, a list output unit 12, and a label generation unit 13.
- the configuration unit 11, the smart analysis unit 21, and the list output unit 12 in this embodiment are the same as the configuration unit 11, the smart analysis unit 21, and the list output unit 12 in the sixth embodiment described above, as described above, and are no longer here. Narration.
- the embodiment further includes a label generating unit 13 configured to generate a corresponding camera label for the camera; the camera label includes at least one of a number of intelligent analysis events, a browsing click rate, and a location.
- the above label generating unit 13 can be provided on the video client.
- an eighth embodiment of the present invention provides a video patrol system, including a configuration unit 11, an intelligent analysis unit 21, a list output unit 12, a label generation unit 13, and a patrol scene generation unit 14.
- the configuration unit 11, the smart analysis unit 21, the list output unit 12, and the tag generation unit 13 in this embodiment are the same as the configuration unit 11, the smart analysis unit 21, the list output unit 12, and the tag generation unit 13 in the seventh embodiment described above. Specifically, as described above, it will not be described here.
- the patrol scene generating unit 14 in this embodiment is configured to automatically set different patrol scenes for selection according to the camera label corresponding to the camera in the patrol camera list; the patrol scene includes:
- the camera in the camera tag that browses the click rate exceeding the second preset value constitutes a second patrol scene
- All the cameras in the patrol camera list constitute a fourth patrol scene.
- the above label generating unit 14 can be disposed on the video client.
- a ninth embodiment of the present invention provides a video patrol system, including a configuration unit 11, an intelligent analysis unit 21, a list output unit 12, and a patrol time calculation unit 15.
- the configuration unit 11, the smart analysis unit 21, the list output unit 12, and the label generation unit 13 in this embodiment are the same as the configuration unit 11, the smart analysis unit 21, and the list output unit 12 in the sixth embodiment described above, specifically as described above. , will not repeat them here.
- the patrol time calculation unit 15 in this embodiment is configured to acquire a historical intelligent analysis event corresponding to the camera in the patrol camera list, and dynamically change the patrol time interval of each round of the camera according to the patrol time interval; In the video client.
- the label generation mentioned in the eighth embodiment above may also be included.
- the unit 13 and the patrol scene generating unit 14 are specifically described above, and are not described herein again.
- a tenth embodiment of the present invention provides a video patrol system, including a configuration unit 11, an intelligent analysis unit 21, a list output unit 12, and a patrol time calculation unit 15.
- the configuration unit 11, the smart analysis unit 21, the list output unit 12, and the label generation unit 13 in this embodiment are the same as the configuration unit 11, the smart analysis unit 21, and the list output unit 12 in the sixth embodiment described above, specifically as described above. , will not repeat them here.
- the patrol time calculation unit 15 is specifically configured to:
- N cn a total number of intelligent analysis events corresponding to the camera C n in a certain period of the previous T n days, forming an array with the time period [T 1 , T 2 , . . . T n ] the number of arrays corresponding to a list of intelligent analysis of events [N c1, N c2, ... N cn]; determining the maximum value and the minimum value N cmax N cmin.
- the video inspection task switches the camera C n , and the camera C n will play the T cnow duration.
- the label generating unit 13 and the patrol scene generating unit 14 mentioned in the foregoing eighth embodiment may be further included, as described above, and details are not described herein again.
- the technical solution of the present invention may be in the form of a software product in essence or in part contributing to the prior art.
- the computer software product is stored in a storage medium (such as ROM/RAM, disk, CD), and includes a plurality of instructions for making a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device). Etc.) Performing the methods described in various embodiments of the invention.
- the video patrol inspection method and system thereof realizes the automatic recommendation of the camera list when the new video patrol task is realized by the statistics of the historical intelligent analysis event, thereby reducing the video patrol configuration due to the uneven artificial experience.
