CN113505709A - Method and system for monitoring dangerous behaviors of human body in real time - Google Patents
Method and system for monitoring dangerous behaviors of human body in real time Download PDFInfo
- Publication number
- CN113505709A CN113505709A CN202110800788.4A CN202110800788A CN113505709A CN 113505709 A CN113505709 A CN 113505709A CN 202110800788 A CN202110800788 A CN 202110800788A CN 113505709 A CN113505709 A CN 113505709A
- Authority
- CN
- China
- Prior art keywords
- dangerous
- behavior
- human body
- dangerous behavior
- behaviors
- 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
- 230000006399 behavior Effects 0.000 title claims abstract description 297
- 238000012544 monitoring process Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000003062 neural network model Methods 0.000 claims abstract description 28
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 28
- 238000013528 artificial neural network Methods 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 12
- 231100001261 hazardous Toxicity 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 239000000779 smoke Substances 0.000 claims description 7
- 230000009189 diving Effects 0.000 claims description 4
- 206010010144 Completed suicide Diseases 0.000 description 5
- 230000009191 jumping Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- -1 flames Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 241001622623 Coeliadinae Species 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Alarm Systems (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention relates to a method and a system for monitoring dangerous behaviors of a human body in real time. The method for monitoring the dangerous behaviors of the human body in real time comprises the steps of firstly, acquiring images of an interested area in real time by using a video acquisition module; then aiming at the acquired image, using the trained deep neural network model to identify whether the image has human dangerous behaviors including a human body and dangerous goods adjacent to the human body, if not, continuing to acquire and identify the image, if so, storing the human dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time, and sending a warning trigger signal; and finally, when the warning trigger signal is received, carrying out dangerous behavior warning on the monitor, and displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor. The invention can monitor the dangerous behaviors of the human body simply, accurately and efficiently with low cost, and can avoid the loss caused by the dangerous behaviors of the human body.
Description
Technical Field
The embodiment of the invention relates to the field of security monitoring, in particular to a method and a system for monitoring dangerous behaviors of a human body in real time.
Background
To inhibit the occurrence of suicide events or other dangerous behaviors, it is desirable to discover and intervene and stop the suicide in time before suicide. In the prior art, the order, personnel safety and intervention suicide of various large places are mainly ensured by artificial daily patrol, monitoring in a monitoring room and the like of security guards or policemen and the like, but the artificial mode is time-consuming and labor-consuming and is difficult to ensure omnibearing 24-hour monitoring.
The time, place and mode of suicide are full of unknowns and uncertainties, so that a method and a system for monitoring dangerous behaviors of a human body in real time are urgently needed, so that the dangerous behaviors of the human body can be monitored simply, accurately and efficiently at low cost.
Disclosure of Invention
In order to solve the problems in the prior art, at least one embodiment of the present invention provides a method and a system for monitoring dangerous behaviors of a human body in real time.
In a first aspect, an embodiment of the present invention provides a method for monitoring dangerous behaviors of a human body in real time, including:
the method comprises the following steps of firstly, acquiring an image of an interested area in real time by using a video acquisition module;
step two, aiming at the collected images, using the trained deep neural network model to identify whether the images have human body dangerous behaviors including human bodies and dangerous articles adjacent to the human bodies, if so, analyzing and obtaining dangerous behavior types and continuing the step three, otherwise, returning to the step one;
step three, storing the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time, and sending a warning trigger signal; and
and fourthly, carrying out dangerous behavior warning on a monitor when the warning trigger signal is received, and displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor.
In some embodiments, when the trained deep neural network model is used to identify that the image has a dangerous behavior of a human body including the human body and dangerous goods immediately adjacent to the human body in the second step, a dangerous behavior frame is generated for the dangerous behavior of the human body, and the dangerous behavior frame is included in the dangerous behavior image of the human body in the third step.
In some embodiments, the deep neural network model in step two includes a YoloV4 neural network and a YoloV5 neural network.
In some embodiments, the area of interest includes a roof edge, a high-rise guardrail, a school classroom, a prison, and a water area edge, the supervisor is a security or police, the human hazardous or dangerous behavior types include a building jump behavior, a knife holding behavior, a fire release behavior, a water jump behavior, and a hang-up behavior, and the hazardous items include a roof edge, a guardrail, a knife, smoke, a flare, a water surface, water splash, and a rope.
