WO2016095374A1 - 一种视频监控方法和装置 - Google Patents

一种视频监控方法和装置 Download PDF

Info

Publication number
WO2016095374A1
WO2016095374A1 PCT/CN2015/076309 CN2015076309W WO2016095374A1 WO 2016095374 A1 WO2016095374 A1 WO 2016095374A1 CN 2015076309 W CN2015076309 W CN 2015076309W WO 2016095374 A1 WO2016095374 A1 WO 2016095374A1
Authority
WO
WIPO (PCT)
Prior art keywords
video
image data
video image
background
group
Prior art date
Application number
PCT/CN2015/076309
Other languages
English (en)
French (fr)
Inventor
邓硕
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to US15/537,488 priority Critical patent/US20180270454A1/en
Priority to EP15868891.1A priority patent/EP3220636A4/en
Publication of WO2016095374A1 publication Critical patent/WO2016095374A1/zh

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
    • H04N7/188—Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19608—Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19678—User interface
    • G08B13/19686—Interfaces masking personal details for privacy, e.g. blurring faces, vehicle license plates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30232—Surveillance

Definitions

  • Embodiments of the present invention relate to monitoring technologies, and in particular, to a video monitoring method and apparatus.
  • Such applications usually connect video cameras and smartphone terminals through a video surveillance system to complete the collection, monitoring and management of video content, thus satisfying the application requirements for security, management, affection and sharing, but also with Some questions include: content privacy protection issues when sharing video or camera permissions to other accounts; or whether there is a solution to enable personalized browsing and management of surveillance video content according to the needs of different types of members. . Therefore, how to achieve more secure video surveillance is an urgent problem to be solved.
  • an embodiment of the present invention provides a video monitoring method and apparatus, which can better ensure the security of video surveillance, thereby protecting the privacy of video surveillance content.
  • an embodiment of the present invention provides a video monitoring method, including:
  • the obtained video image data is filtered according to the permission mapping table and the obtained group;
  • the video image data obtained after the screening is displayed on the video monitor.
  • the method before the acquiring the video image data, the method further includes: preset the permission mapping table;
  • the filtering rules of the permission mapping table are correspondingly configured with corresponding rights of each member group.
  • the acquiring video image data includes: acquiring the video image data
  • the preprocessing the collected video image data includes:
  • the screening of the obtained video image data includes:
  • the background of the video image data and the moving target foreground are filtered according to the rights of the corresponding member groups configured in the filtering rule of the permission mapping table.
  • displaying the video image data obtained after the screening on the video monitor comprises:
  • the video stream is displayed to a video monitor.
  • the embodiment of the invention further provides a video monitoring device, which at least includes:
  • Obtaining a module configured to obtain video image data and a group in which the member is located;
  • a screening module configured to filter the obtained video image data according to the permission mapping table permission mapping table and the obtained group
  • the display module is configured to display the filtered video image data on the video monitor.
  • it also includes:
  • the preset module is configured as a preset permission mapping table.
  • the filtering rules of the permission mapping table are correspondingly configured with corresponding permissions of each member group.
  • the acquiring module includes an acquiring unit and a pre-processing unit; wherein
  • An acquisition unit configured to collect the video image data
  • a pre-processing unit configured to pre-process the collected video image data.
  • the pre-processing unit comprises at least an extracting subunit and an identifying subunit;
  • Extracting a subunit configured to extract a background of the collected video image data and a moving target foreground
  • the recognition sub-unit is set to the foreground and moving target foreground extracted in the recognition and annotation.
  • the screening module is specifically configured to:
  • the background of the video image data and the moving target foreground are filtered according to the rights of the corresponding member groups configured in the filtering rule of the permission mapping table.
  • the display module includes at least a fusion unit and a transmission unit;
  • a merging unit configured to fuse the moving target foreground and the corresponding background that meet the screening condition into a video stream
  • the sending unit is configured to send the obtained video stream to the video monitor display.
  • the video monitoring method provided by the embodiment of the present invention obtains the video image data and the group in which the member is located; and filters the video image data according to the permission of the corresponding member group configured in the filtering rule of the permission mapping table; The obtained video image data is performed on the video monitor display.
  • the embodiment of the invention reconstructs the video monitoring content, and generates video data required by the user according to the customization of different types of member groups, thereby achieving the purpose of personalized browsing and privacy protection of the customized content; substantially reducing the actual workload of the monitoring personnel ; saves transmission bandwidth.
  • FIG. 1 is a schematic flowchart of a first embodiment of a video monitoring method according to the present invention
  • FIG. 2 is a schematic flowchart of a second embodiment of a video monitoring method according to the present invention.
  • FIG. 3 is a schematic diagram showing an implementation flow of acquiring video image data in FIG. 1;
  • FIG. 4 is a schematic diagram showing an implementation process of preprocessing the collected video image data in FIG. 3;
  • FIG. 5 is a schematic flowchart diagram of a third embodiment of a video monitoring method according to the present invention.
  • FIG. 6 is a schematic diagram showing an implementation flow of displaying video image data obtained after screening on a video monitor in FIG. 1;
  • FIG. 7 is a schematic structural diagram of a first embodiment of a video monitoring apparatus according to the present invention.
  • FIG. 8 is a schematic structural diagram of a second embodiment of a video monitoring apparatus according to the present invention.
  • FIG. 9 is a schematic structural diagram of an acquisition module of FIG. 7;
  • FIG. 10 is a schematic structural diagram of a preprocessing unit of FIG. 9;
  • FIG. 11 is a schematic structural diagram of the display module of FIG. 7.
  • FIG. 1 is a schematic flowchart of a first embodiment of a video monitoring method according to the present invention, as shown in FIG. In the first embodiment, the following steps are included:
  • Step 100 Obtain video image data and a group in which the member is located.
  • the video monitoring system includes an acquisition front end, a service processing server, and a client.
  • the video monitoring device collects video image data by collecting the front end.
  • the front end of the collection can adopt a 360-degree rotatable night light camera, and can automatically collect video images of any angle in a night environment.
  • the optical flow can be adopted.
  • Method or inter-frame difference method or background difference method dynamically extracts the background of video image data and the foreground of moving objects, and performs face recognition by face feature method to identify specific people or animals in the moving target, and further to the moving target Track and capture video image data of a specific target and save it in the database.
  • the video monitoring device can log in to the service processing server through the client administrator account of the WEB or the application, and establish different members and member groups, and divide different members into corresponding member groups.
  • the group to which the member belongs can be determined according to the current member, such as the identification information.
  • Step 200 Filter the obtained video image data according to the permission mapping table and the obtained group.
  • the video monitoring device pre-configures the rights mapping table, and filters the obtained video images through the service processing server according to the rights of the corresponding member groups configured in the filtering rules of the rights mapping table.
  • the member component can be a member group. , group B, group C, group A is set to track only specific people, group B is set to track only certain valuables, and group C is set to only the number of the user's license plate Tracking.
  • Each member can be added to multiple groups after being authenticated. If member A is authorized to authenticate, it can be added to group A, group B, or group C, or group A or group B at the same time. And the C member group. If the level of member A is high, you can also modify the permission mapping table to change the permissions of the member group.
  • the video monitoring device filters the video image data according to the filtering rule of the permission mapping table. For example, if the member A is divided into the A member group, the video image data of the corresponding specific person is filtered according to the permission mapping table, so that the member A is In the video monitoring permission, only the video image data of the corresponding specific person can be monitored.
  • Step 300 Display the video image data obtained after the screening on the video monitor.
  • the video monitoring device sends the filtered video image data to the customer service terminal, and the video stream of the filtered video image is decoded by the customer service terminal, and then presented through the interactive interface of the video monitor for the user to monitor and view.
  • the video monitoring method provided in this embodiment reconstructs the video content and filters out the video data required by the user according to the customization of different types of member groups.
  • the purpose of personalized browsing and privacy protection monitoring of the customized content is achieved.
  • the video surveillance security is better ensured, thereby protecting the privacy of the video surveillance content; on the other hand, the actual workload of the monitoring personnel is greatly reduced and the transmission bandwidth is saved without reducing the user's perception.
  • FIG. 2 is a schematic flowchart of a video monitoring method according to a second embodiment of the present invention. As shown in FIG. 2, in the video monitoring method provided by the second embodiment, before the step 100, the method further includes:
  • the preset permission mapping table, the filtering rules of the permission mapping table are correspondingly configured with the corresponding rights of each member group.
  • the corresponding one of the member groups is configured in the filtering rule of the permission mapping table.
  • Each member can correspond to multiple member groups.
  • each member can correspond to multiple permissions, forming a many-to-many mapping.
  • the A member group is set to track only specific people
  • the B member group is set to track only certain valuable items
  • the C member group is set to track only the user's license plate number. If the A member corresponds to the A member group and the B member group, the video of the specific person and the specific valuable item can be monitored.
