CN110459027A - A kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion - Google Patents
A kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion Download PDFInfo
- Publication number
- CN110459027A CN110459027A CN201910755673.0A CN201910755673A CN110459027A CN 110459027 A CN110459027 A CN 110459027A CN 201910755673 A CN201910755673 A CN 201910755673A CN 110459027 A CN110459027 A CN 110459027A
- Authority
- CN
- China
- Prior art keywords
- target
- people
- heterogeneous data
- data fusion
- model
- 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
-
- 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
-
- 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
- G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
- G08B13/19613—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
- G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0438—Sensor means for detecting
- G08B21/0476—Cameras to detect unsafe condition, e.g. video cameras
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Psychiatry (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Alarm Systems (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to community security protection technical fields, specifically disclose a kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion, comprising: obtain monitoring camera image；Target following model, Analysis model of network behaviors and number of people detection model are input an image into respectively；The positioning of target is obtained, while carrying out behavioural analysis, behavioural information is obtained, the number of people number in image is counted, the coordinate of each target is obtained；The early warning for carrying out target tracking, carrying out abnormal behaviour carries out number of people distance and calculates；Demographics are carried out when number of people distance is less than distance threshold；Early warning of assembling a crowd is carried out when the number of statistics is more than number threshold value；Fusion infrared sensor passes through record, pedestrian detection information, target following, comprehensive descision illegal invasion；Merge target following and behavioral value as a result, carrying out key monitoring.
Description
Technical field
The present invention relates to community security protection technical field, specifically a kind of Community Safety based on multi-source heterogeneous data fusion is anti-
Maintaining method and system.
Background technique
No matter Community Safety protection, be all the module most to merit attention in traditional community or intelligence community.One community
Quality, always be important reference point safely.The development of multi-source heterogeneous data fusion and deep learning, can solve wisdom
Many problems of community's automatic safeguarding.
Convolutional neural networks are a kind of methods of deep learning, and using it, in fields such as image procossings, outstanding performance is combined
Traditional algorithm largely reduces manual intervention, and training high quality model is for improving accuracy rate.
Multi-source heterogeneous data fusion is used for military field earliest, but its thought can be used for solving the anti-protector for collar of Community Safety
Domain.The data of multisensor are handled respectively, then by treated, data utilize blending algorithm comprehensive descision, and it is automatic to reach community
Change the purpose of security protection.
The aspect of Community Safety protection at present, mostly uses the mode of traditional monitoring device, sensor and safe manual patrol, respectively
The data of a equipment are not effectively combined, and are formed data silo one by one, can not be automated to form security protection system, manually
Cost is big, and can not find the generation of Community Safety event in time.In consideration of it, the present invention proposes a kind of multi-source heterogeneous data fusion
Community Safety means of defence.
Summary of the invention
The purpose of the present invention is to provide a kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion,
To solve the problems mentioned in the above background technology.
To achieve the above object, the embodiment of the invention provides a kind of safety of community based on multi-source heterogeneous data fusion
Means of defence, the safety protecting method specifically includes the following steps:
S10, monitoring camera image is obtained；
S20, target following model, Analysis model of network behaviors and number of people detection model are input an image into respectively；
S30, the positioning for obtaining target, while behavioral value is carried out, behavioural information is obtained, is united to the number of people number in image
Count and obtain the coordinate of each target；
S40, the prompt for carrying out target tracking, carrying out abnormal behaviour carry out number of people distance and calculate；
S50, demographics are carried out when number of people distance is less than distance threshold；
S60, it carries out assembling a crowd to prompt when the number of statistics is more than number threshold value；
S70, fusion infrared sensor pass through record, the information of behavioral value, target following, comprehensive descision illegal invasion；
Infrared sensor is placed in community's fence and building periphery more, and bonding behavior analysis and target tracking reduce wrong report, improve into
Invade the accuracy rate of detection；
S80, fusion target following and behavioral value as a result, carry out key monitoring；
Target following is combined with behavioral value, behavioural information is assigned for target, targetedly carries out key monitoring.
As a further solution of the present invention: in step slo, the video of monitoring camera is constantly pulled by RTSP
Frame, picture are pre-processed accordingly before entering concrete model.