In some embodiments, the video capture module includes a plurality of cameras aligned with the region of interest and a setting module, the monitor sets monitoring sub-regions for each camera through the setting module, and each camera captures images of its corresponding monitoring sub-region.
In a second aspect, an embodiment of the present invention further provides a system for monitoring dangerous behaviors of a human body in real time, including:
the video acquisition module is used for acquiring images of the region of interest in real time;
the dangerous behavior detection module is used for identifying whether the images have dangerous behaviors of the human body including the human body and dangerous goods close to the human body by using the trained deep neural network model according to the acquired images, analyzing and acquiring dangerous behavior types and sending dangerous behavior trigger signals if the images have the dangerous behaviors and continuing to acquire and identify the images if the images do not have the dangerous behaviors;
the dangerous behavior processing module is used for storing corresponding human dangerous behavior images, dangerous behavior types, dangerous behavior positions and dangerous behavior time when receiving the dangerous behavior trigger signal and sending a warning trigger signal; and
and the dangerous behavior warning module is used for warning dangerous behaviors of a monitor when the warning trigger signal is received, and displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor.
In some embodiments, when the dangerous behavior detection module identifies that the image has a dangerous behavior of a human body including the human body and dangerous goods in close proximity thereto by using the trained deep neural network model, the deep neural network model generates a dangerous behavior frame aiming at the dangerous behavior of the human body, and the dangerous behavior image stored by the dangerous behavior processing module includes the dangerous behavior frame.
In some embodiments, the deep neural network model includes a YoloV4 neural network and a YoloV5 neural network.
In some embodiments, the dangerous behavior warning module includes a warning unit and a display unit, the warning unit is configured to perform dangerous behavior warning on a monitor when receiving the warning trigger signal, and the display unit is configured to display the human dangerous behavior image, the dangerous behavior type, the dangerous behavior position, and the dangerous behavior time to the monitor.
In some embodiments, the area of interest includes a roof edge, a high-rise guardrail, a school classroom, a prison, and a water area edge, the supervisor is a security or police, the human hazardous or dangerous behavior types include a building jump behavior, a knife holding behavior, a fire release behavior, a water jump behavior, and a hang-up behavior, and the hazardous items include a roof edge, a guardrail, a knife, smoke, a flare, a water surface, water splash, and a rope.
In some embodiments, the video capture module includes a plurality of cameras aligned with the region of interest and a setting module, the monitor sets monitoring sub-regions for each camera through the setting module, and each camera captures images of its corresponding monitoring sub-region.
Compared with the prior art that no effective monitoring means exists for the human body dangerous behaviors in dangerous areas such as high buildings, water area edges and the like, the method for monitoring the human body dangerous behaviors in real time of the embodiment of the invention firstly uses the video acquisition module to acquire images of the region of interest in real time; then aiming at the acquired image, using a trained deep neural network model to identify whether the image has a human body dangerous behavior including a human body and dangerous goods adjacent to the human body, if not, continuing to acquire and identify the image, if so, analyzing and acquiring a dangerous behavior type, storing the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time, and sending a warning trigger signal; and finally, when the warning trigger signal is received, carrying out dangerous behavior warning on a monitor, and displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor. The invention can monitor the dangerous behaviors of the human body simply, accurately and efficiently with low cost, and can avoid the loss caused by the dangerous behaviors of the human body.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
Fig. 1 is a schematic structural diagram illustrating a system for monitoring dangerous behaviors of a human body in real time according to an embodiment of the present invention;
FIG. 2 is an exemplary diagram of the human risk behavior image 140A of FIG. 1; and
fig. 3 is a schematic flowchart of a method for monitoring dangerous behaviors of a human body in real time according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
Referring to fig. 1, a component structure of a system 1 for real-time monitoring dangerous behaviors of a human body in an embodiment of the present invention is shown. The dangerous behaviors of the human body comprise a building jumping behavior, a knife holding behavior, a fire releasing behavior, a water jumping behavior, a hanging behavior and the like. The dangerous behaviors of the human body can be set and changed by a monitor according to specific conditions. As shown in fig. 1, a system 1 for monitoring dangerous behaviors of a human body in real time includes a video capture module 10, a dangerous behavior detection module 12, a dangerous behavior processing module 14, and a dangerous behavior warning module 16. The components of the system 1 for real-time monitoring of dangerous behaviour of a human being are explained in detail below.