- the difference and irrationality improve the rational application of video surveillance system resources.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
本发明公开一种视频巡检方法及其系统，该方法包括：接收对摄像头的智能分析任务的配置；获取摄像头实时视频流并根据智能分析任务进行视频分析，检测智能分析事件；接收到视频巡检任务时，获取摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。该系统包括：配置单元，用于接收对摄像头的智能分析任务的配置；智能分析单元，用于获取摄像头实时视频流并根据智能分析任务检测智能分析事件；列表输出单元，用于获取摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。本发明能自动生成巡检摄像头列表减少由于人工经验参差不齐，导致视频巡检配置的差异性及不合理性，提高视频监控系统资源的合理应用性。
Description
本发明涉及智能分析技术与视频监控系统技术领域，尤其涉及一种视频巡检方法及其系统。
视频巡检是视频监控系统技术中重要的组成模块，其约定一组摄像头列表，按照规定的巡检时间间隔，依次在指定的视频窗口中进行轮巡播放实时视频，配合人工值守，对视频画面中异常情况进行处理。
智能分析是一种计算机图像视觉分析技术，通过将场景中背景和目标分离进而分析并追踪在摄像机场景内出现的目标。用户可以根据的视频内容分析功能，通过在不同摄像机的场景中预设不同的报警规则，一旦目标在场景中出现了违反预定义规则的行为，形成相应的智能分析事件。目前主要的智能分析事件：周界防范、人数统计、物品预留、物品盗移、人群聚集、人群驱散、徘徊、交通事件检测、烟火检测、人群密度、打架等。
目前视频巡检的创建方式主要是由人工先选择将要被巡检的摄像头，设置巡检时间间隔，同时再指定视频窗口，最后设置视频巡检的启动条件，当启动条件满足时，自动启动视频巡检任务。该创建方法存在以下问题：
1、巡检的摄像头是由人工选择，存在一定的主观性。例如选取了视频画面人车稀疏的摄像头，巡检这些摄像头，会浪费部分系统资源。
2、巡检时间间隔不能自动根据视频画面中的人车密度调节，所有摄像头的巡检时间间隔都相同。如问题1所述，人车稠密与稀疏的摄像头的巡检时间间隔相同，浪费部分系统资源。
发明内容
本发明的主要目的在于提出一种视频巡检方法及其系统，旨在充分利用系统资源，更好地实现视频巡视。
为实现上述目的，本发明提出一种视频巡检方法，其中，所述方法包括步骤：
接收对摄像头的智能分析任务的配置；
获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；
接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。
可选地，所述接收对摄像头的智能分析任务的配置之后，所述方法还包括步骤：为所述摄像头生成相应摄像头标签；所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项。
可选地，所述接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表之后，所述方法还包括步骤：根据所述巡检摄像头列表中的所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：
将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；
将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；
将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；
将所述巡检摄像头列表中的所有摄像头构成第四巡检场景。
可选地，所述接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表之后，所述方法还包括步骤：获取所
述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔。
可选地，所述获取所述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔包括：
将一天的时间分成n个时间段，形成时间段数组[T1,T2,...Tn]；
获取前Tn天中的某一所述时间段内的摄像头Cn对应的智能分析事件的总数量Ncn，形成与所述时间段数组[T1,T2,...Tn]相对应的智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]；确定最大值Ncmax与最小值Ncmin。
当视频巡检队列中的摄像头Cn轮巡切换前，确定当前时刻在所述时间段数组[T1,T2,...Tn]中对应的时间段Tcnow；在所述智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]查询到相应的值Ncnow，代入公式：
设置摄像头Cn的巡检时间间隔为Tcnow，视频巡检任务切换摄像头Cn，摄像头Cn将播放Tcnow时长。
此外，为实现上述目的，本发明还提供一种视频巡检系统，其中，包括：
配置单元，用于接收对摄像头的智能分析任务的配置；
智能分析单元，用于获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；
列表输出单元，用于接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。
可选地，还包括标签生成单元，用于为所述摄像头生成相应摄像头标签；所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项。
