Abnormal intrusion monitoring method and system for smart community

Abstract

The invention relates to the field of smart community monitoring, and discloses an abnormal intrusion monitoring method and system for a smart community, and the method comprises the steps: determining key frames of a plurality of monitoring videos corresponding to a target person, the time sequence of the key frames of the plurality of monitoring videos, and the time span between the key frames of the plurality of monitoring videos; determining event types and event monitoring indexes respectively corresponding to the key frames of the plurality of monitoring videos according to the key frames of the plurality of monitoring videos through an event identification model; constructing a monitoring event graph corresponding to the target person according to the event types and the event monitoring indexes corresponding to the key frames of the plurality of monitoring videos; and according to the monitoring event graph, determining an abnormal intrusion probability value of the target person, and when the abnormal intrusion probability value of the target person is greater than or equal to a preset abnormal intrusion probability value, sending out abnormal intrusion prompt information corresponding to the target person. The abnormal intrusion monitoring accuracy of the smart community can be improved, and the checking burden of security personnel is reduced.

Description

Technical Field

[0001] This application relates to the technical field of intelligent community monitoring. More specifically, it relates to an abnormal intrusion monitoring method and system for an intelligent community. Background Art

[0002] Currently, the security system of intelligent communities mainly relies on the combination of traditional video monitoring and manual patrol. In this mode, monitoring devices usually use fixed cameras to monitor each area of the community in real time. At the same time, the response process for intrusion events depends on manual review. Once the monitoring system detects an abnormality, security personnel will immediately conduct a review and take corresponding measures.

[0003] However, this community security system still has significant technical bottlenecks in the field of abnormal intrusion monitoring. First, limited by the physical installation location, traditional fixed cameras are prone to form monitoring blind spots in complex areas such as community corners and green belts, and cannot achieve full-scenario coverage. Second, in the intrusion event response process, the time difference from abnormality trigger to alarm confirmation often exceeds 30 seconds, making it difficult to meet the real-time disposal requirements for emergency events. In addition, the monitoring of abnormal personnel activities in the community mainly relies on an intelligent security system based on face recognition, but this system has the following core defects: the target recognition range is limited and it cannot identify unregistered suspicious personnel; the adaptability to dynamic scenarios is insufficient, being sensitive to lighting conditions, occlusions, and posture changes, resulting in a false alarm rate of over 40% at night or in complex environments. This not only increases the verification burden on security personnel but may also lead to potential safety hazards due to missed alarms. Summary of the Invention

[0004] The purpose of this application is to provide an abnormal intrusion monitoring method and system for an intelligent community, which solves the technical problem of poor monitoring effect of abnormal intrusion in existing intelligent communities and achieves the technical effect of improving the accuracy of abnormal intrusion monitoring in intelligent communities.

[0005] An abnormal intrusion monitoring method for an intelligent community provided by an embodiment of the present application, the method includes: respectively obtaining a plurality of monitoring videos corresponding to a target person captured by a plurality of cameras, and determining key frames of the plurality of monitoring videos corresponding to the target person, the time sequence of the key frames of the plurality of monitoring videos, and the time span between the key frames of the plurality of monitoring videos; through an event recognition model, determining, according to the key frames of the plurality of monitoring videos, the event types and event monitoring indicators respectively corresponding to the key frames of the plurality of monitoring videos; constructing a monitoring event graph corresponding to the target person according to the event types and event monitoring indicators respectively corresponding to the key frames of the plurality of monitoring videos; wherein, the monitoring event graph includes event nodes and edges connecting the event nodes, the attributes of the event nodes include event types and event monitoring indicators, the edges connecting the event nodes connect the event nodes in the time sequence of the key frames of the plurality of monitoring videos, and the attributes of the edges connecting the event nodes include the time span between the event nodes corresponding to the key frames of the monitoring videos; through an abnormal intrusion detection model, determining, according to the monitoring event graph, the abnormal intrusion probability value of the target person, and when the abnormal intrusion probability value of the target person is greater than or equal to a preset abnormal intrusion probability value, sending out an abnormal intrusion prompt message corresponding to the target person; wherein, the abnormal intrusion detection model is a graph neural network model.