As a further solution of the present invention: in step S20, same target is used in the training of target following model
A pair of of image template is separately input in convolutional neural networks, and training obtains target following model；
It is trained using longer video by LTC3D network in the training of Analysis model of network behaviors, obtains Analysis model of network behaviors；
Number of people detection model is trained using the SSD network of default parameters, is constantly adjusted, is obtained to parameter in training
Effect best's head detection model.
As a further solution of the present invention: in step s 30,5 kinds of abnormal behaviours be respectively run, jump, falling, bending over and
It is armed, when detecting any one in this 5 kinds of abnormal behaviours, it can all carry out early warning.
As a further solution of the present invention: in step s 30, all numbers of people detected being demarcated, mould is utilized
The central point of the boundingbox confirmation target of type output；
In step s 40, using the coordinate of all number of people central points, the distance between all the points are calculated.
As a further solution of the present invention: in step s 50, distance threshold refers to the reference mark of the distance between two o'clock
It is quasi-；
Then meet the characteristic that is mutually related when the central point of two numbers of people is less than distance threshold, to meeting the characteristic that is mutually related
Number counted.
As a further solution of the present invention: in step S60, number threshold value refers to the number setting for being judged to assembling a crowd, when
It is judged to assembling a crowd when more than number threshold value, carries out warning information and report.
The embodiment of the invention also provides a kind of security protection system of community based on multi-source heterogeneous data fusion, packets
It includes:
A pair of of image is obtained Feature Dimension Reduction mapping by 2 CNN networks with same structure, then by target tracking module
Correlation calculations are realized by convolution, obtain corresponding target position, are returned original image size in final corresponding figure interpolation and are realized mesh
Target positioning, realizes multi-target detection, tracking and the classification of people and vehicle；
Unusual checking module, the network structure of the module contain 5 3D convolutional layers, and every layer all includes 64,128,256,
256,256 convolution kernels finally include 3 layers of full articulamentum, and convolution kernel size is 3*3*3, and each layer of convolutional layer all follows relu
Activation and maximum pond layer, the size of pond layer are 2x2x1 in addition to first layer, and remainder layer size is all 2x2x2, are connected entirely for first two layers
Dropout is all followed after connecing layer；Classify to race, the abnormal behaviours such as jump, fall, bend over, time longer video is passed through
LTC3D network is trained, and obtains identification model, realizes identification, the identification of human body abnormal operation of moving target abnormal behaviour, real
It now gives warning in advance to hazardous act, what discovery in time needed to look after group the abnormalities such as falls down, cries for help；
Detection module of assembling a crowd all is identified the number of people in image using number of people detection model, and saves each number of people
Coordinate calculates each number of people and is counted in the distance of other numbers of people when distance is less than the distance threshold of setting；Work as the number of people
Quantity carries out early warning of assembling a crowd when being more than the number threshold value of setting；
Multi-source heterogeneous data fusion module, after infrared sensor information, target following information, behavioral value information are handled
Fusion association is carried out, whether comprehensive descision forms invasion；Emphasis prison can be carried out for the result of target following and behavioral value
Control.
Compared with prior art, the embodiment of the present invention, which is utilized, provides automation height based on multi-source heterogeneous data fusion for community
The security protection of precision, comprising: obtain monitoring camera image；Target following model, behavioural analysis are input an image into respectively
Model and number of people detection model；The positioning of target is obtained, while carrying out behavioural analysis, behavioural information is obtained, to the people in image
Head number is counted, and the coordinate of each target is obtained；Carry out target tracking, carry out abnormal behaviour early warning, carry out the number of people away from
From calculating；Demographics are carried out when number of people distance is less than distance threshold；Gathered when the number of statistics is more than number threshold value
Many early warning；Fusion infrared sensor passes through record, pedestrian detection information, target following, comprehensive descision illegal invasion；Merge mesh
Mark tracking is with behavioral value as a result, carrying out key monitoring.
In conclusion the embodiment of the present invention is by multi-source heterogeneous Data fusion technique to being related to Community Safety guard technology
Target following, behavioral value and infrared camera data are merged, comprehensive descision illegal invasion behavior.Target following and different
Normal behavioral value mutually merges, and carries out key monitoring.It is greatly improved by fused accuracy rate.The embodiment of the present invention subtracts significantly
The workload of few Security Personnel, precisely detects the generation of Community Safety event, automatic high-efficiency security protection early warning.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment.
Fig. 1 is the process of the Community Safety means of defence provided in an embodiment of the present invention based on multi-source heterogeneous data fusion
Figure.