可选地，还包括巡检场景生成单元，用于根据所述巡检摄像头列表中的
所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：
将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；
将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；
将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；
将所述巡检摄像头列表中的所有摄像头构成第四巡检场景。
可选地，还包括巡检时间计算单元，用于获取所述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔。
可选地，所述巡检时间计算单元具体用于：
将一天的时间分成n个时间段，形成时间段数组[T1,T2,...Tn]；
获取前Tn天中的某一所述时间段内的摄像头Cn对应的智能分析事件的总数量Ncn，形成与所述时间段数组[T1,T2,...Tn]相对应的智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]；确定最大值Ncmax与最小值Ncmin。
当视频巡检队列中的摄像头Cn轮巡切换前，确定当前时刻在所述时间段数组[T1,T2,...Tn]中对应的时间段Tcnow；在所述智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]查询到相应的值Ncnow，代入公式：
设置摄像头Cn的巡检时间间隔为Tcnow，视频巡检任务切换摄像头Cn，摄像头Cn将播放Tcnow时长。
本发明提出的一种视频巡检方法及其系统，通过对历史智能分析事件的
统计，实现了新建视频巡检任务时，自动推荐摄像头列表，减少由于人工经验参差不齐，导致视频巡检配置的差异性及不合理性，提高视频监控系统资源的合理应用性。
图1为实现本发明各个实施例的视频巡检系统的硬件结构示意图；
图2为本发明第一实施例的视频巡检方法的流程示意图；
图3为本发明第二实施例的视频巡检方法的流程示意图；
图4为本发明第三实施例的视频巡检方法的流程示意图；
图5为本发明第四实施例的视频巡检方法的流程示意图；
图6为本发明第六实施例的视频巡检系统的结构示意图；
图7为本发明第七实施例的视频巡检系统的结构示意图；
图8为本发明第八实施例的视频巡检系统的结构示意图；
图9为本发明第九实施例的视频巡检系统的结构示意图；
本发明目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
在后续的描述中，使用用于表示元件的诸如“模块”、“部件”或“单元”的后缀仅为了有利于本发明的说明，其本身并没有特定的意义。因此，"模块"与"部件"可以混合地使用。
本发明提供的视频巡检系统的硬件结构包括视频客户端10、中心服务器20、智能分析服务器30、数据库40、转发服务器50、摄像头70。
其中，中心服务器20是视频巡检系统的核心，用于处理设备管理、智能分析事件数据保存等多种业务功能。智能分析服务器30可以按照一定的智能分析规则，对摄像头实时视频进行智能分析，并将智能分析事件上报到中心服务器20，中心服务器20再上报到视频客户端10。视频客户端10是视频巡检系统业务客户端，具有查看摄像头的实时视频、接收智能分析事件、视频巡检等业务功能。转发服务器50为多媒体转发服务器，负责分发前端摄像头的视频流。数据库40可以存储视频监控系统的业务数据，例如设备信息、智能分析事件等所有业务相关数据。
基于上述视频巡检系统的硬件结构，提出本发明方法各个实施例。
如图2所示，本发明第一实施例提供一种视频巡检方法，包括步骤：
S11、接收对摄像头的智能分析任务的配置；
一般地，用户可以根据摄像头照射的场景及智能分析规则通过视频客户端完成对对摄像头的智能分析任务的配置；配置好的智能分析任务可以通过中心服务器，保存到数据库；
S12、获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；
更具体地，智能分析服务器从中心服务器读取智能分析任务后，通过转发服务器获取摄像头的实时视频流，进行视频分析，检测到智能分析事件时，上报到中心服务器，中心服务器保存智能分析事件到数据库，同时通知视频客户端。上述智能分析事件可以包括：周界防范、人数统计、物品预留、物品盗移、人群聚集、人群驱散、徘徊、交通事件检测、烟火检测、人群密度、打架等；
S13、接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。
更具体地，视频客户端接收到用户配置的视频巡检任务时，可以通过中
心服务器从数据库中获取所有摄像头的智能分析事件数量并排序，根据排序结果，将出现智能分析事件频率比较高的摄像头加入视频巡检队列，同时排除不在线的摄像头。
如图3所示，本发明第二实施例提供一种视频巡检方法，包括步骤：
S21、接收对摄像头的智能分析任务的配置；
一般地，用户可以根据摄像头照射的场景及智能分析规则通过视频客户端完成对对摄像头的智能分析任务的配置；配置好的智能分析任务可以通过中心服务器，保存到数据库；
S22、为摄像头生成相应摄像头标签；
所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项；
S23、获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；
更具体地，智能分析服务器从中心服务器读取智能分析任务后，通过转发服务器获取摄像头的实时视频流，进行视频分析，检测到智能分析事件时，上报到中心服务器，中心服务器保存智能分析事件到数据库，同时通知视频客户端。上述智能分析事件可以包括：周界防范、人数统计、物品预留、物品盗移、人群聚集、人群驱散、徘徊、交通事件检测、烟火检测、人群密度、打架等；