[0006] In a possible implementation manner, the method includes: obtaining the regional scenes respectively corresponding to the key frames of the plurality of monitoring videos, and obtaining the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of the plurality of monitoring videos adjacent in time sequence, and adding the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of the plurality of monitoring videos to the attributes of the edges connecting the event nodes; through the abnormal intrusion detection model, determining, according to the monitoring event graph, the abnormal intrusion probability value of the target person.

[0007] In another possible implementation manner, the method includes: determining the monitoring event graph at the moments respectively corresponding to the key frames of the plurality of monitoring videos, determining, through the abnormal intrusion detection model, the abnormal intrusion probability value of the target person corresponding to each monitoring event graph, and determining the cumulative value of the abnormal intrusion probability values; when the cumulative value of the abnormal intrusion probability values of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability values, sending out an abnormal intrusion prompt message corresponding to the target person.

[0008] In another possible implementation, the method includes: obtaining the regional scenarios to which the key frames of multiple surveillance videos respectively belong, obtaining the intrusion probability weight value corresponding to each regional scenario, and determining the sum of the products of the abnormal intrusion probability values and the intrusion probability weight values corresponding to the regional scenarios to which the key frames of the multiple surveillance videos respectively belong, as the cumulative value of the abnormal intrusion probability value; when the cumulative value of the abnormal intrusion probability value of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability value, sending out the abnormal intrusion prompt information corresponding to the target person.

[0009] In another possible implementation, the method includes: obtaining the regional scenarios to which the key frames of multiple surveillance videos respectively belong, and obtaining the key frame quantity threshold and the time span threshold corresponding to each regional scenario; determining the number of key frames of the surveillance videos within each regional scenario; when the number of key frames of the surveillance video corresponding to the regional scenario is less than the key frame quantity threshold of the surveillance video, or when the time span of the key frames of the surveillance video corresponding to the regional scenario is less than the preset time span threshold, deleting the event nodes corresponding to the key frames of the surveillance video in the surveillance event graph corresponding to the regional scenario.

[0010] In another possible implementation, the method includes: obtaining the time periods to which the key frames of multiple surveillance videos belong, and obtaining the time period adjustment factors corresponding to the time periods to which the key frames of the multiple surveillance videos belong; determining the product of the key frame quantity threshold corresponding to each regional scenario and the time period adjustment factor to adjust the key frame quantity threshold; determining the product of the time span threshold corresponding to each regional scenario and the time period adjustment factor to adjust the time period adjustment factor.

[0011] In another possible implementation, the method includes: determining multiple associated persons of the target person, and constructing surveillance event graphs respectively corresponding to the multiple associated persons; through the abnormal intrusion detection model, determining the group abnormal intrusion probability values corresponding to the target person and the multiple associated persons according to the surveillance event graphs respectively corresponding to the target person and the multiple associated persons. When the group abnormal intrusion probability values corresponding to the target person and the multiple associated persons are greater than or equal to the preset group abnormal intrusion probability value, sending out the group abnormal intrusion prompt information corresponding to the target person and the multiple associated persons.

[0012] In another possible implementation, determining multiple associated persons of the target person includes: constructing surveillance event graphs respectively corresponding to multiple persons, and determining the event nodes respectively corresponding to the same regional scenarios in the surveillance event graphs of each person and the target person; determining the similarity of the event nodes respectively corresponding to the same regional scenarios of each person and the target person, and determining the sum of the similarities of all the event nodes of each person and the target person as the person association degree; when the person association degree of the first person is greater than or equal to the preset person association degree, taking the first person as the associated person of the target person.

[0013] In another possible implementation, the method includes: obtaining a similarity adjustment factor corresponding to each regional scenario, and determining the product of the similarity between the event nodes corresponding to each person and the target person in the same regional scenario and the similarity adjustment factor as the person correlation degree.