Fig. 2 is the structure of the Community Safety guard system provided in an embodiment of the present invention based on multi-source heterogeneous data fusion
Figure.
Specific embodiment
In order to which technical problems, technical solutions and advantages to be solved are more clearly understood, tie below
Accompanying drawings and embodiments are closed, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only
To explain the present invention, it is not intended to limit the present invention.
As shown in Figure 1, in embodiment provided by the invention, a kind of safety of the community based on multi-source heterogeneous data fusion
Means of defence, the safety protecting method specifically includes the following steps:
S10, monitoring camera image is obtained；Specifically, constantly pull the video frame of monitoring camera by RTSP, picture into
It is pre-processed accordingly before entering concrete model.
S20, target following model, Analysis model of network behaviors and number of people detection model are input an image into respectively；Specifically, mesh
It marks and is separately input in convolutional neural networks in the training of trace model using a pair of of image template of same target, training obtains
Target following model.
Wherein, it is trained using longer video by LTC3D network in the training of Analysis model of network behaviors, obtains behavior
Analysis model.In addition, number of people detection model, is trained using the SSD network of default parameters, parameter is constantly carried out in training
Adjustment obtains effect best's head detection model.
S30, the positioning for obtaining target, while carrying out behavioral value obtain behavioural information, to the number of people number in image into
Row counts and obtains the coordinate of each target；
Specifically, behavior is divided into race, jump, falls, bend over and armed, 5 kinds of abnormal behaviours, this 5 kinds of abnormal behaviours are being detected
In any one when, can all carry out early warning.All numbers of people detected are demarcated, model output is utilized
The central point of boundingbox confirmation target.
S40, the prompt for carrying out target tracking, carrying out abnormal behaviour carry out number of people distance and calculate；Specifically, using all
The coordinate of number of people central point calculates the distance between all the points.
S50, demographics are carried out when number of people distance is less than distance threshold；Specifically, distance threshold refers between two o'clock
The reference standard of distance then meets the characteristic that is mutually related when the central point of two numbers of people is less than distance threshold, to meeting phase
The number of the characteristic of mutual correlation is counted.
S60, it carries out assembling a crowd to prompt when the number of statistics is more than number threshold value；It is judged to gathering specifically, number threshold value refers to
Many number settings, are judged to assembling a crowd when being more than number threshold value, carry out warning information and report.
S70, fusion infrared sensor pass through record, the information of behavioral value, target following, and comprehensive descision illegally enters
It invades；
Infrared sensor is placed in community's fence and building periphery more, and bonding behavior analysis and target tracking reduce wrong report, improve into
Invade the accuracy rate of detection.
S80, fusion target following and behavioral value as a result, carry out key monitoring；
Target following is combined with behavioral value, behavioural information is assigned for target, targetedly carries out key monitoring.
As shown in Fig. 2, the safety protecting method based on above-mentioned intelligence community, the embodiment of the invention also provides one kind to be based on
The security protection system of the community of multi-source heterogeneous data fusion, comprising: target tracking module, is assembled a crowd at unusual checking module
Detection module and multi-source heterogeneous data fusion module.Wherein:
A pair of of image is obtained Feature Dimension Reduction mapping by 2 CNN networks with same structure, then by target tracking module
Correlation calculations are realized by convolution, obtain corresponding target position, are returned original image size in final corresponding figure interpolation and are realized mesh
Target positioning, realizes multi-target detection, tracking and the classification of people and vehicle；
Unusual checking module, the network structure of the module contain 5 3D convolutional layers, and every layer all includes 64,128,256,
256,256 convolution kernels finally include 3 layers of full articulamentum, and convolution kernel size is 3*3*3, and each layer of convolutional layer all follows relu
Activation and maximum pond layer, the size of pond layer are 2x2x1 in addition to first layer, and remainder layer size is all 2x2x2, are connected entirely for first two layers
Dropout is all followed after connecing layer.Classify to race, the abnormal behaviours such as jump, fall, bend over, time longer video is passed through
LTC3D network is trained, and obtains identification model, realizes identification, the identification of human body abnormal operation of moving target abnormal behaviour, real
It now gives warning in advance to hazardous act, what discovery in time needed to look after group the abnormalities such as falls down, cries for help.