S24、接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。
更具体地，视频客户端接收到用户配置的视频巡检任务时，可以通过中心服务器从数据库中获取所有摄像头的智能分析事件数量并排序，根据排序结果，将出现智能分析事件频率比较高的摄像头加入视频巡检队列，同时排除不在线的摄像头。
如图4所示，本发明第三实施例提供一种视频巡检方法，包括步骤：
S31、接收对摄像头的智能分析任务的配置；
一般地，用户可以根据摄像头照射的场景及智能分析规则通过视频客户端完成对对摄像头的智能分析任务的配置；配置好的智能分析任务可以通过中心服务器，保存到数据库；
S32、为摄像头生成相应摄像头标签；
所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项；
S33、获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；
更具体地，智能分析服务器从中心服务器读取智能分析任务后，通过转发服务器获取摄像头的实时视频流，进行视频分析，检测到智能分析事件时，上报到中心服务器，中心服务器保存智能分析事件到数据库，同时通知视频客户端。上述智能分析事件可以包括：周界防范、人数统计、物品预留、物品盗移、人群聚集、人群驱散、徘徊、交通事件检测、烟火检测、人群密度、打架等；
S34、接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。
更具体地，视频客户端接收到用户配置的视频巡检任务时，可以通过中心服务器从数据库中获取所有摄像头的智能分析事件数量并排序，根据排序结果，将出现智能分析事件频率比较高的摄像头加入视频巡检队列，同时排除不在线的摄像头。
S35、根据所述巡检摄像头列表中的所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：
将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；
将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；
将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；
将所述巡检摄像头列表中的所有摄像头构成第四巡检场景；
还可以在所述巡检摄像头列表中的自定义至少一个摄像头构成第五巡检场景；
用户可以在上述巡检场景选择一个来执行。
如图5所示，本发明第四实施例提供一种视频巡检方法，包括步骤：
S41、接收对摄像头的智能分析任务的配置；
一般地，用户可以根据摄像头照射的场景及智能分析规则通过视频客户端完成对对摄像头的智能分析任务的配置；配置好的智能分析任务可以通过中心服务器，保存到数据库；
S42、获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；
更具体地，智能分析服务器从中心服务器读取智能分析任务后，通过转发服务器获取摄像头的实时视频流，进行视频分析，检测到智能分析事件时，上报到中心服务器，中心服务器保存智能分析事件到数据库，同时通知视频客户端。上述智能分析事件可以包括：周界防范、人数统计、物品预留、物品盗移、人群聚集、人群驱散、徘徊、交通事件检测、烟火检测、人群密度、打架等；
S43、接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表；
更具体地，视频客户端接收到用户配置的视频巡检任务时，可以通过中心服务器从数据库中获取所有摄像头的智能分析事件数量并排序，根据排序结果，将出现智能分析事件频率比较高的摄像头加入视频巡检队列，同时排
除不在线的摄像头。
S44、获取所述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔；
此时，每个摄像头每次对应的轮巡的巡检时间间隔都是不同的。
在本发明的另一实施例中，在步骤S41之后，还包括步骤：为摄像头生成相应摄像头标签；所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项；在步骤S43之后，步骤S44之前，还包括步骤：根据所述巡检摄像头列表中的所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；将所述巡检摄像头列表中的所有摄像头构成第四巡检场景；还可以在所述巡检摄像头列表中的自定义至少一个摄像头构成第五巡检场景；用户可以在上述巡检场景选择一个来执行。此时，步骤S44中，只需获取被选择的巡检场景中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔。
本发明第五实施例提供一种视频巡检方法，包括的步骤与第四实施例中提及的S41至S44相同，具体如上所述，此处不再赘述。
需要说明的是，本实施例中，步骤S44包括：
将一天的时间分成n个时间段，形成时间段数组[T1,T2,...Tn]；
获取前Tn天中的某一所述时间段内的摄像头Cn对应的智能分析事件的总数量Ncn，形成与所述时间段数组[T1,T2,...Tn]相对应的智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]；确定最大值Ncmax与最小值Ncmin。
当视频巡检队列中的摄像头Cn轮巡切换前，确定当前时刻在所述时间段数组[T1,T2,...Tn]中对应的时间段Tcnow；在所述智能分析事件的数量数组列表
[Nc1,Nc2,...Ncn]查询到相应的值Ncnow，代入公式：
设置摄像头Cn的巡检时间间隔为Tcnow，视频巡检任务切换摄像头Cn，摄像头Cn将播放Tcnow时长。