[0014] The embodiment of the present application also provides an abnormal intrusion monitoring system for a smart community, including a unit for executing the method described in any one of the above.

[0015] The beneficial effects of the embodiment of the present application compared with the prior art are:

[0016] The embodiment of the present application provides an abnormal intrusion monitoring method for a smart community. The method includes: respectively obtaining a plurality of monitoring videos corresponding to a target person captured by a plurality of cameras, and determining the key frames of the plurality of monitoring videos corresponding to the target person, the time sequence of the key frames of the plurality of monitoring videos, and the time span between the key frames of the plurality of monitoring videos; through an event recognition model, determining the event types and event monitoring indicators respectively corresponding to the key frames of the plurality of monitoring videos according to the key frames of the plurality of monitoring videos; constructing a monitoring event graph corresponding to the target person according to the event types and event monitoring indicators respectively corresponding to the key frames of the plurality of monitoring videos; wherein, the monitoring event graph includes event nodes and edges connecting the event nodes, the attributes of the event nodes include event types and event monitoring indicators, the edges connecting the event nodes connect the event nodes according to the time sequence of the key frames of the plurality of monitoring videos, and the attributes of the edges connecting the event nodes include the time span between the event nodes corresponding to the key frames of the monitoring videos; through an abnormal intrusion detection model, determining an abnormal intrusion probability value of the target person according to the monitoring event graph, and when the abnormal intrusion probability value of the target person is greater than or equal to a preset abnormal intrusion probability value, sending out an abnormal intrusion prompt message corresponding to the target person; wherein, the abnormal intrusion detection model is a graph neural network model. The embodiment of the present application can construct a monitoring event graph of the target person, and combine and identify the monitoring event graph by using a graph neural network, realizing the accurate identification of the activity pattern of the target person in the community, improving the accuracy of identifying the abnormal intrusion of the target person in the community, and improving the abnormal intrusion recognition effect of the smart community. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0018] Figure 1Schematic flowchart of the first method for abnormal intrusion monitoring in an intelligent community provided by an embodiment of the present application;

[0019] Figure 2 Schematic flowchart of the second method for abnormal intrusion monitoring in an intelligent community provided by an embodiment of the present application;

[0020] Figure 3 Schematic flowchart of the third method for abnormal intrusion monitoring in an intelligent community provided by an embodiment of the present application;

[0021] Figure 4 Schematic flowchart of the fourth method for abnormal intrusion monitoring in an intelligent community provided by an embodiment of the present application;

[0022] Figure 5 Schematic flowchart of the fifth method for abnormal intrusion monitoring in an intelligent community provided by an embodiment of the present application;

[0023] Figure 6 Schematic flowchart of the sixth method for abnormal intrusion monitoring in an intelligent community provided by an embodiment of the present application;

[0024] Figure 7 Schematic logical structure diagram of an abnormal intrusion monitoring system for an intelligent community provided by an embodiment of the present application. Detailed implementation manners

[0025] It should be understood that when used in the specification and appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0026] It should also be understood that the term "and / or" as used in the specification and appended claims of the present application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0027] As used in the specification and appended claims of the present application, the term "if" may be construed as "when" or "once" or "in response to determining" or "in response to detecting" depending on the context. Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be construed as meaning "once determined" or "in response to determining" or "once detected [the described condition or event]" or "in response to detecting [the described condition or event]" depending on the context.

[0028] In addition, in the description of the specification and the appended claims of the present application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0029] The reference to "one embodiment" or "some embodiments" etc. described in the specification of the present application means that a specific feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, the statements "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0030] The current community security system has the following core defects: the target recognition range is limited and it cannot recognize unregistered suspicious persons; the adaptability to dynamic scenarios is insufficient, being sensitive to lighting conditions, occlusions, and pose changes, resulting in a false alarm rate of over 40% at night or in complex environments. This not only increases the verification burden on security personnel but may also pose security risks due to missed alarms.