  • the member with a high level can modify the permission mapping table, and correspondingly modify the permission of the member group. For example, when member A is at the highest level, and currently in the group B group, only the specific valuable items set by the group B group can be performed. The tracked permissions are modified to the right to track specific animals.
  • the video monitoring method provided in this embodiment selects a permission mapping table, filters the collected video image data, and filters out video data required by the user according to different types of member groups.
  • the customized content is achieved.
  • the purpose of personalized browsing and privacy protection monitoring is to better ensure the security of video surveillance, thus protecting the privacy of video surveillance content; on the other hand, it greatly reduces the actual situation of monitoring personnel without reducing the user's perception. The workload and the transmission bandwidth are saved.
  • FIG. 3 is a schematic diagram of the implementation of acquiring video image data in FIG. 1 , as shown in FIG. 3 , including:
  • Step 110 Collect video image data.
  • the video monitoring device receives the control command from the service processing server, and collects the video data through the collecting front end.
  • the collecting front end may be a 360-degree rotatable night light camera, which can be arbitrarily in the night environment. Angled video images are intelligently captured.
  • Step 120 Perform pre-processing on the collected video image data.
  • the video monitoring device preprocesses the collected video image data through the acquisition front end.
  • the optical image method or the background difference method may be used to dynamically extract the video image data background and the moving target foreground, and pass the facial features.
  • the method performs face recognition to identify a specific person or animal in the moving target, and further tracks the moving target, acquires video image data of a specific target, and saves it in a database. And collecting the background of each angle, respectively storing the background of the corresponding angle of the video image data in the database.
  • FIG. 4 is a schematic diagram of an implementation process of preprocessing the collected video image data in FIG. 3, as shown in FIG. 4, including:
  • Step 121 Extract the background of the collected video image data and the foreground of the moving target.
  • the video monitoring device uses the method of moving object detection such as the optical flow method to separate the foreground and background of the current video image data, extracts the foreground moving target, and recognizes the contour of the moving target.
  • moving object detection such as the optical flow method to separate the foreground and background of the current video image data, extracts the foreground moving target, and recognizes the contour of the moving target.
  • optical flow method is only used as an example here, and of course, it is not limited thereto, and other methods such as an interframe difference method or a background difference method may be used.
  • Step 122 Identify and extract the extracted background and the moving target foreground on the video image.
  • the video monitoring device identifies and extracts a body contour of the moving target, and processes a preview frame sequence of a continuously moving moving target; for the preview frame sequence, uses a target detecting method to detect a foreground target that may appear; if a certain frame appears
  • the foreground target is found, and its representative key feature points are found.
  • the key feature points can be randomly generated, or the corner points can be used as the key feature points.
  • the corner points are the local maximum points of the curvature on the body contour.
  • the contour feature has a decisive effect, and once the contour feature of the moving target is found, the shape of the target is roughly grasped.
  • the corner feature is an important feature of the image.
  • the corner points have the advantages of rotation invariance and no change with the illumination condition.
  • the amount of data involved in the calculation can be reduced without Loss of important grayscale information of the image, using the corner feature to match can greatly improve the speed of matching; for any two adjacent video frames afterwards, find the most important feature points in the previous frame in the current frame. Good position, so as to get the position coordinates of the moving target in the current frame; so iteratively, the foreground moving target can be tracked, and the acquired foreground and background can be obtained.
  • filter some objects filter out the desired video image data, and identify moving targets in the foreground.
  • FIG. 5 is a schematic diagram of a functional module of a third embodiment of a video monitoring method according to the present invention. As shown in FIG. 5, based on the first embodiment or the second embodiment (FIG. 5 shows the basis of the second embodiment). ), step 200 specifically includes:
  • Step 200A Filter the background of the video image data and the moving target foreground according to the rights of the corresponding member groups configured in the filtering rule of the permission mapping table.
  • the video monitoring device obtains the permission of the corresponding member group from the filtering rule according to the preset permission mapping table, and filters the extracted background and the moving target foreground according to the permission of the corresponding member group. For example, a member group selects a specific person in the video image data, and a member group selects a specific valuable item in the video image data, and the group C group tracks the license plate number of the user in the video image data. Thereby identifying specific people, cars, and objects.
  • FIG. 6 is a schematic diagram of an implementation process of displaying video image data obtained by screening in a video monitor in FIG. 1 , as shown in FIG. 6 , specifically including:
  • Step 310 Encode the moving target foreground and the corresponding background that meet the screening condition into a video stream.
  • the video monitoring device combines the identified background and the moving target foreground to generate a new video stream based on the IP packet to describe an XML (eXtensible Markup Language) file.
  • the principle of the video stream fusion coding is: firstly, the video data of the selected specific moving target foreground is obtained; and the background corresponding to the video data of the specific moving target foreground is obtained; and then the background identifier corresponding to the moving target foreground is obtained,
  • the background identifier may be an angle identifier of the background corresponding to the foreground of the moving target. For example, the angle of the background corresponding to the foreground of the moving target is 50 degrees.
  • the filtered moving target foreground and the corresponding background are merged to form a video stream.
  • the video data carried by the video stream only needs to include the video data of the foreground of the specific moving target and the corresponding background image, that is, only one corresponding background image needs to be transmitted in the background of the video stream transmission.
  • the background of different angles it is only necessary to transmit multiple background images of corresponding angles, thereby greatly reducing the amount of information transmission and greatly improving the efficiency of analysis and processing.
  • Step 320 Send the obtained video stream to the video monitor, and the video monitor decodes the video stream and presents the video stream to the interactive interface for the user to perform monitoring and viewing.
  • the video monitoring device sends the generated video stream to the video monitor, and the video monitor pair
  • the video stream is decoded and presented to the interactive interface for monitoring and viewing by the user.
  • Decoding the video stream includes: first decomposing the merged video stream according to the foreground and the background; and then acquiring the corresponding background image according to the background identifier corresponding to the foreground of the moving target; and then moving the target of the moving target and the corresponding background video data Real-time integration.
  • the video monitoring method provided by the embodiment by analyzing and reconstructing the video content, and generating the video stream according to the customized permissions of different types of user groups, on the one hand, achieves the purpose of customized content personalized browsing and privacy protection monitoring, and The security of the video surveillance is well ensured, thereby protecting the privacy of the video surveillance content; on the other hand, the actual workload of the monitoring personnel is greatly reduced and the transmission bandwidth is saved without reducing the user's perception.
  • FIG. 7 is a schematic structural diagram of a first embodiment of a video monitoring apparatus according to the present invention. As shown in FIG. 7, in the first embodiment, the method includes:
  • the obtaining module 10 is configured to acquire video image data and a group in which the member is located;
  • the screening module 20 is configured to filter the obtained video image data according to the permission mapping table and the obtained group;
  • the display module 30 is configured to display the video image data obtained after the screening on the video monitor.
  • the acquiring module 10 is specifically configured to: collect video image data through the collecting front end, and adopt a 360-degree rotatable night light camera in the front end to intelligently collect video images of any angle in a night environment, and may adopt an optical flow method or a frame.
  • the difference between the difference method or the background difference method dynamically extracts the background of the video image data and the foreground of the moving target, and performs face recognition by the facial feature method to identify a specific person or animal in the moving target, and further track and acquire the moving target.
  • the video monitoring device logs in to the service processing server through the client administrator account of the WEB or the application, and establishes different members and member groups, and divides different members into corresponding member groups.
  • the screening module 20 of the video monitoring device is specifically configured to: pre-configure the permission mapping table, and filter the video image through the service processing server according to the permission of the corresponding member group configured in the filtering rule of the permission mapping table, for example,
  • the member components are group A, group B, and group C.
  • Group A is set to track only specific people.
  • Group B is set to track only certain valuable items.
  • the order only tracks the user's license plate number.
  • Each member can be admitted to multiple groups after being authenticated. For example, member A is authorized to be authenticated.
  • the video monitoring device filters the video image data according to the filtering rule of the permission mapping table, for example, the member A is divided into a group member group, and the video image data of the corresponding specific person is filtered out, so that the video monitoring authority of the member A is Only the video image data of the corresponding specific person can be monitored.