The video capture module 10 is used to capture images of the region of interest in real time. The video acquisition module 10 includes a plurality of cameras 100 aligned with an area of interest and a setting module 102, the monitor sets monitoring sub-regions for each camera 100 through the setting module 102, and each camera 100 performs image acquisition for its corresponding monitoring sub-region. The region of interest includes a roof edge (as shown in fig. 2), a high-rise guardrail, a school classroom, a prison, a water area edge, and the like, and can be set and changed by a monitor according to actual conditions, for example, a street can also be listed as the region of interest.
The dangerous behavior detection module 12 is configured to, for the image acquired by the video acquisition module 10, use the deep neural network model 120 whose training is completed to identify whether there is a dangerous behavior of the human body including the human body and dangerous goods immediately adjacent to the human body in the image, analyze and acquire a dangerous behavior type and send a dangerous behavior trigger signal if there is a dangerous behavior type, and continue to perform image acquisition and identification if not. The deep neural network model 120 is trained with a large number of images of dangerous behavior of the human body. The dangerous behavior types comprise a building jumping behavior, a knife holding behavior, a fire releasing behavior, a water jumping behavior, a hanging behavior and the like. The dangerous goods comprise roof edges, guardrails, cutters, smoke, flames, water surfaces, water flowers, ropes and the like, wherein the roof edges and the guardrails correspond to the building jump behavior, the cutters correspond to the cutter holding behavior, the smoke and the flames correspond to the fire releasing behavior, the water surfaces and the water flowers correspond to the water jump behavior, and the ropes correspond to the upper hanging behavior.
When the dangerous behavior detection module 12 identifies that there is a dangerous behavior of the human body in the image by using the deep neural network model 120 that has been trained, the deep neural network model 120 generates a dangerous behavior frame 120A as shown in fig. 2 for the dangerous behavior of the human body, the dangerous behavior frame 120A frames the human body and the dangerous goods in close proximity thereto, and the deep neural network model 120 may be a YoloV4 neural network or a YoloV5 neural network.
The YOLOv5 neural network is the most advanced target detection network in 2020, has higher speed and higher precision, and can reduce the calculation amount and improve the detection speed by using the CSPDarknet53 network. The method is characterized in that the Neck part of the YOLOv5 neural network is added with FPN and PAN enhancement feature fusion to enhance the small target detection effect, a new box loss function is used to accelerate model convergence and improve box accuracy, meanwhile, multiple data enhancement technologies are added in the aspect of data processing, the network detection accuracy is improved by the technologies such as Mosaic and Mixup, and models with multiple sizes are set to flexibly select different models according to different computational powers. The image of the region of interest is input into a Yolov5 neural network, coordinates of a dangerous behavior border 120A are output if a human dangerous behavior appears in the image, and a null is fed back if no human dangerous behavior appears, and detection of a situation where multiple human dangerous behaviors appear simultaneously is supported.
The dangerous behavior processing module 14 comprises a database 140, when receiving the dangerous behavior trigger signal, the dangerous behavior processing module 14 stores the human body dangerous behavior image 140A, the dangerous behavior type 140B, the dangerous behavior position 140C and the dangerous behavior time 140D in the database 140, and sends a warning trigger signal to the dangerous behavior warning module 16, and when storing, the dangerous behavior types such as a building jump behavior, a knife holding behavior, a fire release behavior, a water jump behavior and an upward hanging behavior can be respectively represented by binary codes 001, 010, 011, 100 and 101. The dangerous behavior position 140C may be a monitoring sub-region corresponding to the camera 100, and may specifically be determined by an identification code of the camera, for example, the image in fig. 2 is an image acquired by the No. 2 camera for the monitoring sub-region corresponding to the camera, and the dangerous behavior position may be determined by determining the monitoring sub-region corresponding to the No. 2 camera.