[0031] For the above reasons, the embodiments of the present application provide an abnormal intrusion monitoring method for a smart community. The method includes: respectively obtaining multiple monitoring videos corresponding to a target person captured by multiple cameras, and determining the key frames of the multiple monitoring videos corresponding to the target person, the time sequence of the key frames of the multiple monitoring videos, and the time span between the key frames of the multiple monitoring videos; through an event recognition model, determining the event types and event monitoring indicators respectively corresponding to the key frames of the multiple monitoring videos according to the key frames of the multiple monitoring videos; constructing a monitoring event graph corresponding to the target person according to the event types and event monitoring indicators respectively corresponding to the key frames of the multiple monitoring videos; wherein, the monitoring event graph includes event nodes and edges connecting the event nodes, the attributes of the event nodes include event types and event monitoring indicators, the edges connecting the event nodes connect the event nodes in the time sequence of the key frames of the multiple monitoring videos, and the attributes of the edges connecting the event nodes include the time span between the event nodes corresponding to the key frames of the monitoring videos; through an abnormal intrusion detection model, determining the abnormal intrusion probability value of the target person according to the monitoring event graph, and when the abnormal intrusion probability value of the target person is greater than or equal to a preset abnormal intrusion probability value, sending out an abnormal intrusion prompt message corresponding to the target person; wherein, the abnormal intrusion detection model is a graph neural network model. The embodiments of the present application can construct a monitoring event graph of the target person, and combine the graph neural network to identify the monitoring event graph, realizing the accurate identification of the activity pattern of the target person in the community, improving the accuracy of identifying the abnormal intrusion of the target person in the community, and improving the abnormal intrusion recognition effect of the smart community.

[0032] In some scenarios, an abnormal intrusion monitoring method for a smart community in the embodiments of the present application can be applied to the abnormal intrusion recognition of a smart community and can be used for intrusion detection in areas such as communities.

[0033] The following specifically describes an abnormal intrusion monitoring method for a smart community provided by the embodiments of the present application with specific examples.

[0034] Figure 1 FIG. is a schematic flowchart of the first abnormal intrusion monitoring method for a smart community provided by the embodiments of the present application. As Figure 1 shown, this method includes S110 to S130, and the following specifically describes S110 to S130.

[0035] S110. Respectively obtain multiple monitoring videos corresponding to a target person captured by multiple cameras, and determine the key frames of the multiple monitoring videos corresponding to the target person, the time sequence of the key frames of the multiple monitoring videos, and the time span between the key frames of the multiple monitoring videos. Through an event recognition model, determine the event types and event monitoring indicators respectively corresponding to the key frames of the multiple monitoring videos according to the key frames of the multiple monitoring videos.

[0036] When performing intrusion detection in a smart community, multiple surveillance videos corresponding to a target person captured by multiple cameras can be obtained respectively. The target person is any person in the community. This application can identify the surveillance videos of the target person and perform event recognition on the surveillance videos to identify the activities of the target person.

[0037] After obtaining multiple surveillance videos corresponding to the target person, the key frames of the multiple surveillance videos corresponding to the target person, the time sequence of the key frames of the multiple surveillance videos, and the time span between the key frames of the multiple surveillance videos can be determined through a person event recognition model. Among them, the key frames of the multiple surveillance videos are the key frames that can represent the activity events of the target person. After obtaining the key frames of the multiple surveillance videos on the time axis, the time sequence of the key frames of the multiple surveillance videos on the time axis can be determined, and the time span between the key frames of the multiple surveillance videos can be determined. The time span is the time interval between the key frames of the multiple surveillance videos.

[0038] Exemplarily, the key frames of the multiple surveillance videos include the key frames corresponding to the staying time of the target person in a certain area, specific actions, and specific behavioral expressions.

[0039] When obtaining the key frames of the multiple surveillance videos corresponding to the target person, the event types and event monitoring indicators respectively corresponding to the key frames of the multiple surveillance videos can be determined through an event recognition model. Furthermore, the activities of the target person can be monitored according to the event types and event monitoring indicators.