  • the display module 30 of the video monitoring device is specifically configured to: send the filtered video image data to the client terminal, and the client terminal decodes the video stream of the filtered video image, and then presents the video monitor through the interactive interface. For users to monitor and view.
  • the video monitoring apparatus reconstructs the video content and filters out the video data required by the user according to the customization of different types of member groups.
  • the purpose of personalized browsing and privacy protection monitoring of the customized content is achieved.
  • the video surveillance security is better ensured, thereby protecting the privacy of the video surveillance content; on the other hand, the actual workload of the monitoring personnel is greatly reduced and the transmission bandwidth is saved without reducing the user's perception.
  • FIG. 8 is a schematic structural diagram of a second embodiment of a video monitoring apparatus according to the present invention. As shown in FIG. 8, according to the first embodiment, the method further includes:
  • the preset module 40 is configured as a preset permission mapping table, wherein the filtering rules of the permission mapping table are correspondingly configured with corresponding rights of each member group.
  • the corresponding one of the member groups is configured in the filtering rule of the permission mapping table.
  • Each member can correspond to multiple member groups.
  • each member can correspond to multiple permissions, forming a many-to-many mapping.
  • the A member group is set to track only specific people
  • the B member group is set to track only certain valuable items
  • the C member group is set to track only the user's license plate number.
  • the member with a high level can modify the permission mapping table, and correspondingly modify the permission of the member group. For example, when member A is at the highest level, and currently in the group B group, only the specific valuable items set by the group B group can be performed.
  • the tracked permissions are modified to the right to track specific animals.
  • the video monitoring apparatus filters the collected video image data by using a preset permission mapping table, and filters out video data required by the user according to customization of different types of member groups.
  • the customized content is achieved.
  • Personalized browsing and privacy protection monitoring for better protection It proves the security of video surveillance, thus protecting the privacy of video surveillance content; on the other hand, it greatly reduces the actual workload of monitoring personnel and saves transmission bandwidth without reducing the user's perception.
  • FIG. 9 is a schematic structural diagram of the acquiring module of FIG. 7. As shown in FIG. 9, the acquiring module 10 specifically includes:
  • the collecting unit 11 is configured to collect video image data
  • the pre-processing unit 12 is configured to pre-process the collected video image data.
  • the collecting unit 11 of the video monitoring device receives the control command from the service processing server, collects the video data through the collecting front end, and the collecting front end can adopt a 360-degree rotatable night light camera, which can be used at any angle in the night environment. Video images are intelligently captured.
  • the pre-processing unit 12 of the video monitoring device pre-processes the collected video image data through the acquisition front end.
  • the optical image data background and the moving target foreground may be dynamically extracted by using an optical flow method or an inter-frame difference method or a background difference method.
  • Face recognition is performed by the facial feature method to identify a specific person or animal in the moving target, and the moving target is further tracked, and the video image data of the specific target is acquired and saved in the database. And collecting the background of each angle, respectively storing the background of the corresponding angle of the video image data in the database.
  • FIG. 10 is a schematic structural diagram of a pre-processing unit of FIG. 9. As shown in FIG. 10, the pre-processing unit 12 specifically includes:
  • the extracting subunit 121 is configured to extract a background of the collected video image data and a moving target foreground;
  • the recognition sub-unit 122 is arranged to identify and extract the extracted background and moving target foreground on the video image.
  • the extraction subunit 121 of the video monitoring device separates the foreground and background of the current video image data by using a method of moving object detection such as the optical flow method, extracts the foreground moving target, and recognizes the contour of the moving target.
  • a method of moving object detection such as the optical flow method
  • the optical flow method is only used as an example here, and of course, it is not limited thereto, and other methods such as an interframe difference method or a background difference method may be used.
  • the identification sub-unit 122 of the video monitoring device identifies and extracts a body contour of the moving object, processes a preview frame sequence of a continuously moving moving object, and detects a possible foreground object by using a target detecting method for the preview frame sequence; If a foreground target appears in a certain frame, find its representative key feature points, wherein the key feature points can be randomly generated or The corner points are used to make the key feature points.
  • the corner points are the local maximum points of the curvature on the main body contour. They have a decisive effect on grasping the contour features of the main body. Once the contour features of the moving objects are found, the shape of the target is roughly grasped.
  • the corner feature is an important feature of the image.
  • the corner points have the advantages of rotation invariance and no change with the illumination condition.
  • the amount of data involved in the calculation can be reduced without Loss of important grayscale information of the image, using the corner feature to match can greatly improve the speed of matching; for any two adjacent video frames afterwards, find the most important feature points in the previous frame in the current frame. Good position, thus obtaining the position coordinates of the moving target in the current frame; thus iteratively, the tracking of the moving target in the foreground can be realized, and the acquired foreground and background are separated, thereby filtering some objects and filtering out the desired video image. Data to identify moving targets in the foreground.
  • the screening module 20 is specifically configured to: view the video according to the rights of the corresponding member group configured in the filtering rule of the permission mapping table.
  • the background of the image data and the foreground of the moving target are screened.
  • the screening module 20 of the video monitoring device obtains the rights of the corresponding member group from the filtering rules according to the preset permission mapping table, and filters the background and the moving target foreground according to the rights of the corresponding member group. For example, a member group selects a specific person in the video image data, and a member group selects a specific valuable item in the video image data, and the group C group tracks the license plate number of the user in the video image data. Thereby identifying specific people, cars, and objects.
  • FIG. 11 is a schematic structural diagram of the display module of FIG. 7. As shown in FIG. 11, the display module 30 specifically includes:
  • the merging unit 31 is configured to fuse the moving target foreground and the corresponding background that meet the screening condition into a video stream;
  • the sending unit 32 is configured to send the obtained video stream to the video monitor, and the video monitor decodes the video stream and presents the video stream to the interactive interface for the user to perform monitoring and viewing.
  • the fusion unit 31 of the video monitoring device is specifically configured to: fuse the identified background and the moving target foreground to generate a new video stream description XML file based on the IP packet, wherein the principle of the video stream fusion coding is: first obtaining the selected specific motion a video data of the target foreground; and obtaining a background corresponding to the video data of the selected moving target foreground; and then acquiring a background identifier corresponding to the foreground of the moving target, where the background identifier may be an angle identifier of the background corresponding to the foreground of the moving target, such as The angle of the background corresponding to the foreground of a certain moving target is 50 degrees; the last selected motion The target foreground and the corresponding background merge to form a video stream.
  • the video data carried by the video stream only needs to include the video data of the foreground of the specific moving target and the corresponding background image, that is, only one corresponding background image needs to be transmitted in the background of the video stream transmission.
  • the background of different angles it is only necessary to transmit multiple background images of corresponding angles, thereby greatly reducing the amount of information transmission and greatly improving the efficiency of analysis and processing.
  • the sending unit 32 of the video monitoring device is specifically configured to: send the generated video stream to the video monitor, and the video monitor decodes the video stream and presents the video stream to the interactive interface for the user to perform monitoring and viewing.
  • Decoding the video stream includes: first decomposing the merged video stream according to the foreground and the background; and then acquiring the corresponding background image according to the background identifier corresponding to the foreground of the moving target; and then moving the target of the moving target and the corresponding background video data Real-time integration.
  • the video monitoring apparatus provided in this embodiment generates video streams according to analysis and reconstruction of video content and according to customized rights of different types of user groups.
  • the purpose of personalized browsing and privacy protection monitoring of customized content is achieved, and The security of the video surveillance is well ensured, thereby protecting the privacy of the video surveillance content; on the other hand, the actual workload of the monitoring personnel is greatly reduced and the transmission bandwidth is saved without reducing the user's perception.
  • the video monitoring method and device provided by the embodiment of the present invention includes: acquiring video image data and a group in which the member is located; screening the obtained video image data according to the permission mapping table and the obtained group; and screening the obtained video
  • the image data is displayed on the video monitor.
  • the video monitoring content is reconstructed, and the video data required by the user is generated according to the customized correspondence of different types of member groups.
  • the purpose of customized content personalized browsing and privacy protection monitoring is achieved, and the present invention is better.
  • the security of the video surveillance is ensured, thereby protecting the privacy of the video surveillance content; on the other hand, the actual workload of the monitoring personnel is greatly reduced and the transmission bandwidth is saved without reducing the user's perception.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Human Computer Interaction (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • Alarm Systems (AREA)