Fig. 2 is a schematic diagram of the dangerous behavior processing module 14 in fig. 1 processing the stored dangerous behavior image of the human body, and the deep neural network model 120 identifies the jumping floor behavior of fig. 2 including the human body and the roof edge in close proximity thereto. As shown in fig. 2, the dangerous behavior frame 120A is included in the human dangerous behavior image 140A, the dangerous behavior frame 120A is obtained by performing dangerous behavior recognition on the image of the region of interest (e.g. the top floor edge) acquired by the video acquisition module 10 by using the deep neural network model 120, which is exemplarily a YoloV5 neural network, and the dangerous behavior frame 120A accurately and completely frames the human body and the top floor edge therein. The deep neural network model 120 identifies the jump behavior of figure 2 including the top edge and the human body in close proximity thereto.
The dangerous behavior warning module 16 is configured to perform dangerous behavior warning on a monitor when receiving the warning trigger signal, and display a human dangerous behavior image 140A, a dangerous behavior type 140B, a dangerous behavior position 140C, and a dangerous behavior time 140D to the monitor. The monitor is a security guard or a police.
Dangerous behavior warning module 16 includes warning unit 160 and display element 162, warning unit 160 is used for receiving carry out dangerous behavior warning to the supervisor during warning trigger signal, display element 162 be used for to the supervisor shows human dangerous behavior image 140A, dangerous behavior position 140C and dangerous behavior time 140D, and warning unit 160 accessible speech mode or the mode of pronunciation mixed light warns.
Fig. 3 is a schematic flowchart of a method for monitoring dangerous behaviors of a human body in real time according to an embodiment of the present invention. Referring to fig. 3, with combined reference to fig. 1-2, the method 30 for real-time monitoring dangerous behaviors of a human body first performs step S300, and acquires an image of a region of interest in real time using the video acquisition module 10. In step S300, the region of interest includes a roof edge, a high-rise guardrail, a school classroom, a prison, a water area edge, and the like. As shown in fig. 1, the video capture module 10 includes a plurality of cameras 100 aligned with a region of interest, the monitor can set monitoring sub-regions for each camera 100 through a setting module 102 of the video capture module 10, and each camera 100 captures an image of its corresponding monitoring sub-region.
The method 30 for real-time monitoring of dangerous behaviors of human body continues to step S310, and for the acquired image, the trained deep neural network model 120 is used to identify whether there is a dangerous behavior of human body including human body and dangerous goods in close proximity thereto, if so, the step S320 is continued, otherwise, the step S300 is returned to. When the deep neural network model 120 after training in step S310 identifies that there is a dangerous behavior in the image, a dangerous behavior border 120A shown in fig. 2 is generated for the dangerous behavior of the human body, and the deep neural network model 120 includes a YoloV4 neural network and a YoloV5 neural network. The dangerous behaviors of the human body in the step S310 include a building jump behavior, a knife holding behavior, a fire releasing behavior, a diving behavior, a hanging behavior and the like, and the dangerous goods include roof edges, guardrails, knives, smoke, flames, water surfaces, water flowers, ropes and the like.
In step S320, the dangerous behavior type 140B is obtained by analyzing using the deep neural network model 120. The dangerous behavior types in step S320 include a jumping-building behavior, a knife holding behavior, a fire releasing behavior, a diving behavior, a hanging-up behavior, and the like.
The method 30 for monitoring the dangerous behavior of the human body in real time continues to step S330, where the image 140A of the dangerous behavior of the human body, the type 140B of the dangerous behavior, the position 140C of the dangerous behavior, and the time 140D of the dangerous behavior are stored in the database 140, and an alert trigger signal is sent. The human dangerous behavior image 140A in step S330 includes a dangerous behavior frame 120A as shown in fig. 2. In step S330, the binary codes 001, 010, 011, 100, and 101 can be used to respectively represent the building jump behavior, the knife holding behavior, the fire release behavior, the water jump behavior, and the hanging behavior in the database 140.