[0040] Exemplarily, the event type can be the event type recorded by text, and the event monitoring indicator can be the marked activity suspiciousness value.

[0041] Exemplarily, the event recognition model can be trained by the marked key frames of the surveillance videos, event types, and event monitoring indicators.

[0042] S120. Construct a surveillance event graph corresponding to the target person according to the event types and event monitoring indicators respectively corresponding to the key frames of the multiple surveillance videos. Among them, the surveillance event graph includes event nodes and edges connecting the event nodes. The attributes of the event nodes include event types and event monitoring indicators. The edges connecting the event nodes connect the event nodes according to the time sequence of the key frames of the multiple surveillance videos. The attributes of the edges connecting the event nodes include the time span between the event nodes corresponding to the key frames of the surveillance videos.

[0043] After obtaining the event types and event monitoring metrics corresponding to the key frames of multiple surveillance videos, a surveillance event graph corresponding to the target person is constructed according to the event types and event monitoring metrics corresponding to the key frames of the multiple surveillance videos. The surveillance event graph characterizes the overall characteristics of the relationships between the activity events of the target person at different locations in the community.

[0044] It should be noted that the surveillance event graph includes event nodes and edges connecting the event nodes. The attributes of the event nodes include event types and event monitoring metrics. Thus, the event types and event detection metrics of the target person can be represented by the event nodes. The edges connecting the event nodes are used to connect the event nodes, and the edges connecting the event nodes connect the event nodes in the time order of the key frames of the multiple surveillance videos. The attributes of the edges connecting the event nodes include the time span between the event nodes corresponding to the key frames of the surveillance videos. Thus, the time span between the key frames of the surveillance videos can be represented by the edges connecting the event nodes.

[0045] S130. Using the abnormal intrusion detection model, based on the surveillance event graph, determine the abnormal intrusion probability value of the target person. When the abnormal intrusion probability value of the target person is greater than or equal to the preset abnormal intrusion probability value, send out the abnormal intrusion prompt information corresponding to the target person. Among them, the abnormal intrusion detection model is a graph neural network model.

[0046] After obtaining the surveillance event graph, the abnormal intrusion probability value of the target person can be determined through the abnormal intrusion detection model based on the surveillance event graph. The abnormal intrusion probability value characterizes the abnormal intrusion probability of the target person in the community. The abnormal intrusion detection model is a graph neural network model. Through the graph neural network model, the features of the event nodes and the edges connecting the event nodes of the surveillance event graph can be extracted and recognized, and the relationships between the event nodes and the edges connecting the event nodes of the target person corresponding to the surveillance event graph can be recognized, that is, the feature recognition of different events and the time span between different events in the surveillance videos of the target person at different locations is realized. Furthermore, the recognition of the abnormal intrusion of the target person can be improved.

[0047] It should be noted that the abnormal intrusion detection model is a graph neural network model, and the abnormal intrusion detection model can be trained through the labeled surveillance event graph and the abnormal intrusion probability value of the person.

[0048] After obtaining the abnormal intrusion probability value of the target person, when the abnormal intrusion probability value of the target person is greater than or equal to the preset abnormal intrusion probability value, it indicates that the probability that the target person is an abnormal intruder is relatively high. At this time, the abnormal intrusion prompt information corresponding to the target person can be sent out to prompt the security personnel to handle or recheck.

[0049] The beneficial effects of the above implementation method are as follows. The monitoring event graph includes event nodes and edges connecting the event nodes. The attributes of the event nodes include event types and event monitoring metrics. The attributes of the edges connecting the event nodes include the time span between the event nodes corresponding to the key frames of the monitoring video. Through the abnormal intrusion detection model, according to the monitoring event graph, the abnormal intrusion probability value of the target person is determined, which improves the recognition accuracy of the abnormal intrusion probability of the target person.