Abstract

一种视频监控方法及装置，包括获取视频图像数据以及成员所在的组别；根据权限映射表和和获得的组别，对获得的视频图像数据进行筛选；将筛选后得到的视频图像数据显示在视频监控器上。本发明实施例通过对视频监控内容进行重构，并根据不同类型成员组的定制对应生成用户所需的视频数据，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。

Description

一种视频监控方法和装置 技术领域
本发明实施例涉及监控技术，尤指一种视频监控方法和装置。
背景技术
随着视频监控应用的发展与终端设备的普及，面向中小企业及家庭安防的监控系统通过与移动互联网结合得到了极大发展，大量结合网络摄像头和智能手机终端的应用(APP，Application)纷纷推出。
此类应用通常是通过视频监控系统将网络摄像头和智能手机终端连接起来，完成视频内容的采集、监控和管理，从而满足了人们对于安防、管理、亲情和分享的应用需求，但随之也带来一些问题，主要包括：在分享视频或摄像机权限给其他账户时带来的内容隐私保护问题；或者说是否可以有一种方案可以按照不同类型成员定制的需求实现监控视频内容的个性化浏览和管理。因此，如何实现更安全的视频监控，是一个亟待解决的问题。
发明内容
为了解决上述技术问题，本发明实施例提供一种视频监控方法和装置，能够更好地保证视频监控的安全性，从而保护视频监控内容的隐私。
为实现上述目的，本发明实施例提供了一种视频监控方法，包括：
获取视频图像数据以及成员所在的组别；
根据权限映射表和获得的组别，对获得的视频图像数据进行筛选；
将筛选后得到的视频图像数据显示在视频监控器上。
可选地，所述获取视频图像数据之前还包括：预设所述权限映射表；
其中，所述权限映射表的过滤规则中对应配置有各个成员组对应的权限。
可选地，所述获取视频图像数据包括：采集所述视频图像数据；
对所述采集到的视频图像数据进行预处理。
可选地，所述对采集到的视频图像数据进行预处理包括：
提取所述采集到的视频图像数据的背景和运动目标前景；
在视频图像上识别与标注提取的背景和运动目标前景。
可选地地，所述对获得的视频图像数据进行筛选包括：
根据所述权限映射表的过滤规则中配置的对应成员组的权限，对所述视频图像数据的背景和运动目标前景进行筛选。
可选地，所述将筛选后得到的视频图像数据显示在视频监控器上包括：
将所述符合筛选条件的运动目标前景和对应的背景融合编码成视频流；
将所述视频流给视频监控器显示。
本发明实施例还提供了一种视频监控装置，至少包括：
获取模块，设置为获取视频图像数据以及成员所在的组别；
筛选模块，设置为根据权限映射表权限映射表和获得的组别，对获得的视频图像数据进行筛选；
显示模块，设置为将筛选后得到的视频图像数据显示在视频监控器上。
可选地，还包括：
预设模块，设置为预设权限映射表；其中，权限映射表的过滤规则中对应配置有各个成员组对应的权限。
可选地，所述获取模块包括采集单元和预处理单元；其中，
采集单元，设置为采集所述视频图像数据；
预处理单元，设置为对所述采集到的视频图像数据进行预处理。
优选地，所述预处理单元包括至少包括提取子单元和识别子单元；其中，
提取子单元，设置为提取所述采集到的视频图像数据的背景和运动目标前景；
识别子单元，设置为在识别与标注提取的背景和运动目标前景。
优选地，所述筛选模块具体设置为：
根据所述权限映射表的过滤规则中配置的对应成员组的权限，对所述视频图像数据的背景和运动目标前景进行筛选。
优选地，所述显示模块至少包括融合单元和发送单元；其中，
融合单元，设置为将所述符合筛选条件的运动目标前景和对应的背景融合编码成视频流；
发送单元，设置为将得到的视频流发送给视频监控器显示。
本发明实施例提供的视频监控方法，通过获取视频图像数据以及成员所在的组别；根据权限映射表的过滤规则中配置的对应成员组的权限，对所述视频图像数据进行筛选；将筛选后得到的视频图像数据在视频监控器上进行 显示。本发明实施例对视频监控内容进行重构，并根据不同类型成员组的定制对应生成用户所需的视频数据，达到定制内容个性化浏览和隐私保护的目的；大量减轻了监控人员的实际工作量；节约了传输带宽。
附图概述
此处所说明的附图用来提供对本发明的进一步理解，构成本申请的一部分，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：
图1为本发明视频监控方法第一实施例的流程示意图；
图2为本发明视频监控方法第二实施例的流程示意图；
图3为图1中获取视频图像数据的实现流程示意图；
图4为图3中对采集的视频图像数据进行预处理的实现流程示意图；
图5为本发明视频监控方法第三实施例的流程示意图；
图6为图1中将筛选后得到的视频图像数据显示在视频监控器上的实现流程示意图；
图7为本发明视频监控装置第一实施例的组成结构示意图；
图8为本发明视频监控装置第二实施例的组成结构示意图；
图9为图7中获取模块的组成结构示意图；
图10为图9中预处理单元的组成结构示意图；
图11为图7中显示模块的组成结构示意图。
本发明的较佳实施方式
为使本发明实施例的目的、技术方案和优点更加清楚明白，下文中将结合附图对本发明的实施例进行详细说明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互任意组合。
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。
图1为本发明视频监控方法第一实施例的流程示意图，如图1所示，在 第一实施例中，包括以下步骤：
步骤100、获取视频图像数据以及成员所在的组别。
需要说明的是，视频监控系统包括采集前端、业务处理服务器和客户端。
视频监控装置通过采集前端采集视频图像数据，较佳地，采集前端可以采用360度可旋转的夜光摄像头，可以在夜晚的环境下对任意角度的视频图像进行智能采集，进一步地，可以采用光流法或者帧间差分法或者背景差法动态地提取视频图像数据背景以及运动目标前景，并通过脸部特征法进行人脸识别，从而识别出运动目标中特定的人或动物，并对运动目标进一步跟踪，获取特定目标的视频图像数据，并在数据库中进行保存。
本步骤之前，还包括：视频监控装置可以通过WEB或应用程序的客户端管理员账户登录到业务处理服务器，并建立不同成员及成员组，将不同成员划分到对应成员组。这样，本步骤中可以根据当前成员如标识信息，确定其所属的组别。
步骤200、根据权限映射表和获得的组别，对获得的视频图像数据进行筛选。
视频监控装置预先配置权限映射表，并根据权限映射表的过滤规则中所配置的对应成员组的权限，经由业务处理服务器对获得的视频图像进行筛选，例如，可将成员组分为甲成员组、乙成员组、丙成员组，甲成员组设定成只对特定的人进行跟踪，乙成员组设定成只对特定的贵重物品进行跟踪，丙成员组设定成只对用户的车牌号进行跟踪。每个成员组内的成员人数没有限制，成员组内可以没有成员参加，也可以多个成员加入。每个成员经过授权认证后可以参入多个组别，如成员A经过授权认证后，既可以单独加入甲成员组、乙成员组、或者丙成员组，也可以同时加入甲成员组、乙成员组和丙成员组。如果成员A的级别高的话，还可以对权限映射表进行修改，更改所在成员组的权限。
视频监控装置根据权限映射表的过滤规则，对视频图像数据进行筛选，比如成员A划分在甲成员组，那么按照权限映射表就会筛选出相应的特定的人的视频图像数据，从而使得成员A的视频监测权限中，只能监测到相应的特定的人的视频图像数据。
步骤300、将筛选后得到的视频图像数据显示在视频监控器上。
视频监控装置将筛选后的视频图像数据发送给客服端，由客服端对筛选后的视频图像的视频流进行解码后，通过视频监控器的交互界面进行呈现以供用户进行监控查看。