Detection module of assembling a crowd all is identified the number of people in image using number of people detection model, and saves everyone
The coordinate of head calculates each number of people and is counted in the distance of other numbers of people when distance is less than the distance threshold of setting.When
Number of people quantity carries out early warning of assembling a crowd when being more than the number threshold value of setting.
Multi-source heterogeneous data fusion module, at infrared sensor information, target following information, behavioral value information
Fusion association is carried out after reason, whether comprehensive descision forms invasion.Weight can be carried out for the result of target following and behavioral value
Point monitoring.
The present invention, which is utilized, provides the security protection of automated high-precision based on multi-source heterogeneous data fusion for community, comprising:
Obtain monitoring camera image；Target following model, Analysis model of network behaviors and number of people detection model are input an image into respectively；
The positioning of target is obtained, while carrying out behavioural analysis, behavioural information is obtained, the number of people number in image is counted simultaneously
It is counted, obtains the coordinate of each target；The early warning for carrying out target tracking, carrying out abnormal behaviour carries out number of people distance and calculates；
Demographics are carried out when number of people distance is less than distance threshold；Assemble a crowd when the number of statistics is more than number threshold value pre-
It is alert；Fusion infrared sensor passes through record, pedestrian detection information, target following, comprehensive descision illegal invasion；Merge target with
Track and behavioral value as a result, carrying out key monitoring.
In conclusion the embodiment of the present invention is by multi-source heterogeneous Data fusion technique to being related to Community Safety guard technology
Target following, behavioral value and infrared camera data are merged, comprehensive descision illegal invasion behavior.Target following and different
Normal behavioral value mutually merges, and carries out key monitoring.It is greatly improved by fused accuracy rate.The embodiment of the present invention subtracts significantly
The workload of few Security Personnel, precisely detects the generation of Community Safety event, automatic high-efficiency security protection early warning.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of safety protecting method of the community based on multi-source heterogeneous data fusion, which is characterized in that the safety protecting method
Specifically includes the following steps:
S10, monitoring camera image is obtained；
S20, target following model, Analysis model of network behaviors and number of people detection model are input an image into respectively；
S30, the positioning for obtaining target, while behavioral value is carried out, behavioural information is obtained, is united to the number of people number in image
Count and obtain the coordinate of each target；
S40, the prompt for carrying out target tracking, carrying out abnormal behaviour carry out number of people distance and calculate；
S50, demographics are carried out when number of people distance is less than distance threshold；
S60, it carries out assembling a crowd to prompt when the number of statistics is more than number threshold value；
S70, fusion infrared sensor pass through record, the information of behavioral value, target following, comprehensive descision illegal invasion；
Infrared sensor is placed in community's fence and building periphery more, and bonding behavior analysis and target tracking reduce wrong report, improve into
Invade the accuracy rate of detection；
S80, fusion target following and behavioral value as a result, carry out key monitoring；
Target following is combined with behavioral value, behavioural information is assigned for target, targetedly carries out key monitoring.
2. the Community Safety means of defence according to claim 1 based on multi-source heterogeneous data fusion, which is characterized in that In
In step S10, the video frame of monitoring camera is constantly pulled by RTSP, picture carries out corresponding pre- before entering concrete model
Processing.
3. the Community Safety means of defence according to claim 2 based on multi-source heterogeneous data fusion, which is characterized in that In
In step S20, convolutional Neural net is separately input to using a pair of of image template of same target in the training of target following model
In network, training obtains target following model.
4. the Community Safety means of defence according to claim 3 based on multi-source heterogeneous data fusion, which is characterized in that In
It in step S20, is trained using longer video by LTC3D network in the training of Analysis model of network behaviors, obtains behavior point
Analyse model.
5. the Community Safety means of defence according to claim 3 or 4 based on multi-source heterogeneous data fusion, feature exist
In, in step S20, number of people detection model is trained using the SSD network of default parameters, training in parameter constantly into
Row adjustment obtains effect best's head detection model.
6. the Community Safety means of defence according to claim 5 based on multi-source heterogeneous data fusion, which is characterized in that In
In step S30,5 kinds of abnormal behaviours be respectively run, jump, falling, bend over it is armed, detecting appointing in this 5 kinds of abnormal behaviours
When anticipating a kind of, early warning can be all carried out.