[0050] Figure 2 FIG. is a schematic flow chart of a second abnormal intrusion monitoring method for a smart community provided by an embodiment of the present application. As Figure 2 shown, the above method includes S210 to S220, and the following is a specific description of S210 to S220.

[0051] S210. Obtain the regional scenes corresponding to the key frames of multiple monitoring videos respectively, and obtain the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of multiple adjacent monitoring videos in chronological order. Add the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of multiple monitoring videos to the attributes of the edges connecting the event nodes.

[0052] When performing abnormal intrusion recognition, the movement speed of the target person can be recognized to identify events such as the target person's ungrounded wandering and abnormal staying, so as to improve the recognition accuracy of abnormal intrusion. During recognition, the regional scenes corresponding to the key frames of multiple monitoring videos can be obtained, and the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of multiple adjacent monitoring videos in chronological order can be obtained. The maximum movement time, minimum movement time, and normal movement time between the regional scenes can be obtained through experiments by security personnel according to the maximum walking speed, minimum walking speed, and normal walking speed.

[0053] After obtaining the maximum movement time, minimum movement time, and normal movement time between the regional scenes, the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of multiple monitoring videos can be added to the attributes of the edges connecting the event nodes. Furthermore, the movement speed characteristics of the target person can be further judged according to the maximum movement time, minimum movement time, and normal movement time between the regional scenes, so as to identify events such as the target person's ungrounded wandering and abnormal staying.

[0054] S220. Through the abnormal intrusion detection model, according to the monitoring event graph, determine the abnormal intrusion probability value of the target person.

[0055] After adding the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of multiple surveillance videos to the attributes of the edges connecting the event nodes, the abnormal intrusion probability value of the target person can be determined according to the surveillance event graph through the abnormal intrusion detection model, realizing high-precision detection of the abnormal intrusion of the target person by combining the movement speed characteristics of the target person.

[0056] The beneficial effect of the above implementation method is that after adding the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of multiple surveillance videos to the attributes of the edges connecting the event nodes, high-precision detection of the abnormal intrusion of the target person is achieved by combining the movement speed characteristics of the target person.

[0057] Figure 3 As shown in the flowchart of the third abnormal intrusion monitoring method for a smart community provided by an embodiment of the present application, Figure 3 As shown, the above method includes S310 to S320, and the following is a specific description of S310 to S320.

[0058] S310. Determine the surveillance event graph at the moments corresponding to the key frames of multiple surveillance videos respectively. Through the abnormal intrusion detection model, determine the abnormal intrusion probability value of the target person corresponding to each surveillance event graph, and determine the cumulative value of the abnormal intrusion probability values.

[0059] When detecting the abnormal intrusion of the target person, in order to detect the abnormal intrusion of the target person in real time, the surveillance event graph at the moments corresponding to the key frames of multiple surveillance videos can be determined. Different key frames of the surveillance videos are at different moments, that is, the surveillance event graph corresponds to different moments. Furthermore, after updating a key frame of a surveillance video, the abnormal intrusion detection can be immediately performed through the abnormal intrusion detection model to achieve the detection of the abnormal intrusion of the target person.

[0060] After obtaining the surveillance event graph at the moments corresponding to the key frames of multiple surveillance videos respectively, the abnormal intrusion probability value of the target person corresponding to each surveillance event graph can be determined through the abnormal intrusion detection model. The abnormal intrusion probability value of the target person corresponding to each surveillance event graph represents the abnormal intrusion probability value at different moments.

[0061] After obtaining the abnormal intrusion probability values at different moments, the cumulative value of the abnormal intrusion probability values can be determined. The cumulative value of the abnormal intrusion probability values represents the sum of the abnormal intrusion probabilities of the target user at different moments.

[0062] S320. When the cumulative value of the abnormal intrusion probability value of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability value, send out the abnormal intrusion prompt information corresponding to the target person.

[0063] After obtaining the cumulative value of the abnormal intrusion probability value of the target person, when the cumulative value of the abnormal intrusion probability value of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability value, it indicates that the total intrusion probability of the target person at different times is too large. Furthermore, an abnormal intrusion prompt message corresponding to the target person can be sent to prompt the abnormal intrusion.