The method 30 for monitoring dangerous behaviors of a human body in real time continues to step S340, and warns a dangerous behavior of a monitor when receiving the warning trigger signal, and displays a dangerous behavior image 140A of the human body, a dangerous behavior type 140B, a dangerous behavior position 140C, and a dangerous behavior time 140D to the monitor. The monitor in step S340 is a security guard or a police, and the warning of dangerous behavior may be performed in a voice manner or a voice and light mixed manner.
The method for monitoring the dangerous behaviors of the human body in real time comprises the steps of firstly, using a video acquisition module to acquire images of an interested area in real time; then aiming at the acquired image, using a trained deep neural network model to identify whether the image has a human body dangerous behavior comprising a human body and dangerous goods adjacent to the human body, if not, continuing to acquire and identify the image, if so, analyzing and acquiring a dangerous behavior type, storing the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time, and sending a warning trigger signal; and finally, when the warning trigger signal is received, carrying out dangerous behavior warning on a monitor, and displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor.
The invention can monitor the dangerous behaviors of the human body simply, accurately and efficiently with low cost, and can avoid the loss caused by the dangerous behaviors of the human body.
For a software implementation, the techniques described herein may be implemented by means of units performing the functions described herein. The software codes may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present application, it should be understood that the execution sequence of the steps of the method embodiments can be arbitrarily adjusted unless there is an explicit precedence sequence. The disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or make a contribution to the prior art, or may be implemented in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Those skilled in the art will appreciate that although some embodiments described herein include some features included in other embodiments instead of others, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (10)
1. A method for real-time monitoring of dangerous behavior of a human, comprising:
the method comprises the following steps of firstly, acquiring an image of an interested area in real time by using a video acquisition module;
step two, aiming at the collected images, using the trained deep neural network model to identify whether the images have human body dangerous behaviors including human bodies and dangerous articles adjacent to the human bodies, if so, analyzing and obtaining dangerous behavior types and continuing the step three, otherwise, returning to the step one;
step three, storing the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time, and sending a warning trigger signal; and
and fourthly, carrying out dangerous behavior warning on a monitor when the warning trigger signal is received, and displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor.
2. The method according to claim 1, wherein when the trained deep neural network model is used to identify dangerous behaviors of the human body including the human body and dangerous goods in close proximity thereto in the image, a dangerous behavior frame is generated for the dangerous behaviors of the human body, and the dangerous behavior frame is included in the dangerous behavior image of the human body in step three.
3. The method for real-time monitoring of dangerous behavior of human body according to claim 1 or 2, wherein the deep neural network model in step two comprises a YoloV4 neural network and a YoloV5 neural network.
4. The method for real-time monitoring of human hazardous behaviors of claim 1, wherein the area of interest comprises a roof edge, a high-rise guardrail, a school classroom, a prison and a water area edge, the guardian is a security or police, the human hazardous behavior or type of hazardous behavior comprises a jumping-up behavior, a knife holding behavior, a fire release behavior, a diving behavior and a hanging-up behavior, and the hazardous articles comprise a roof edge, a guardrail, a knife, smoke, a flame, a water surface, a water splash and a rope.
5. The method for real-time monitoring of dangerous behaviors of human bodies according to claim 4, wherein the video capture module comprises a plurality of cameras aligned with the regions of interest and a setting module, the monitor sets a monitoring sub-region for each camera through the setting module, and each camera captures images of its corresponding monitoring sub-region.
6. A system for real-time monitoring of dangerous behavior of a human, comprising:
the video acquisition module is used for acquiring images of the region of interest in real time;
the dangerous behavior detection module is used for identifying whether the images have dangerous behaviors of the human body including the human body and dangerous goods close to the human body by using the trained deep neural network model according to the acquired images, analyzing and acquiring dangerous behavior types and sending dangerous behavior trigger signals if the images have the dangerous behaviors and continuing to acquire and identify the images if the images do not have the dangerous behaviors;
the dangerous behavior processing module is used for storing corresponding human dangerous behavior images, dangerous behavior types, dangerous behavior positions and dangerous behavior time when receiving the dangerous behavior trigger signal and sending a warning trigger signal; and
and the dangerous behavior warning module is used for warning dangerous behaviors of a monitor when the warning trigger signal is received, and displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor.