本实施例提供的视频监控方法，通过对视频内容进行重构，并根据不同类型成员组的定制筛选出用户所需的视频数据，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。
图2为本发明视频监控方法第二实施例的流程示意图，如图2所示，在第一实施例的基础上，第二实施例提供的视频监控方法中，步骤100之前还包括：
步骤100A、预设权限映射表，权限映射表的过滤规则中对应配置有各个成员组对应的权限。
其中，权限映射表的过滤规则中对应配置有各个成员组对应的权限。而每个成员可对应多个成员组。从而使得每个成员可对应多个权限，形成多对多的映射。例如甲成员组设定成只对特定的人进行跟踪，乙成员组设定成只对特定的贵重物品进行跟踪，丙成员组设定成只对用户的车牌号进行跟踪。如果A成员对应甲成员组和乙成员组，则可以对特定的人和特定的贵重物品的视频进行监控。并且，级别高的成员可以修改权限映射表，对应修改成员组的权限，例如成员A为最高级别时，并且当前在乙成员组时，可以将乙成员组设定的只对特定的贵重物品进行跟踪的权限修改为对特定的动物进行跟踪的权限。
本实施例提供的视频监控方法，通过预设权限映射表，并对采集的视频图像数据进行筛选，并根据不同类型成员组的定制筛选出用户所需的视频数据，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。
图3为图1中获取视频图像数据的实现示意图，如图3所示，包括：
步骤110、采集视频图像数据。
本步骤中，视频监控装置在接收到来自业务处理服务器发出的控制指令，会通过采集前端对视频数据进行采集，采集前端可以是采用360度可旋转的夜光摄像头，可以在夜晚的环境下对任意角度的视频图像进行智能采集。
步骤120、对采集到的视频图像数据进行预处理。
视频监控装置通过采集前端对采集到的视频图像数据进行预处理，例如，可以采用光流法或者帧间差分法或者背景差法动态地提取视频图像数据背景以及运动目标前景，并通过脸部特征法进行人脸识别，从而识别出运动目标中特定的人或动物，并对运动目标进一步跟踪，获取特定目标的视频图像数据，并在数据库中进行保存。并且针对各个角度的背景进行采集，分别将视频图像数据对应角度的背景存储在数据库中。
图4为图3中对采集的视频图像数据进行预处理的实现流程示意图，如图4所示，包括：
步骤121、提取采集到的视频图像数据的背景和运动目标前景。
视频监控装置利用光流法等运动物体检测的方法，分离出当前视频图像数据的前景和背景，对前景运动目标进行提取，识别运动目标的轮廓。需要注意的是，此处仅是以光流法来举例，当然并不局限于此，还可以使用其他方式，比如帧间差分法或者背景差法等。
步骤122、在视频图像上识别与标注提取的背景和运动目标前景。
视频监控装置识别和提取所述运动目标的主体轮廓，对一个连续运动的运动目标的预览帧序列进行处理；针对预览帧序列，利用目标检测方法，检测可能出现的前景目标；如果某一帧出现了前景目标，找到其具有代表性的关键特征点，其中，关键特征点可以随机产生，也可以利用角点来做关键特征点，角点是主体轮廓上曲率的局部极大点，对掌握主体轮廓特征具有决定作用，一旦找到了运动目标的轮廓特征也就大致掌握了目标的形状。角点特征是影像的重要特征，在各种影像特征中角点具有旋转不变性和不随光照条件改变而改变的优点，在目标识别和目标跟踪中，可以减少参与计算的数据量，同时又不损失图像的重要灰度信息，利用角点特征进行匹配可以大大提高匹配的速度；对之后的任意两个相邻视频帧而言，寻找上一帧中出现的关键特征点在当前帧中的最佳位置，从而得到运动目标在当前帧中的位置坐标；如此迭代进行，便可实现前景的运动目标的跟踪，将获取的前景和背景进行 分离，从而过滤一些物体，筛选出所需的视频图像数据，识别前景中运动目标。
图5为本发明视频监控方法第三实施例的功能模块示意图，如图5所示，在第一实施例或第二实施例的基础上(图5示出了在第二实施例的基础上)，步骤200具体包括：
步骤200A、根据权限映射表的过滤规则中配置的对应成员组的权限，对所述视频图像数据的背景和运动目标前景进行筛选。
视频监控装置根据预设的权限映射表，从过滤规则中获取对应成员组的权限，并根据对应成员组的权限，对提取的背景和运动目标前景进行筛选。例如甲成员组对视频图像数据中特定的人进行筛选，乙成员组对视频图像数据中特定的贵重物品进行筛选，丙成员组对视频图像数据中用户的车牌号进行跟踪。从而识别特定的人、车、物。
图6为图1中将筛选后得到的视频图像数据显示在视频监控器上的实现流程示意图，如图6所示，具体包括：
步骤310、将符合筛选条件的运动目标前景和对应的背景融合编码成视频流。
视频监控装置将识别的背景和运动目标前景进行融合生成基于IP包的新的视频流描述可扩展标识语言(XML，eXtensible Markup Language)文件。其中，视频流融合编码的原则为：首先获取筛选后的特定运动目标前景的视频数据；再获取筛选后的特定运动目标前景的视频数据对应的背景；然后获取与运动目标前景对应的背景标识，所述背景标识可以为运动目标前景对应的背景的角度标识，比如说某一运动目标前景对应的背景的角度标识为50度；最后将筛选后的运动目标前景和对应背景汇合以形成视频流。例如在同一背景下，视频流所携带的视频数据只需要包括特定运动目标前景的视频数据和对应的一张背景图像即可，即视频流传输的背景中只需要传输一张对应背景图像即可。而在不同角度的背景下，也只需传输对应角度的多张背景图像而已，从而大大减小了信息的传输量，并大幅提高了分析处理效率。
步骤320、将得到的视频流发送给视频监控器，由视频监控器对视频流进行解码后呈现到交互界面上，以供用户进行监控查看。
视频监控装置将生成的视频流发送给视频监控器，由所述视频监控器对 所述视频流进行解码后呈现到交互界面上供用户进行监控查看。所述对视频流进行解码包括：首先对汇合的视频流按照前景和背景进行分解；然后根据运动目标前景对应的背景标识，获取对应的背景图像；然后将运动目标前景和对应的背景的视频数据进行实时结合。
本实施例提供的视频监控方法，通过对视频内容的分析和重构，并根据不同类型用户组的定制权限生成视频流，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。
图7为本发明视频监控装置第一实施例的组成结构示意图，如图7所示，在第一实施例中，包括：
获取模块10，设置为获取视频图像数据以及成员所在的组别；
筛选模块20，设置为根据权限映射表和获得的组别，对获得的视频图像数据进行筛选；
显示模块30，设置为将筛选后得到的视频图像数据显示在视频监控器上。
获取模块10，具体设置为：通过采集前端采集视频图像数据，采集前端采用360度可旋转的夜光摄像头，可以在夜晚的环境下对任意角度的视频图像进行智能采集，可以采用光流法或者帧间差分法或者背景差法动态地提取视频图像数据背景以及运动目标前景，并通过脸部特征法进行人脸识别，从而识别出运动目标中特定的人或动物，并对运动目标进一步跟踪，获取特定目标的视频图像数据，并在数据库中进行保存。视频监控装置通过WEB或应用程序的客户端管理员账户登录到业务处理服务器，并建立不同成员及成员组，将不同成员划分到对应成员组。
视频监控装置的筛选模块20具体设置为：预先配置权限映射表，并根据权限映射表的过滤规则中所配置的对应成员组的权限，经由业务处理服务器对所述视频图像进行筛选，例如，可将成员组分为甲成员组、乙成员组、丙成员组，甲成员组设定成只对特定的人进行跟踪，乙成员组设定成只对特定的贵重物品进行跟踪，丙成员组设定成只对用户的车牌号进行跟踪。每个成员组内的成员人数没有限制，成员组内可以没有成员参加，也可以多个成员加入。每个成员经过授权认证后可以参入多个组别，如成员A经过授权认证 后，既可以单独加入甲成员组、乙成员组、或者丙成员组，也可以同时加入甲成员组、乙成员组和丙成员组。如果成员A的级别高的话，还可以对权限映射表进行修改，更改所在成员组的权限。视频监控装置根据权限映射表的过滤规则，对视频图像数据进行筛选，比如将成员A划分到甲成员组，筛选出相应的特定的人的视频图像数据，从而使得成员A的视频监测权限中，只能监测到相应的特定的人的视频图像数据。