在本发明的另一实施例中，在步骤S41之后，还包括步骤：为摄像头生成相应摄像头标签；所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项；在步骤S43之后，步骤S44之前，还包括步骤：根据所述巡检摄像头列表中的所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；将所述巡检摄像头列表中的所有摄像头构成第四巡检场景；还可以在所述巡检摄像头列表中的自定义至少一个摄像头构成第五巡检场景；用户可以在上述巡检场景选择一个来执行。此时，步骤S44中，只需获取被选择的巡检场景中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔。
上面对本发明实施例中的视频巡检方法进行了描述，下面对本发明实施例中的视频巡检系统进行描述。
如图6所示，本发明第六实施例提出一种视频巡检系统，包括配置单元11、智能分析单元21、列表输出单元12。
其中，配置单元11用于接收对摄像头的智能分析任务的配置；在具体实施时，配置单元11可以设置在视频客户端。用户可以根据摄像头照射的场景
及智能分析规则通过视频客户端完成对对摄像头的智能分析任务的配置；配置好的智能分析任务可以通过中心服务器，保存到数据库；
智能分析单元21用于获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；更具体地，智能分析单元21为智能分析服务器的单元，智能分析服务器从中心服务器读取智能分析任务后，通过转发服务器获取摄像头的实时视频流，进行视频分析，检测到智能分析事件时，上报到中心服务器，中心服务器保存智能分析事件到数据库，同时通知视频客户端。上述智能分析事件可以包括：周界防范、人数统计、物品预留、物品盗移、人群聚集、人群驱散、徘徊、交通事件检测、烟火检测、人群密度、打架等；
列表输出单元12，用于接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。列表输出单元12可以设置在视频客户端上，更具体地，视频客户端接收到用户配置的视频巡检任务时，可以通过中心服务器从数据库中获取所有摄像头的智能分析事件数量并排序，根据排序结果，将出现智能分析事件频率比较高的摄像头加入视频巡检队列，同时排除不在线的摄像头。
如图7所示，本发明第七实施例提出一种视频巡检系统，包括配置单元11、智能分析单元21、列表输出单元12、标签生成单元13。
本实施例中的配置单元11、智能分析单元21、列表输出单元12与上述第六实施例中的配置单元11、智能分析单元21、列表输出单元12相同，具体如上所述，此处不再赘述。
不同的是，本实施例中还包括标签生成单元13，用于对所述摄像头生成相应摄像头标签；所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项。上述标签生成单元13可以设置在视频客户端上。
如图8所示，本发明第八实施例提出一种视频巡检系统，包括配置单元11、智能分析单元21、列表输出单元12、标签生成单元13、巡检场景生成单元14。
本实施例中的配置单元11、智能分析单元21、列表输出单元12、标签生成单元13与上述第七实施例中的配置单元11、智能分析单元21、列表输出单元12、标签生成单元13相同，具体如上所述，此处不再赘述。
本实施例中的巡检场景生成单元14用于根据所述巡检摄像头列表中的所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：
将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；
将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；
将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；
将所述巡检摄像头列表中的所有摄像头构成第四巡检场景。
上述标签生成单元14可以设置在视频客户端上。
如图9所示，本发明第九实施例提出一种视频巡检系统，包括配置单元11、智能分析单元21、列表输出单元12、巡检时间计算单元15。
本实施例中的配置单元11、智能分析单元21、列表输出单元12、标签生成单元13与上述第六实施例中的配置单元11、智能分析单元21、列表输出单元12相同，具体如上所述，此处不再赘述。
本实施例中的巡检时间计算单元15用于获取所述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔；可以设置在视频客户端。
在本发明的另一实施例中，还可以包括上述第八实施例提及的标签生成
单元13和巡检场景生成单元14，具体如上所述，在此不再赘述。
本发明第十实施例提出一种视频巡检系统，包括配置单元11、智能分析单元21、列表输出单元12、巡检时间计算单元15。
本实施例中的配置单元11、智能分析单元21、列表输出单元12、标签生成单元13与上述第六实施例中的配置单元11、智能分析单元21、列表输出单元12相同，具体如上所述，此处不再赘述。
与上一实施例不同的是，本实施例中，巡检时间计算单元15具体用于：
将一天的时间分成n个时间段，形成时间段数组[T1,T2,...Tn]；
获取前Tn天中的某一所述时间段内的摄像头Cn对应的智能分析事件的总数量Ncn，形成与所述时间段数组[T1,T2,...Tn]相对应的智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]；确定最大值Ncmax与最小值Ncmin。
当视频巡检队列中的摄像头Cn轮巡切换前，确定当前时刻在所述时间段数组[T1,T2,...Tn]中对应的时间段Tcnow；在所述智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]查询到相应的值Ncnow，代入公式：
设置摄像头Cn的巡检时间间隔为Tcnow，视频巡检任务切换摄像头Cn，摄像头Cn将播放Tcnow时长。
