CN108076311A - Personnel based on depth camera, which fall down to the ground, monitors identifying system - Google Patents
Personnel based on depth camera, which fall down to the ground, monitors identifying system Download PDFInfo
- Publication number
- CN108076311A CN108076311A CN201611001975.1A CN201611001975A CN108076311A CN 108076311 A CN108076311 A CN 108076311A CN 201611001975 A CN201611001975 A CN 201611001975A CN 108076311 A CN108076311 A CN 108076311A
- Authority
- CN
- China
- Prior art keywords
- video
- function
- ground
- module
- real
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
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
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Alarm Systems (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
A kind of personnel based on depth camera, which fall down to the ground, monitors identifying system, and including front-end camera sensor, center processing server is further included with lower module：(1) vision operation module：Vision operation function is provided, including real-time video preview, video taking turn prediction scheme, history video frequency searching, camera control and video export function；(2) good application module：Offer personnel fall down to the ground the real-time dynamic posture detection function of abnormal behaviour；(3) system function module：It provides necessary including alarm event information retrieval, system equipment monitoring running state, device configuration and the system functions such as user authority management and log management；(4) data memory module：System sets special central database to carry out the storage of Various types of data, and video, system information data are exported including necessary business datum, the image data downloaded, alarm.
Description
Technical field
The invention belongs to technical field of video monitoring, are related to a kind of video analysis processing platform based on depth camera,
Particular for prison institute application environment, make computer based on indoor three-dimensional information it will be appreciated that with the personnel's under analysis complex scene
Behavior and action.
Background technology
In current prison institute monitor mode, the application of video monitoring becomes increasingly popular, but most of is all conventional video prison
Control, checks real-time video by way of manual retrieval, and inefficiency and easily fatigue supervise effect with the growth of event
Sharp-decay.In key monitoring region, some places have started using intelligent video mode, and common intelligent monitoring algorithm is based on
Two-dimensional video image is analyzed and processed, and be there is a situation where wrong report, is failed to report, and causes people's police on duty hard to tell whether it is true or false in face of alert, visitor
The pressure of people's police is added in sight.Two dimensional image analysis mainly has following shortcoming：
Conventional two-dimensional image analysis judges abnormal behaviour by the severe degree changed, when scene changes complexity is to trigger greatly
Amount wrong report；
Conventional two-dimensional image is analyzed carries out target detection by the pattern-recognition of personnel characteristics, lacks depth data, simultaneously
Influenced by shadow and blocking big, detection result is bad in crowd scene；
Conventional two-dimensional image analysis is based on two-dimensional image pixel coordinate when defining virtual perimeter protection, when object is in image
When entering region into setting regions in coordinate but not in real space, a large amount of wrong reports can be caused；
Particularly with night, illumination is unknown etc., scenes, the analysis of bipartite graph picture are difficult to accomplish Real-time Alarm.
The content of the invention
It falls down to the ground it is an object of the invention to provide a kind of personnel based on depth camera and monitors identifying system, which can
Complete identification alarm system is established in the form of by ad hoc network, while the api interface of full standard and prison institute are also provided
Existing security protection video monitoring system or operation platform docking, realize alarm linkage.
Technical scheme is as follows：
A kind of personnel based on depth camera fall down to the ground monitoring identifying system, including front-end camera sensor, at center
Manage server, it is characterised in that further include with lower module：
(1) vision operation module：Vision operation function is provided, is regarded including real-time video preview, video taking turn prediction scheme, history
Frequency retrieval, camera control and video export function, wherein, real-time video preview is deployed troops on garrison duty in real time including the more videos of multiwindow, branch
The picture composition mode for holding 1,4,9,16 picture plurality of specifications is played in real time；Video taking turn prediction scheme provides taking turn in group, divides
It organizes taking turn, timesharing taking turn and combination taking turn mode and combines video camera distribution situation in key area and formulate Video emergency side
Case, during emergency case can a key transfer urgent prediction scheme browsing video；History video frequency searching, provide conventional history video playing and
F.F., pause, slow play, sectional drawing function and can be real with the screens switch point position when history video is watched in some monitoring point
When video；Camera control can be supported to carry out the controllable video camera of access diminution amplification, cloud platform rotation, presetting bit setting etc.
Control operation；Video export function can export history video according to specified time section and be stored in specified file road
Under footpath, and the particular alarm video for the alarm that falls down to the ground is triggered, system can be stored automatically；
(2) good application module：Offer personnel fall down to the ground the real-time dynamic posture detection function of abnormal behaviour, in depth camera
In the range of machine monitoring, when system by into personnel three-dimensional information carry out analysis identify fall down to the ground behavior when, automatic trigger
Warning function carries out Realtime Alerts in system front end user interface, and the common controllable cameras capture for the access system that links shows
Field video, while the video segment of triggering alarm is automatically saved；
(3) system function module：There is provided it is necessary including alarm event information retrieval, system equipment monitoring running state,
The system functions such as device configuration and user authority management and log management；
(4) data memory module：System sets the storage of special central database progress Various types of data, including necessary
Business datum, the image data downloaded, alarm export video, system information data.