7. the Community Safety means of defence according to claim 3 based on multi-source heterogeneous data fusion, which is characterized in that In
In step S30, all numbers of people detected are demarcated, utilize the center of the boundingbox confirmation target of model output
Point；
In step s 40, using the coordinate of all number of people central points, the distance between all the points are calculated.
8. the Community Safety means of defence according to claim 7 based on multi-source heterogeneous data fusion, which is characterized in that In
In step S50, distance threshold refers to the reference standard of the distance between two o'clock；
Then meet the characteristic that is mutually related when the central point of two numbers of people is less than distance threshold, to meeting the characteristic that is mutually related
Number counted.
9. the Community Safety means of defence according to claim 8 based on multi-source heterogeneous data fusion, which is characterized in that In
In step S60, number threshold value refers to the number setting for being judged to assembling a crowd, and is judged to assembling a crowd when being more than number threshold value, carries out early warning
Information reporting.
10. a kind of security protection system of the community based on multi-source heterogeneous data fusion characterized by comprising
A pair of of image is obtained Feature Dimension Reduction mapping by 2 CNN networks with same structure, then by target tracking module
Correlation calculations are realized by convolution, obtain corresponding target position, are returned original image size in final corresponding figure interpolation and are realized mesh
Target positioning, realizes multi-target detection, tracking and the classification of people and vehicle；
Unusual checking module, the network structure of the module contain 5 3D convolutional layers, and every layer all includes 64,128,256,
256,256 convolution kernels finally include 3 layers of full articulamentum, and convolution kernel size is 3*3*3, and each layer of convolutional layer all follows relu
Activation and maximum pond layer, the size of pond layer are 2x2x1 in addition to first layer, and remainder layer size is all 2x2x2, are connected entirely for first two layers
Dropout is all followed after connecing layer；Classify to race, the abnormal behaviours such as jump, fall, bend over, time longer video is passed through
LTC3D network is trained, and obtains identification model, realizes identification, the identification of human body abnormal operation of moving target abnormal behaviour, real
It now gives warning in advance to hazardous act, what discovery in time needed to look after group the abnormalities such as falls down, cries for help；
Detection module of assembling a crowd all is identified the number of people in image using number of people detection model, and saves each number of people
Coordinate calculates each number of people and is counted in the distance of other numbers of people when distance is less than the distance threshold of setting；Work as the number of people
Quantity carries out early warning of assembling a crowd when being more than the number threshold value of setting；
Multi-source heterogeneous data fusion module, after infrared sensor information, target following information, behavioral value information are handled
Fusion association is carried out, whether comprehensive descision forms invasion；Emphasis prison can be carried out for the result of target following and behavioral value
Control.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910755673.0A CN110459027A (en) | 2019-08-15 | 2019-08-15 | A kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910755673.0A CN110459027A (en) | 2019-08-15 | 2019-08-15 | A kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110459027A true CN110459027A (en) | 2019-11-15 |
Family
ID=68486902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910755673.0A Pending CN110459027A (en) | 2019-08-15 | 2019-08-15 | A kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110459027A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111260525A (en) * | 2020-01-16 | 2020-06-09 | 深圳市广道高新技术股份有限公司 | Community security situation perception and early warning method, system and storage medium |
| CN112489359A (en) * | 2020-12-09 | 2021-03-12 | 江西珉轩大数据有限公司 | Abnormal event early warning system for smart community |