在本发明的另一实施例中，还可以包括上述第八实施例提及的标签生成单元13和巡检场景生成单元14，具体如上所述，在此不再赘述。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体
现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本发明各个实施例所述的方法。
以上仅为本发明的优选实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。
本发明提出的一种视频巡检方法及其系统，通过对历史智能分析事件的统计，实现了新建视频巡检任务时，自动推荐摄像头列表，减少由于人工经验参差不齐，导致视频巡检配置的差异性及不合理性，提高视频监控系统资源的合理应用性。
Claims (10)
- 一种视频巡检方法，所述方法包括步骤：接收对摄像头的智能分析任务的配置；获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。
- 根据权利要求1所述视频巡检方法，其中，所述接收对摄像头的智能分析任务的配置之后，所述方法还包括步骤：为所述摄像头生成相应摄像头标签；所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项。
- 根据权利要求2所述视频巡检方法，其中，所述接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表之后，所述方法还包括步骤：根据所述巡检摄像头列表中的所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；将所述巡检摄像头列表中的所有摄像头构成第四巡检场景。
- 根据权利要求1所述视频巡检方法，其中，所述接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表之后，所述方法还包括步骤：获取所述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔。
- 根据权利要求4所述视频巡检方法，其中，所述获取所述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔包括：将一天的时间分成n个时间段，形成时间段数组[T1,T2,...Tn]；获取前Tn天中的某一所述时间段内的摄像头Cn对应的智能分析事件的总数量Ncn，形成与所述时间段数组[T1,T2,...Tn]相对应的智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]；确定最大值Ncmax与最小值Ncmin。当视频巡检队列中的摄像头Cn轮巡切换前，确定当前时刻在所述时间段数组[T1,T2,...Tn]中对应的时间段Tcnow；在所述智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]查询到相应的值Ncnow，代入公式：若Ncmax＝Ncmin，则巡检时间间隔
设置摄像头Cn的巡检时间间隔为Tcnow，视频巡检任务切换摄像头Cn，摄像头Cn将播放Tcnow时长。 - 一种视频巡检系统，包括：配置单元，用于接收对摄像头的智能分析任务的配置；智能分析单元，用于获取所述摄像头实时视频流并根据所述智能分析任务进行视频分析，检测智能分析事件；列表输出单元，用于接收到视频巡检任务时，获取所述摄像头对应的智能分析事件的数量并据此给出巡检摄像头列表。
- 根据权利要求6所述视频巡检系统，其中，还包括标签生成单元，用于为所述摄像头生成相应摄像头标签；所述摄像头标签包括智能分析事件的数量、浏览点击率、地点中的至少一项。
- 根据权利要求7所述视频巡检系统，其中，还包括巡检场景生成单元， 用于根据所述巡检摄像头列表中的所述摄像头对应的摄像头标签自动设置不同的巡检场景以供选择；所述巡检场景包括：将所述摄像头标签中智能分析事件的数量超过第一预设值的摄像头构成第一巡检场景；将所述摄像头标签中浏览点击率超过第二预设值的摄像头构成第二巡检场景；将所述摄像头标签中地点为预设地点的摄像头构成第三巡检场景；将所述巡检摄像头列表中的所有摄像头构成第四巡检场景。
- 根据权利要求6所述视频巡检系统，其中，还包括巡检时间计算单元，用于获取所述巡检摄像头列表中的摄像头对应的历史智能分析事件，并据此动态改变所述摄像头每次轮巡的巡检时间间隔。
- 根据权利要求9所述视频巡检系统，其中，所述巡检时间计算单元具体用于：将一天的时间分成n个时间段，形成时间段数组[T1,T2,...Tn]；获取前Tn天中的某一所述时间段内的摄像头Cn对应的智能分析事件的总数量Ncn，形成与所述时间段数组[T1,T2,...Tn]相对应的智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]；确定最大值Ncmax与最小值Ncmin。当视频巡检队列中的摄像头Cn轮巡切换前，确定当前时刻在所述时间段数组[T1,T2,...Tn]中对应的时间段Tcnow；在所述智能分析事件的数量数组列表[Nc1,Nc2,...Ncn]查询到相应的值Ncnow，代入公式：若Ncmax＝Ncmin，则巡检时间间隔
设置摄像头Cn的巡检时间间隔为Tcnow，视频巡检任务切换摄像头Cn，摄像头Cn将播放Tcnow时长。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2016/110938 WO2018112722A1 (zh) | 2016-12-20 | 2016-12-20 | 一种视频巡检方法及其系统 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2016/110938 WO2018112722A1 (zh) | 2016-12-20 | 2016-12-20 | 一种视频巡检方法及其系统 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2018112722A1 true WO2018112722A1 (zh) | 2018-06-28 |