7. The system according to claim 6, wherein the dangerous behavior detection module generates a dangerous behavior frame for the dangerous behavior of the human body when the trained deep neural network model is used to identify that the image has the dangerous behavior of the human body including the human body and dangerous goods adjacent to the human body, and the dangerous behavior processing module stores the image of the dangerous behavior of the human body including the dangerous behavior frame.
8. The system for real-time monitoring of dangerous behavior of human body according to claim 6, wherein said deep neural network model comprises a yoloV4 neural network and a yoloV5 neural network; the dangerous behavior warning module comprises a warning unit and a display unit, the warning unit is used for warning dangerous behaviors of a monitor when the warning trigger signal is received, and the display unit is used for displaying the human body dangerous behavior image, the dangerous behavior type, the dangerous behavior position and the dangerous behavior time to the monitor.
9. The system for real-time monitoring of human hazardous behaviors of claim 6, wherein the area of interest comprises a roof edge, a high-rise guardrail, a school classroom, a prison and a water edge, the guardian is a security or police, the human hazardous behavior or type of hazardous behavior comprises a jumping-building behavior, a knife holding behavior, a fire release behavior, a diving behavior and a hanging-up behavior, and the hazardous articles comprise a roof edge, a guardrail, a knife, smoke, a flame, a water surface, a water splash and a rope.
10. The system for real-time monitoring of dangerous behaviors of human bodies according to claim 6, wherein said video capture module comprises a plurality of cameras aligned with the regions of interest and a setting module, said monitor sets a monitoring sub-region for each camera through said setting module, and each camera captures images of its corresponding monitoring sub-region.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110800788.4A CN113505709A (en) | 2021-07-15 | 2021-07-15 | Method and system for monitoring dangerous behaviors of human body in real time |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110800788.4A CN113505709A (en) | 2021-07-15 | 2021-07-15 | Method and system for monitoring dangerous behaviors of human body in real time |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113505709A true CN113505709A (en) | 2021-10-15 |
Family
ID=78013015
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110800788.4A Pending CN113505709A (en) | 2021-07-15 | 2021-07-15 | Method and system for monitoring dangerous behaviors of human body in real time |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113505709A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114092865A (en) * | 2022-01-21 | 2022-02-25 | 北京清鹤科技有限公司 | Prisoner supervision method and system based on neural network |
CN114724246A (en) * | 2022-04-11 | 2022-07-08 | 中国人民解放军东部战区总医院 | Dangerous behavior identification method and device |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107045623A (en) * | 2016-12-30 | 2017-08-15 | 厦门瑞为信息技术有限公司 | A kind of method of the indoor dangerous situation alarm based on human body attitude trace analysis |
CN109871775A (en) * | 2019-01-22 | 2019-06-11 | 北京影谱科技股份有限公司 | A kind of the ice rink monitoring method and device of Behavior-based control detection |
CN110084228A (en) * | 2019-06-25 | 2019-08-02 | 江苏德劭信息科技有限公司 | A kind of hazardous act automatic identifying method based on double-current convolutional neural networks |
CN110969130A (en) * | 2019-12-03 | 2020-04-07 | 厦门瑞为信息技术有限公司 | Driver dangerous action identification method and system based on YOLOV3 |
CN110969078A (en) * | 2019-09-17 | 2020-04-07 | 博康智能信息技术有限公司 | Abnormal behavior identification method based on human body key points |
CN111083435A (en) * | 2019-10-27 | 2020-04-28 | 恒大智慧科技有限公司 | Security monitoring method and device and computer readable storage medium |
CN210895500U (en) * | 2019-12-31 | 2020-06-30 | 吉首大学 | Dangerous behavior action recognition system |
CN111353365A (en) * | 2019-08-19 | 2020-06-30 | 深圳市鸿合创新信息技术有限责任公司 | Behavior early warning method and device and electronic equipment |
CN111881865A (en) * | 2020-08-03 | 2020-11-03 | 南京奥拓电子科技有限公司 | Self-adaptive dangerous behavior monitoring method and system and intelligent equipment |