视频监控装置的显示模块30具体设置为：将筛选后的视频图像数据发送给客服端，由所述客服端对筛选后的视频图像的视频流进行解码后，通过视频监控器的交互界面进行呈现以供用户进行监控查看。
本实施例提供的视频监控装置，通过对视频内容进行重构，并根据不同类型成员组的定制筛选出用户所需的视频数据，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。
图8为本发明视频监控装置第二实施例的组成结构示意图，如图8所示，在第一实施例的基础上，还包括：
预设模块40，设置为预设权限映射表，其中，权限映射表的过滤规则中对应配置有各个成员组对应的权限。
其中，权限映射表的过滤规则中对应配置有各个成员组对应的权限。而每个成员可对应多个成员组。从而使得每个成员可对应多个权限，形成多对多的映射。例如甲成员组设定成只对特定的人进行跟踪，乙成员组设定成只对特定的贵重物品进行跟踪，丙成员组设定成只对用户的车牌号进行跟踪。如果成员A对应甲成员组和乙成员组，则可以对特定的人和特定的贵重物品的视频进行监控。并且，级别高的成员可以修改权限映射表，对应修改成员组的权限，例如成员A为最高级别时，并且当前在乙成员组时，可以将乙成员组设定的只对特定的贵重物品进行跟踪的权限修改为对特定的动物进行跟踪的权限。
本实施例提供的视频监控装置，通过预设权限映射表，并对采集的视频图像数据进行筛选，并根据不同类型成员组的定制筛选出用户所需的视频数据，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保 证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。
图9为图7中获取模块的组成结构示意图，如图9所示，获取模块10具体包括：
采集单元11，设置为采集视频图像数据；
预处理单元12，设置为对采集到的视频图像数据进行预处理。
视频监控装置的采集单元11在接收到来自业务处理服务器发出的控制指令，通过采集前端对视频数据进行采集，采集前端可以采用360度可旋转的夜光摄像头，可以在夜晚的环境下对任意角度的视频图像进行智能采集。
视频监控装置的预处理单元12通过采集前端对采集到的视频图像数据进行预处理，例如，可以采用光流法或者帧间差分法或者背景差法动态地提取视频图像数据背景以及运动目标前景，并通过脸部特征法进行人脸识别，从而识别出运动目标中特定的人或动物，并对运动目标进一步跟踪，获取特定目标的视频图像数据，并在数据库中进行保存。并且针对各个角度的背景进行采集，分别将视频图像数据对应角度的背景存储在数据库中。
图10为图9中预处理单元的组成结构示意图，如图10所示，预处理单元12具体包括：
提取子单元121，设置为提取采集到的视频图像数据的背景和运动目标前景；
识别子单元122，设置为在视频图像上识别与标注提取的背景和运动目标前景。
视频监控装置的提取子单元121利用光流法等运动物体检测的方法，分离出当前视频图像数据的前景和背景，对前景运动目标进行提取，识别运动目标的轮廓。需要注意的是，此处仅是以光流法来举例，当然并不局限于此，还可以使用其他方式，比如帧间差分法或者背景差法等。
视频监控装置的识别子单元122识别和提取所述运动目标的主体轮廓，对一个连续运动的运动目标的预览帧序列进行处理；针对预览帧序列，利用目标检测方法，检测可能出现的前景目标；如果某一帧出现了前景目标，找到其具有代表性的关键特征点，其中，关键特征点可以随机产生，也可以利 用角点来做关键特征点，角点是主体轮廓上曲率的局部极大点，对掌握主体轮廓特征具有决定作用，一旦找到了运动目标的轮廓特征也就大致掌握了目标的形状。角点特征是影像的重要特征，在各种影像特征中角点具有旋转不变性和不随光照条件改变而改变的优点，在目标识别和目标跟踪中，可以减少参与计算的数据量，同时又不损失图像的重要灰度信息，利用角点特征进行匹配可以大大提高匹配的速度；对之后的任意两个相邻视频帧而言，寻找上一帧中出现的关键特征点在当前帧中的最佳位置，从而得到运动目标在当前帧中的位置坐标；如此迭代进行，便可实现前景的运动目标的跟踪，将获取的前景和背景进行分离，从而过滤一些物体，筛选出所需的视频图像数据，识别前景中运动目标。
进一步参见图7，图7中提供的视频监控装置，在第一实施例的基础上，筛选模块20具体设置为：根据权限映射表的过滤规则中配置的对应成员组的权限，对所述视频图像数据的背景和运动目标前景进行筛选。
视频监控装置的筛选模块20根据预设的权限映射表，从过滤规则中获取对应成员组的权限，并根据对应成员组的权限，对所述的背景和运动目标前景进行筛选。例如甲成员组对视频图像数据中特定的人进行筛选，乙成员组对视频图像数据中特定的贵重物品进行筛选，丙成员组对视频图像数据中用户的车牌号进行跟踪。从而识别特定的人、车、物。
图11为图7中显示模块的组成结构示意图，如图11所示，显示模块30具体包括：
融合单元31，设置为将符合筛选条件的运动目标前景和对应的背景融合编码成视频流；
发送单元32，设置为将得到的视频流发送给视频监控器，由视频监控器对视频流进行解码后呈现到交互界面上，以供用户进行监控查看。
视频监控装置的融合单元31具体设置为：将识别的背景和运动目标前景进行融合生成基于IP包的新的视频流描述XML文件，其中视频流融合编码的原则为：首先获取筛选后的特定运动目标前景的视频数据；再获取筛选后的特定运动目标前景的视频数据对应的背景；然后获取与运动目标前景对应的背景标识，所述背景标识可以为运动目标前景对应的背景的角度标识，比如说某一运动目标前景对应的背景的角度标识为50度；最后将筛选后的运动 目标前景和对应背景汇合以形成视频流。例如在同一背景下，视频流所携带的视频数据只需要包括特定运动目标前景的视频数据和对应的一张背景图像即可，即视频流传输的背景中只需要传输一张对应背景图像即可。而在不同角度的背景下，也只需传输对应角度的多张背景图像而已，从而大大减小了信息的传输量，并可大幅提高分析处理效率。
视频监控装置的发送单元32具体设置为：将生成的视频流发送给视频监控器，由所述视频监控器对所述视频流进行解码后呈现到交互界面上供用户进行监控查看。所述对视频流进行解码包括：首先对汇合的视频流按照前景和背景进行分解；然后根据运动目标前景对应的背景标识，获取对应的背景图像；然后将运动目标前景和对应的背景的视频数据进行实时结合。
本实施例提供的视频监控装置，通过对视频内容的分析和重构，并根据不同类型用户组的定制权限生成视频流，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。
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。
工业实用性
本发明实施例提出的视频监控方法及装置，包括获取视频图像数据以及成员所在的组别；根据权限映射表和和获得的组别，对获得的视频图像数据进行筛选；将筛选后得到的视频图像数据显示在视频监控器上。本发明实施例通过对视频监控内容进行重构，并根据不同类型成员组的定制对应生成用户所需的视频数据，一方面，达到了定制内容个性化浏览和隐私保护监控的目的，更好地保证了视频监控的安全性，从而保护了视频监控内容的隐私；另一方面，在不降低用户观感的情况下，大量减轻了监控人员的实际工作量、节约了传输带宽。