Family
ID=62624273
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2016/110938 WO2018112722A1 (zh) | 2016-12-20 | 2016-12-20 | 一种视频巡检方法及其系统 |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2018112722A1 (zh) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110750424A (zh) * | 2019-10-17 | 2020-02-04 | 北京百度网讯科技有限公司 | 资源巡检方法和装置 |
| CN110971865A (zh) * | 2018-09-28 | 2020-04-07 | 杭州海康威视数字技术股份有限公司 | 视频数据获取方法、装置、电子设备及可读存储介质 |
| CN111193904A (zh) * | 2019-12-17 | 2020-05-22 | 视联动力信息技术股份有限公司 | 巡检控制方法、装置及可读存储介质 |
| CN112261440A (zh) * | 2020-10-28 | 2021-01-22 | 成都华栖云科技有限公司 | 基于镜头识别及场景识别的分布式超高清视频识别方法和系统 |
| CN113365018A (zh) * | 2020-03-06 | 2021-09-07 | 杭州海康威视数字技术股份有限公司 | 一种周期调整方法、装置及设备 |
| CN113542672A (zh) * | 2021-05-25 | 2021-10-22 | 浙江大华技术股份有限公司 | 摄像机巡航方法、电子设备及存储介质 |
| CN113591549A (zh) * | 2021-06-16 | 2021-11-02 | 浙江大华技术股份有限公司 | 一种视频事件检测方法、计算机设备以及装置 |
| CN114582041A (zh) * | 2022-05-07 | 2022-06-03 | 绿城科技产业服务集团有限公司 | 一种基于视频分析算法的园区巡更系统 |
| CN115484320A (zh) * | 2022-10-27 | 2022-12-16 | 陕西思极科技有限公司 | 应用于智能识别终端群的数据传输方法及监控设备 |
| CN116863272A (zh) * | 2023-07-11 | 2023-10-10 | 广州兴合科技有限公司 | 基于工程分类的巡检图像处理方法及系统 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103795987A (zh) * | 2014-02-27 | 2014-05-14 | 福州华森智能电网技术有限公司 | 嵌入式视频采集处理系统及其实现方法 |
| CN104574722A (zh) * | 2013-10-12 | 2015-04-29 | 北京航天长峰科技工业集团有限公司 | 一种基于多传感器的海港安全控制系统 |
| CN104700609A (zh) * | 2013-12-06 | 2015-06-10 | 大连灵动科技发展有限公司 | 智能交通监控方法 |
| CN106209462A (zh) * | 2016-07-18 | 2016-12-07 | 四川君逸数码科技股份有限公司 | 基于智能视频技术的智慧金融管理系统 |
-
2016
- 2016-12-20 WO PCT/CN2016/110938 patent/WO2018112722A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104574722A (zh) * | 2013-10-12 | 2015-04-29 | 北京航天长峰科技工业集团有限公司 | 一种基于多传感器的海港安全控制系统 |
| CN104700609A (zh) * | 2013-12-06 | 2015-06-10 | 大连灵动科技发展有限公司 | 智能交通监控方法 |
| CN103795987A (zh) * | 2014-02-27 | 2014-05-14 | 福州华森智能电网技术有限公司 | 嵌入式视频采集处理系统及其实现方法 |
| CN106209462A (zh) * | 2016-07-18 | 2016-12-07 | 四川君逸数码科技股份有限公司 | 基于智能视频技术的智慧金融管理系统 |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110971865A (zh) * | 2018-09-28 | 2020-04-07 | 杭州海康威视数字技术股份有限公司 | 视频数据获取方法、装置、电子设备及可读存储介质 |
| CN110750424A (zh) * | 2019-10-17 | 2020-02-04 | 北京百度网讯科技有限公司 | 资源巡检方法和装置 |
| CN110750424B (zh) * | 2019-10-17 | 2023-07-04 | 北京百度网讯科技有限公司 | 资源巡检方法和装置 |
| CN111193904A (zh) * | 2019-12-17 | 2020-05-22 | 视联动力信息技术股份有限公司 | 巡检控制方法、装置及可读存储介质 |
| CN111193904B (zh) * | 2019-12-17 | 2022-12-13 | 视联动力信息技术股份有限公司 | 巡检控制方法、装置及可读存储介质 |
| CN113365018B (zh) * | 2020-03-06 | 2022-07-01 | 杭州海康威视数字技术股份有限公司 | 一种周期调整方法、装置及设备 |