| CN112990254A (en) * | 2020-12-17 | 2021-06-18 | 北京以萨技术股份有限公司 | Fusion analysis method, system, equipment and medium based on multi-source heterogeneous data |
| CN113989324A (en) * | 2021-10-12 | 2022-01-28 | 中国农业大学 | Method, device, electronic device and medium for detecting and tracking abnormal behavior of fish |
| CN114550074A (en) * | 2022-04-25 | 2022-05-27 | 成都信息工程大学 | Image recognition method and system based on computer vision |
| CN115410324A (en) * | 2022-10-28 | 2022-11-29 | 山东世拓房车集团有限公司 | Car as a house night security system and method based on artificial intelligence |
| CN116168344A (en) * | 2023-02-21 | 2023-05-26 | 航天正通汇智(北京)科技股份有限公司 | Security monitoring method and device based on array computing vision |
| CN117176910A (en) * | 2023-09-04 | 2023-12-05 | 深圳市海成智联科技有限公司 | Video management system based on video monitoring platform |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017591A (en) * | 2007-02-06 | 2007-08-15 | 重庆大学 | Video safety prevention and monitoring method based on biology sensing and image information fusion |
| CN101093603A (en) * | 2007-07-03 | 2007-12-26 | 北京智安邦科技有限公司 | Module set of intellective video monitoring device, system and monitoring method |
| CN101635834A (en) * | 2008-07-21 | 2010-01-27 | 侯荣琴 | Automatic tracing identification system for artificial neural control |
| CN101753992A (en) * | 2008-12-17 | 2010-06-23 | 深圳市先进智能技术研究所 | Multi-mode intelligent monitoring system and method |
| WO2015058219A1 (en) * | 2013-09-18 | 2015-04-23 | Coetzer Barend Hendrik | System for identification and tracking of humans |
| CN104881637A (en) * | 2015-05-09 | 2015-09-02 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Multimode information system based on sensing information and target tracking and fusion method thereof |
| CN106127814A (en) * | 2016-07-18 | 2016-11-16 | 四川君逸数码科技股份有限公司 | A kind of wisdom gold eyeball identification gathering of people is fought alarm method and device |
| KR20170081623A (en) * | 2017-07-03 | 2017-07-12 | 엘지이노텍 주식회사 | Method of tracing object using network camera and network camera therefor |
| US20180033024A1 (en) * | 2016-07-28 | 2018-02-01 | Cisco Technology, Inc. | Behavioral Analytic System |
| CN109034020A (en) * | 2018-07-12 | 2018-12-18 | 重庆邮电大学 | A kind of community's Risk Monitoring and prevention method based on Internet of Things and deep learning |
| US20190034735A1 (en) * | 2017-07-25 | 2019-01-31 | Motionloft, Inc. | Object detection sensors and systems |
| CN109325404A (en) * | 2018-08-07 | 2019-02-12 | 长安大学 | A kind of demographic method under public transport scene |
| CN109508657A (en) * | 2018-10-29 | 2019-03-22 | 重庆中科云丛科技有限公司 | Crowd massing analysis method, system, computer readable storage medium and equipment |
| CN110119655A (en) * | 2018-02-07 | 2019-08-13 | 中国石油化工股份有限公司 | A kind of petroleum chemical enterprise's on-site Vehicle And Personnel assembles early warning system and method for early warning |
-
2019
- 2019-08-15 CN CN201910755673.0A patent/CN110459027A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017591A (en) * | 2007-02-06 | 2007-08-15 | 重庆大学 | Video safety prevention and monitoring method based on biology sensing and image information fusion |
| CN101093603A (en) * | 2007-07-03 | 2007-12-26 | 北京智安邦科技有限公司 | Module set of intellective video monitoring device, system and monitoring method |
| CN101635834A (en) * | 2008-07-21 | 2010-01-27 | 侯荣琴 | Automatic tracing identification system for artificial neural control |
| CN101753992A (en) * | 2008-12-17 | 2010-06-23 | 深圳市先进智能技术研究所 | Multi-mode intelligent monitoring system and method |
| WO2015058219A1 (en) * | 2013-09-18 | 2015-04-23 | Coetzer Barend Hendrik | System for identification and tracking of humans |
| CN104881637A (en) * | 2015-05-09 | 2015-09-02 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Multimode information system based on sensing information and target tracking and fusion method thereof |
| CN106127814A (en) * | 2016-07-18 | 2016-11-16 | 四川君逸数码科技股份有限公司 | A kind of wisdom gold eyeball identification gathering of people is fought alarm method and device |
| US20180033024A1 (en) * | 2016-07-28 | 2018-02-01 | Cisco Technology, Inc. | Behavioral Analytic System |
| KR20170081623A (en) * | 2017-07-03 | 2017-07-12 | 엘지이노텍 주식회사 | Method of tracing object using network camera and network camera therefor |
| US20190034735A1 (en) * | 2017-07-25 | 2019-01-31 | Motionloft, Inc. | Object detection sensors and systems |
| CN110119655A (en) * | 2018-02-07 | 2019-08-13 | 中国石油化工股份有限公司 | A kind of petroleum chemical enterprise's on-site Vehicle And Personnel assembles early warning system and method for early warning |