The analyzing and processing of depth camera is to increase one-dimensional depth information on the basis of two-dimensional image information so that is calculated
Machine can based on three-dimensional information it will be appreciated that and people under analysis complex scene action and behavior, track to people, action and
So as to accurately judge anomalous event, and Realtime Alerts, the monitoring efficiency of key area is substantially improved in the analysis of time-space relationship.This
Invention is had the following advantages by introducing depth camera：
Precisely search：Structuring risk case information, afterwards can by precisely search for realization event it is quick search with
Positioning；
Intelligent taking turn：Real time propelling movement anomalous event carries out Display on the same screen in a manner of intelligent taking turn, avoids largely useless regard
Frequency information is disturbed, and the identified probability of hazard event has been significantly greatly increased；
Open platform：Event support is exported in a manner of API, can be directly displayed in the system, also access client is supported to show
There is platform, personnel's trace information secondary development customized event also can be used.
The present invention is by introducing depth camera, and compared to conventional two-dimensional image intellectual analysis means, that detects is accurate
Property will greatly promote so that energy can be placed in the place paid close attention to without omitting critical event by staff, be carried
Reduce operating pressure while high workload efficiency, and the application of depth information, it will also bring new technological innovation and application wound
Newly.
Description of the drawings
Fig. 1 is the system logic structure schematic diagram of the present invention.
Specific embodiment
As shown in Figure 1, the personnel based on depth camera fall down to the ground, the system logic structure of monitoring identifying system is divided into four layers,
Wherein：Access layer is responsible for the access of equipment and data, and network layer is responsible for the transmission of data, and service layer is responsible for the deployment of all kinds of services
With the storage of data, application layer provides types of applications operation to the user.
Personnel based on depth camera fall down to the ground monitoring identifying system by the three of depth camera sensor acquisition human body
Dimension space data so as to further identify the body work posture of people, further accurately detect the abnormal behaviours such as fall down to the ground.
System needs user to manually set concern period startup identification function, is specifying in section, the depth information that system passes through acquisition
The limbs state of automatic testing staff, while be compared with the state of ground in the initial scene of system, once there is personnel's limbs appearance
State exception or limbs with ground is improper contacts, system can carry out risk to report prompting at the first time.
Personnel based on depth camera, which fall down to the ground, monitors identifying system, including front-end camera sensor, center processing clothes
Business device equipment further includes vision operation, good application, system function and data and stores four modules.
(1) vision operation module：Vision operation function is provided, is regarded including real-time video preview, video taking turn prediction scheme, history
Frequency retrieval, camera control and video export function.
Real-time video preview, in order to enhance the universality of the system, system supports real-time video preview function, including：It is more
The more videos of window are deployed troops on garrison duty in real time, support that the picture composition mode of 1,4,9,16 picture plurality of specifications is played in real time.
Video taking turn prediction scheme provides taking turn in group, grouping taking turn, timesharing taking turn and combination taking turn mode and in emphasis
Region combines video camera distribution situation and formulates Video emergency scheme, during emergency case can a key transfer urgent prediction scheme browsing and regard
Frequently.
History video frequency searching provides conventional history video playing and F.F., pause, slow play, sectional drawing function and at some
It monitoring point can be with the screens switch point position real-time video when watching history video.
Camera control can be supported to carry out the controllable video camera of access diminution amplification, cloud platform rotation, presetting bit setting
Etc. control operations.
Video export function can export history video according to specified time section and be stored in specified file path
Under, and the particular alarm video for the alarm that falls down to the ground is triggered, system can be stored automatically.
(2) good application module：
Offer personnel fall down to the ground the real-time dynamic posture detection function of abnormal behaviour, in depth camera monitoring range, when
System by into personnel three-dimensional information carry out analysis identify fall down to the ground behavior when, automatic trigger warning function, in system
Realtime Alerts are carried out in front-end user interface, the common controllable cameras capture live video for the access system that links, while to touching
The video segment of hair alarm is automatically saved.
In addition, system is supported to carrying out target trajectory tracking into the personnel in depth camera monitoring range, number is united
The functions such as meter further support the good application of system.
(3) system function module：
It provides necessary including alarm event information retrieval, system equipment monitoring running state, device configuration and user power
The system functions such as limit management and log management.
(4) data memory module：
System sets special central database to carry out the storage of Various types of data, including necessary business datum, downloads
Image data, alarm export video, system information data etc..
The present invention in application, by be deployed in prison in depth camera by network insertion central server, in it is genuinely convinced
Business program obtains depth data by unified API, obtains abnormal behaviour recognition result, as a result, should to upper strata according to differentiation
With good application function is provided, application services is deployed in central server, and user interaction software is by calling service to carry
All kinds of interfaces supplied realize types of functionality, and the split open functional interface of simultaneity factor supports the access with existing monitoring device,
Such as NVR, DVR provide existing intelligent system for monitoring function access and support, coarseness, loose couplings so as to fulfill system
With it is service-oriented so that system is easy to maintain, High Availabitity, retractility are good.