| CN113365018A (zh) * | 2020-03-06 | 2021-09-07 | 杭州海康威视数字技术股份有限公司 | 一种周期调整方法、装置及设备 |
| CN112261440B (zh) * | 2020-10-28 | 2022-11-25 | 成都华栖云科技有限公司 | 基于镜头识别及场景识别的分布式超高清视频识别方法和系统 |
| CN112261440A (zh) * | 2020-10-28 | 2021-01-22 | 成都华栖云科技有限公司 | 基于镜头识别及场景识别的分布式超高清视频识别方法和系统 |
| CN113542672A (zh) * | 2021-05-25 | 2021-10-22 | 浙江大华技术股份有限公司 | 摄像机巡航方法、电子设备及存储介质 |
| CN113542672B (zh) * | 2021-05-25 | 2023-08-18 | 浙江大华技术股份有限公司 | 摄像机巡航方法、电子设备及存储介质 |
| CN113591549A (zh) * | 2021-06-16 | 2021-11-02 | 浙江大华技术股份有限公司 | 一种视频事件检测方法、计算机设备以及装置 |
| CN114582041A (zh) * | 2022-05-07 | 2022-06-03 | 绿城科技产业服务集团有限公司 | 一种基于视频分析算法的园区巡更系统 |
| CN114582041B (zh) * | 2022-05-07 | 2023-02-17 | 绿城科技产业服务集团有限公司 | 一种基于视频分析算法的园区巡更系统 |
| CN115484320A (zh) * | 2022-10-27 | 2022-12-16 | 陕西思极科技有限公司 | 应用于智能识别终端群的数据传输方法及监控设备 |
| CN116863272A (zh) * | 2023-07-11 | 2023-10-10 | 广州兴合科技有限公司 | 基于工程分类的巡检图像处理方法及系统 |
| CN116863272B (zh) * | 2023-07-11 | 2024-01-30 | 广州兴合科技有限公司 | 基于工程分类的巡检图像处理方法及系统 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018112722A1 (zh) | 一种视频巡检方法及其系统 | |
| RU2484529C1 (ru) | Способ ранжирования видеоданных | |
| EP2936388B1 (en) | System and method for detection of high-interest events in video data | |
| US9226037B2 (en) | Inference engine for video analytics metadata-based event detection and forensic search | |
| US20140375819A1 (en) | Autonomous video management system | |
| CN110263613A (zh) | 监控视频处理方法及装置 | |
| CN111209331A (zh) | 目标对象的检索方法、装置及电子设备 | |
| CN105210048A (zh) | 基于社交媒体的内容识别引擎 | |
| KR20080075091A (ko) | 실시간 경보 및 포렌식 분석을 위한 비디오 분석 데이터의저장 | |
| US20170294213A1 (en) | Method for video investigation | |
| CN110572617B (zh) | 一种环境监控的处理方法、装置及存储介质 | |
| CN107920223B (zh) | 一种对象行为检测方法及装置 | |
| US20180150683A1 (en) | Systems, methods, and devices for information sharing and matching | |
| CN112182295B (zh) | 基于行为预测的业务处理方法、装置及电子设备 | |
| CN111401239B (zh) | 一种视频分析方法、装置、系统、设备及存储介质 | |
| CN109800329B (zh) | 一种监控方法及装置 | |
| WO2021114985A1 (zh) | 一种同行对象识别方法、装置、服务器及系统 | |
| DE112016004160T5 (de) | Benutzeroberfläche für Videozusammenfassungen | |
| CN106781167B (zh) | 监测物体运动状态的方法及设备 | |
| CN104253981B (zh) | 一种用于视频侦查的运动目标按颜色排序的方法 | |
| CN109783663B (zh) | 一种归档方法及装置 | |
| KR20160082915A (ko) | 미디어 데이터 분류 방법 및 그 장치 | |
| CN103731639A (zh) | 一种通过智能视频安防系统提供图形出入记录的方法 | |
| CN109670079B (zh) | 基于监控系统的督察方法和系统 | |
| CN109120896B (zh) | 安防视频监控卫士系统 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16924295 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 16924295 Country of ref document: EP Kind code of ref document: A1 |