Claims (12)

  1. 一种视频监控方法，其特征在于，包括：
    获取视频图像数据以及成员所在的组别；
    根据权限映射表和获得的组别，对获得的视频图像数据进行筛选；
    将筛选后得到的视频图像数据显示在视频监控器上。
  2. 根据权利要求1所述的视频监控方法，其特征在于，所述获取视频图像数据之前还包括：预设所述权限映射表；
    其中，所述权限映射表的过滤规则中对应配置有各个成员组对应的权限。
  3. 根据权利要求1所述的视频监控方法，其特征在于，所述获取视频图像数据包括：采集所述视频图像数据；
    对采集到的视频图像数据进行预处理。
  4. 根据权利要求3所述的视频监控方法，其特征在于，所述对采集到的视频图像数据进行预处理包括：
    提取所述采集到的视频图像数据的背景和运动目标前景；
    在视频图像上识别与标注提取的背景和运动目标前景。
  5. 根据权利要求1或4所述的视频监控方法，其特征在于，所述对获得的视频图像数据进行筛选包括：
    根据所述权限映射表的过滤规则中配置的对应成员组的权限，对所述视频图像数据的背景和运动目标前景进行筛选。
  6. 根据权利要求5所述的视频监控方法，其特征在于，所述将筛选后得到的视频图像数据显示在视频监控器上包括：
    将所述符合筛选条件的运动目标前景和对应的背景融合编码成视频流；
    将所述视频流发送给视频监控器显示。
  7. 一种视频监控装置，其特征在于，至少包括：
    获取模块，设置为获取视频图像数据以及成员所在的组别；
    筛选模块，设置为根据权限映射表权限映射表和获得的组别，对获得的 视频图像数据进行筛选；
    显示模块，设置为将筛选后得到的视频图像数据显示在视频监控器上。
  8. 根据权利要求7所述的视频监控装置，其特征在于，还包括：
    预设模块，设置为预设权限映射表；其中，所述权限映射表的过滤规则中对应配置有各个成员组对应的权限。
  9. 根据权利要求7所述的视频监控装置，其特征在于，所述获取模块包括采集单元和预处理单元；其中，
    采集单元，设置为采集所述视频图像数据；
    预处理单元，设置为对所述采集到的视频图像数据进行预处理。
  10. 根据权利要求9所述的视频监控装置，其特征在于，所述预处理单元至少包括提取子单元和识别子单元；其中，
    提取子单元，设置为提取所述采集到的视频图像数据的背景和运动目标前景；
    识别子单元，设置为识别与标注提取的背景和运动目标前景。
  11. 根据权利要求7或8所述的视频监控装置，其特征在于，所述筛选模块具体设置为：
    根据所述权限映射表的过滤规则中配置的对应成员组的权限，对所述视频图像数据的背景和运动目标前景进行筛选。
  12. 根据权利要求11所述的视频监控装置，其特征在于，所述显示模块至少包括融合单元和发送单元；其中，
    融合单元，设置为将所述符合筛选条件的运动目标前景和对应的背景融合编码成视频流；
    发送单元，设置为将得到的视频流发送给视频监控器显示。
PCT/CN2015/076309 2014-12-18 2015-04-10 一种视频监控方法和装置 WO2016095374A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US15/537,488 US20180270454A1 (en) 2014-12-18 2015-04-10 Video monitoring method and device
EP15868891.1A EP3220636A4 (en) 2014-12-18 2015-04-10 Video monitoring method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410802364.1 2014-12-18
CN201410802364.1A CN105791747A (zh) 2014-12-18 2014-12-18 视频监控方法和装置