CN111967400A (en) * | 2020-08-19 | 2020-11-20 | 顾晓东 | Campus safety monitoring and early warning management system based on artificial intelligence |
-
2021
- 2021-07-15 CN CN202110800788.4A patent/CN113505709A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107045623A (en) * | 2016-12-30 | 2017-08-15 | 厦门瑞为信息技术有限公司 | A kind of method of the indoor dangerous situation alarm based on human body attitude trace analysis |
CN109871775A (en) * | 2019-01-22 | 2019-06-11 | 北京影谱科技股份有限公司 | A kind of the ice rink monitoring method and device of Behavior-based control detection |
CN110084228A (en) * | 2019-06-25 | 2019-08-02 | 江苏德劭信息科技有限公司 | A kind of hazardous act automatic identifying method based on double-current convolutional neural networks |
CN111353365A (en) * | 2019-08-19 | 2020-06-30 | 深圳市鸿合创新信息技术有限责任公司 | Behavior early warning method and device and electronic equipment |
CN110969078A (en) * | 2019-09-17 | 2020-04-07 | 博康智能信息技术有限公司 | Abnormal behavior identification method based on human body key points |
CN111083435A (en) * | 2019-10-27 | 2020-04-28 | 恒大智慧科技有限公司 | Security monitoring method and device and computer readable storage medium |
CN110969130A (en) * | 2019-12-03 | 2020-04-07 | 厦门瑞为信息技术有限公司 | Driver dangerous action identification method and system based on YOLOV3 |
CN210895500U (en) * | 2019-12-31 | 2020-06-30 | 吉首大学 | Dangerous behavior action recognition system |
CN111881865A (en) * | 2020-08-03 | 2020-11-03 | 南京奥拓电子科技有限公司 | Self-adaptive dangerous behavior monitoring method and system and intelligent equipment |
CN111967400A (en) * | 2020-08-19 | 2020-11-20 | 顾晓东 | Campus safety monitoring and early warning management system based on artificial intelligence |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114092865A (en) * | 2022-01-21 | 2022-02-25 | 北京清鹤科技有限公司 | Prisoner supervision method and system based on neural network |
CN114724246A (en) * | 2022-04-11 | 2022-07-08 | 中国人民解放军东部战区总医院 | Dangerous behavior identification method and device |
CN114724246B (en) * | 2022-04-11 | 2024-01-30 | 中国人民解放军东部战区总医院 | Dangerous behavior identification method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10275657B2 (en) | Video surveillance system, video processing apparatus, video processing method, and video processing program | |
CN211293956U (en) | AI-based identification and alarm system for abnormal agent on construction site | |
WO2019058379A1 (en) | Systems and methods for preventing work accidents | |
CN107437318B (en) | Visible light intelligent recognition algorithm | |
CN109598885A (en) | Monitoring system and its alarm method | |
CN113505709A (en) | Method and system for monitoring dangerous behaviors of human body in real time | |
CN104091354A (en) | Fire detection method based on video images and fire detection device thereof | |
KR102149832B1 (en) | Automated Violence Detecting System based on Deep Learning | |
CN112488483A (en) | AI technology-based EHS transparent management system and management method | |
JP2021132267A (en) | Video monitoring system and video monitoring method | |
CN109886129B (en) | Prompt message generation method and device, storage medium and electronic device | |
CN108319892A (en) | A kind of vehicle safety method for early warning and system based on genetic algorithm | |
US20210364356A1 (en) | System and method for using artificial intelligence to enable elevated temperature detection of persons using commodity-based thermal cameras | |
KR20230004421A (en) | System for detecting abnormal behavior based on artificial intelligence | |
CN112071007A (en) | Museum safety prevention and control management system | |
CN111985428A (en) | Security detection method and device, electronic equipment and storage medium | |
US10650651B1 (en) | Automated geospatial security event grouping | |
CN111698473A (en) | Method and device for detecting staff on duty | |
CN111860187A (en) | High-precision worn mask identification method and system | |
CN114155470A (en) | River channel area intrusion detection method, system and storage medium | |
CN112576310B (en) | Tunnel security detection method and system based on robot | |
CN117636169A (en) | Forest patrol method and system | |
Hameete et al. | Intelligent Multi-Camera Video Surveillance. | |
CN116246401A (en) | Monitoring system for community public area management | |
Van Schaik et al. | Assessment of visual cues by tower controllers, with implications for a remote tower control centre |
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 |