| CN109034020A (en) * | 2018-07-12 | 2018-12-18 | 重庆邮电大学 | A kind of community's Risk Monitoring and prevention method based on Internet of Things and deep learning |
| CN109325404A (en) * | 2018-08-07 | 2019-02-12 | 长安大学 | A kind of demographic method under public transport scene |
| CN109508657A (en) * | 2018-10-29 | 2019-03-22 | 重庆中科云丛科技有限公司 | Crowd massing analysis method, system, computer readable storage medium and equipment |
Non-Patent Citations (4)
| Title |
|---|
| 北北南北: ""《Long-Term Temporal Convolutions for Action Recognition》 论文翻译"", 《CSDN博客》 * |
| 张睿鑫: ""基于人头检测与跟踪的人群行为异常检测"", 《电子测量技术》 * |
| 苏亚涛: "《高校信息资源管理研究》", 30 September 2017, 合肥工业大学出版社 * |
| 陈旭: ""基于深度学习的目标跟踪算法综述"", 《计算机系统应用》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111260525A (en) * | 2020-01-16 | 2020-06-09 | 深圳市广道高新技术股份有限公司 | Community security situation perception and early warning method, system and storage medium |
| CN112489359A (en) * | 2020-12-09 | 2021-03-12 | 江西珉轩大数据有限公司 | Abnormal event early warning system for smart community |
| CN112990254A (en) * | 2020-12-17 | 2021-06-18 | 北京以萨技术股份有限公司 | Fusion analysis method, system, equipment and medium based on multi-source heterogeneous data |
| CN113989324A (en) * | 2021-10-12 | 2022-01-28 | 中国农业大学 | Method, device, electronic device and medium for detecting and tracking abnormal behavior of fish |
| CN114550074A (en) * | 2022-04-25 | 2022-05-27 | 成都信息工程大学 | Image recognition method and system based on computer vision |
| CN115410324A (en) * | 2022-10-28 | 2022-11-29 | 山东世拓房车集团有限公司 | Car as a house night security system and method based on artificial intelligence |
| CN116168344A (en) * | 2023-02-21 | 2023-05-26 | 航天正通汇智(北京)科技股份有限公司 | Security monitoring method and device based on array computing vision |
| CN117176910A (en) * | 2023-09-04 | 2023-12-05 | 深圳市海成智联科技有限公司 | Video management system based on video monitoring platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110459027A (en) | A kind of Community Safety means of defence and system based on multi-source heterogeneous data fusion | |
| CN110188724B (en) | Method and system for helmet positioning and color recognition based on deep learning | |
| US8253564B2 (en) | Predicting a future location of a moving object observed by a surveillance device | |
| CN105141885B (en) | Carry out the method and device of video monitoring | |
| US9524426B2 (en) | Multi-view human detection using semi-exhaustive search | |
| Lee et al. | Hierarchical abnormal event detection by real time and semi-real time multi-tasking video surveillance system | |
| JP2007272488A (en) | Image processor, monitor camera and image monitoring system | |
| US20150294143A1 (en) | Vision based monitoring system for activity sequency validation | |
| CN110852183B (en) | Method, system, device and storage medium for identifying person without wearing safety helmet | |
| US20140119608A1 (en) | System and Methods for Improving Accuracy and Robustness of Abnormal Behavior Detection | |
| CN107645652A (en) | A kind of illegal geofence system based on video monitoring | |
| CN104933436A (en) | Vision-based multi-camera factory monitoring including dynamic integrity grading | |
| KR102174784B1 (en) | Method for Recognizing and Tracking Large-scale Object using Deep learning and Multi-Agent | |
| JP2013088870A (en) | Suspicious person detecting device, and suspicious person detection method and program | |
| EP2858057A1 (en) | System for traffic behaviour surveillance | |
| CN101339688A (en) | Intrusion checking method and system | |
| JP7282186B2 (en) | situational awareness surveillance | |
| CN115527158B (en) | Method and device for detecting abnormal behaviors of personnel based on video monitoring | |
| CN110491060A (en) | A kind of robot and its method for safety monitoring, device and storage medium | |
| CN112711960B (en) | Gate alarm method and device, control processor and readable storage medium | |
| CN106600628A (en) | Target object identification method and device based on infrared thermal imaging system | |
| KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN114022810A (en) | Method, system, medium and terminal for detecting working state of climbing frame protective net in construction site | |
| KR100879622B1 (en) | Method for detecting objects passing through virtual paths on video | |
| CN207530963U (en) | A kind of illegal geofence system based on video monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191115 |