Claims (1)
- The monitoring identifying system 1. a kind of personnel based on depth camera fall down to the ground, including front-end camera sensor, center processing Server, it is characterised in that further include with lower module：(1) vision operation module：Vision operation function is provided, including real-time video preview, video taking turn prediction scheme, the inspection of history video Rope, camera control and video export function, wherein, real-time video preview is deployed troops on garrison duty in real time including the more videos of multiwindow, support 1, 4th, the picture composition mode of 9,16 picture plurality of specifications is played in real time；Video taking turn prediction scheme provides taking turn in group, grouping wheel Patrol, timesharing taking turn and combination taking turn mode and key area combine video camera distribution situation formulate Video emergency scheme, dash forward During heat condition can a key transfer urgent prediction scheme browsing video；History video frequency searching, provide conventional history video playing and F.F., Pause, slow play, sectional drawing function and some monitoring point watch history video when can in real time be regarded the point position with screens switch Frequently；Camera control can be supported to carry out the controllable video camera of access the controls such as diminution amplification, cloud platform rotation, presetting bit setting Operation；Video export function can export history video according to specified time section and is stored under specified file path, And the particular alarm video for the alarm that falls down to the ground is triggered, system can be stored automatically；(2) good application module：Offer personnel fall down to the ground the real-time dynamic posture detection function of abnormal behaviour, are supervised in depth camera In the range of control, when system to carrying out analysis into the three-dimensional information of personnel by identifying that automatic trigger is alarmed when falling down to the ground behavior Function carries out Realtime Alerts in system front end user interface, and the common controllable cameras capture scene for the access system that links regards Frequently, while to the video segment of triggering alarm automatically save；(3) system function module：It provides necessary including alarm event information retrieval, system equipment monitoring running state, equipment Configuration and user authority management and log management function；(4) data memory module：System sets the storage of special central database progress Various types of data, including necessary business Data, the image data downloaded, alarm export video, system information data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611001975.1A CN108076311A (en) | 2016-11-14 | 2016-11-14 | Personnel based on depth camera, which fall down to the ground, monitors identifying system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611001975.1A CN108076311A (en) | 2016-11-14 | 2016-11-14 | Personnel based on depth camera, which fall down to the ground, monitors identifying system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108076311A true CN108076311A (en) | 2018-05-25 |
Family
ID=62161913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611001975.1A Pending CN108076311A (en) | 2016-11-14 | 2016-11-14 | Personnel based on depth camera, which fall down to the ground, monitors identifying system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108076311A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881842A (en) * | 2018-07-05 | 2018-11-23 | 盎锐（上海）信息科技有限公司 | Monitoring system and information processing method based on 3D video camera |
-
2016
- 2016-11-14 CN CN201611001975.1A patent/CN108076311A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881842A (en) * | 2018-07-05 | 2018-11-23 | 盎锐（上海）信息科技有限公司 | Monitoring system and information processing method based on 3D video camera |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108073858A (en) | Crowd massing monitoring identifying system based on depth camera | |
| US7633520B2 (en) | Method and apparatus for providing a scalable multi-camera distributed video processing and visualization surveillance system | |
| US8289390B2 (en) | Method and apparatus for total situational awareness and monitoring | |
| CN108074363A (en) | Perimeter protection monitoring identifying system based on depth camera | |
| CN107018362B (en) | Anti-theft hunting monitoring method and system | |
| US7479980B2 (en) | Monitoring system | |
| CN106331657A (en) | Video analysis and detection method and system for crowd gathering and moving | |
| CN106781165A (en) | A kind of indoor multi-cam intelligent linkage supervising device based on depth sensing | |
| CN106408833A (en) | Perimeter intrusion detection method and system | |
| US20070058842A1 (en) | Storage of video analysis data for real-time alerting and forensic analysis | |
| CN111222373B (en) | Personnel behavior analysis method and device and electronic equipment | |
| AU2011352414A1 (en) | Inference engine for video analytics metadata-based event detection and forensic search | |
| CN104574722A (en) | Harbor safety control system based on multiple sensors | |
| GB2450478A (en) | A security device and system | |
| CN106303469A (en) | Video analysis detection method and system to indoor and outdoor surroundings Flame | |
| KR102149832B1 (en) | Automated Violence Detecting System based on Deep Learning | |
| CN106412522A (en) | Video analysis detection method and system of object in indoor and outdoor environment | |
| CN102270375A (en) | Time based visual review of multi-polar incidents | |
| US20140240455A1 (en) | System and Method to Create Evidence of an Incident in Video Surveillance System | |
| CN111010547A (en) | Target object tracking method and device, storage medium and electronic device | |
| KR20140141025A (en) | Video Searching Apparatus and Method | |
| JP4578044B2 (en) | Image data processing | |
| KR20160093253A (en) | Video based abnormal flow detection method and system | |
| CA2393932C (en) | Human object surveillance using size, shape, movement | |
| CN108076311A (en) | Personnel based on depth camera, which fall down to the ground, monitors identifying system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180525 |