Publications (1)

Publication Number Publication Date
WO2016095374A1 true WO2016095374A1 (zh) 2016-06-23

Family

ID=56125743

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/076309 WO2016095374A1 (zh) 2014-12-18 2015-04-10 一种视频监控方法和装置

Country Status (4)

Country Link
US (1) US20180270454A1 (zh)
EP (1) EP3220636A4 (zh)
CN (1) CN105791747A (zh)
WO (1) WO2016095374A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10762659B2 (en) * 2017-07-10 2020-09-01 Electronics And Telecommunications Research Institute Real time multi-object tracking apparatus and method using global motion
CN112532936A (zh) * 2020-11-25 2021-03-19 华能国际电力股份有限公司日照电厂 一种热力站智能监控方法、系统、设备及存储介质

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108040230B (zh) * 2017-12-19 2020-02-07 司马大大（北京）智能系统有限公司 一种保护隐私的监控方法及装置
JP7030534B2 (ja) * 2018-01-16 2022-03-07 キヤノン株式会社 画像処理装置および画像処理方法
CN110427865B (zh) * 2019-07-29 2023-08-25 三峡大学 高电压禁止区域人类行为视频特征图片提取与重构方法
CN113286170A (zh) * 2020-12-11 2021-08-20 苏州律点信息科技有限公司 一种视频内容分析方法、装置及云服务器
CN113642381A (zh) * 2021-06-10 2021-11-12 临沂边锋自动化设备有限公司 基于大数据的视频监控系统
CN113625603B (zh) * 2021-07-27 2023-06-23 金鹏电子信息机器有限公司 基于大数据的安防监控管理系统及管理方法
CN116362564B (zh) * 2023-04-04 2023-09-08 重庆海辉星科技有限公司 酒店公共卫生视频监管结果的生成方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101778260A (zh) * 2009-12-29 2010-07-14 公安部第三研究所 一种基于结构化描述的视频监控管理方法及系统
CN102340650A (zh) * 2010-07-19 2012-02-01 中国移动通信集团公司 终端视频监控的方法及系统
CN102523433A (zh) * 2011-12-27 2012-06-27 浙江省电力公司 一种视频监控系统与数据采集监控系统联动的设备与方法
CN102547241A (zh) * 2011-12-31 2012-07-04 深圳市永达电子股份有限公司 家庭异常检测方法和装置以及基于无线网络摄像机的家庭异常检测系统
CN103152556A (zh) * 2013-03-15 2013-06-12 云南大学 一种用于隐私保护的监控视频加密方法
CN103841374A (zh) * 2012-11-27 2014-06-04 华为技术有限公司 一种视频监控图像的显示方法及系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7079177B2 (en) * 1995-02-27 2006-07-18 Canon Kabushiki Kaisha Remote control system and access control method for information input apparatus with limitation by user for image access and camemremote control
US7940302B2 (en) * 2004-09-15 2011-05-10 The Regents Of The University Of California Apparatus and method for privacy protection of data collection in pervasive environments
US8576282B2 (en) * 2008-12-12 2013-11-05 Honeywell International Inc. Security system with operator-side privacy zones
CN101964901B (zh) * 2010-10-11 2012-05-30 杭州海康威视数字技术股份有限公司 一种视频监控设备的权限管理方法及设备
WO2013010177A2 (en) * 2011-07-14 2013-01-17 Surfari Inc. Online groups interacting around common content
CN102831166B (zh) * 2012-07-24 2015-05-27 武汉大千信息技术有限公司 一种基于色彩特征检测的刑侦视频预处理方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101778260A (zh) * 2009-12-29 2010-07-14 公安部第三研究所 一种基于结构化描述的视频监控管理方法及系统
CN102340650A (zh) * 2010-07-19 2012-02-01 中国移动通信集团公司 终端视频监控的方法及系统
CN102523433A (zh) * 2011-12-27 2012-06-27 浙江省电力公司 一种视频监控系统与数据采集监控系统联动的设备与方法
CN102547241A (zh) * 2011-12-31 2012-07-04 深圳市永达电子股份有限公司 家庭异常检测方法和装置以及基于无线网络摄像机的家庭异常检测系统
CN103841374A (zh) * 2012-11-27 2014-06-04 华为技术有限公司 一种视频监控图像的显示方法及系统
CN103152556A (zh) * 2013-03-15 2013-06-12 云南大学 一种用于隐私保护的监控视频加密方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3220636A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10762659B2 (en) * 2017-07-10 2020-09-01 Electronics And Telecommunications Research Institute Real time multi-object tracking apparatus and method using global motion
CN112532936A (zh) * 2020-11-25 2021-03-19 华能国际电力股份有限公司日照电厂 一种热力站智能监控方法、系统、设备及存储介质

Also Published As

Publication number Publication date
CN105791747A (zh) 2016-07-20
EP3220636A4 (en) 2017-10-18
EP3220636A1 (en) 2017-09-20
US20180270454A1 (en) 2018-09-20

Similar Documents

Publication Publication Date Title
WO2016095374A1 (zh) 一种视频监控方法和装置
Xiao et al. Video-based evidence analysis and extraction in digital forensic investigation
CN109271554B (zh) 一种智能视频识别系统及其应用
IL256885A (en) Device and methods for facial recognition and video analysis for identifying individuals in the context of video streaming
JP2022048147A5 (zh)
AU2022246412A2 (en) A method and apparatus for conducting surveillance
US11776308B2 (en) Frictionless access control system embodying satellite cameras for facial recognition
CN108364374A (zh) 基于深度学习的人脸门禁控制装置及方法
CN104318217A (zh) 一种基于分布式云计算的人脸识别情报分析方法和系统
CN110705395A (zh) 一种基于树莓派的检测系统
CN112307817B (zh) 人脸活体检测方法、装置、计算设备及计算机存储介质
WO2019234498A1 (en) A network switching appliance, process and system for performing visual analytics for a streamng video
CN102779274A (zh) 一种基于双目摄像头的智能电视人脸识别方法
CN105825568A (zh) 便携式智能交互设备
US20180151039A1 (en) Neighborhood Security Cameras
CN105930780A (zh) 基于近红外和面部微表情的真人活体身份验证方法
CN110852306A (zh) 一种基于人工智能的安全监控系统
Yimyam et al. Video Surveillance System Using IP Camera for Target Person Detection
CN103986882A (zh) 一种实时监控系统中的图像分级﹑传输和处理的方法
KR102091088B1 (ko) Ip 카메라를 이용한 얼굴 영상 획득 방법 및 장치
Chen et al. An active security system based on ar smart classes and face recognition technology
CN207020684U (zh) 基于人脸识别的便携式警用单兵套装
CN103345625B (zh) 一种三维图像分析方法及系统
US10395137B2 (en) Method and processing unit for correlating image data content from disparate sources
TW201508651A (zh) 人臉偵測之雲端智慧監視系統

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: 15868891

Country of ref document: EP

Kind code of ref document: A1

REEP Request for entry into the european phase

Ref document number: 2015868891

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 15537488

Country of ref document: US