[0064] The beneficial effect of the above implementation method is that the cumulative value of the abnormal intrusion probability value represents the sum of the abnormal intrusion probabilities of the target user at different times, and the abnormal intrusion of the target user is accurately identified in real time through the cumulative value of the abnormal intrusion probability value.

[0065] Figure 4 As shown in the following, this is the flowchart of the fourth abnormal intrusion monitoring method for smart communities provided by the embodiments of this application. Figure 4 As shown, the above method includes S410 to S420, and the following will specifically describe S410 to S420.

[0066] S410: Obtain the area scenes to which the key frames of multiple surveillance videos respectively belong, obtain the intrusion probability weight value corresponding to each area scene, and determine the sum of the products of the abnormal intrusion probability value and the intrusion probability weight value corresponding to the area scenes to which the key frames of multiple surveillance videos respectively belong, as the cumulative value of the abnormal intrusion probability value.

[0067] When detecting the abnormal intrusion of a target person, the importance of different areas is different. Therefore, the key frames of the surveillance videos in different areas can be detected for intrusion according to different importance weights. Different areas can include key surveillance areas such as the areas around the surveillance room and the perimeter wall, and can also include areas of general importance such as the square and the community entrance.

[0068] When performing intrusion detection, the area scenes to which the key frames of multiple surveillance videos respectively belong can be obtained. The area scene is the area photographed by the cameras corresponding to the key frames of multiple surveillance videos, and the intrusion probability weight value corresponding to each area scene can be obtained. The intrusion probability weight value represents the importance characteristics corresponding to different area scenes.

[0069] After obtaining the intrusion probability weight value corresponding to each area scene, the sum of the products of the abnormal intrusion probability value and the intrusion probability weight value corresponding to the area scenes to which the key frames of multiple surveillance videos respectively belong can be determined, realizing the summation of the abnormal intrusion probability values corresponding to multiple area scenes according to the intrusion probability weight value, and taking it as the cumulative value of the abnormal intrusion probability value, realizing the adjustment of the abnormal intrusion detection weight of the key area.

[0070] S420. When the cumulative value of the abnormal intrusion probability value of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability value, an abnormal intrusion prompt message corresponding to the target person is issued.

[0071] After obtaining the cumulative value of the abnormal intrusion probability value of the target person, when the cumulative value of the abnormal intrusion probability value of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability value, an abnormal intrusion prompt message corresponding to the target person can be issued.

[0072] The beneficial effect of the above implementation method is that the abnormal intrusion probability values corresponding to multiple regional scenarios are summed according to the intrusion probability weight value and used as the cumulative value of the abnormal intrusion probability value, realizing the adjustment of the abnormal intrusion detection weight of key areas and improving the accuracy of abnormal intrusion detection of the target person.

[0073] Figure 5 It is a schematic flowchart of the fifth abnormal intrusion monitoring method for a smart community provided by an embodiment of the present application. As Figure 5 shown, the above method includes S510 to S520, and S510 to S520 will be specifically described below.

[0074] S510. Obtain the regional scenarios to which the key frames of multiple monitoring videos belong respectively, and obtain the key frame quantity threshold and time span threshold of the monitoring videos corresponding to each regional scenario. Determine the number of key frames of the monitoring videos within each regional scenario.

[0075] When monitoring different regions, different regions may have different requirements for the number of key frames and time span when the monitoring importance is different, so as to ensure the monitoring accuracy of regions with different importance.

[0076] When monitoring regions with different importance, the regional scenarios to which the key frames of multiple monitoring videos belong respectively can be obtained, and the key frame quantity threshold and time span threshold of the monitoring videos corresponding to each regional scenario can be obtained. The key frame quantity threshold and time span threshold of the monitoring videos corresponding to each regional scenario can be preset.

[0077] After obtaining the key frame quantity threshold of the monitoring videos corresponding to each regional scenario, the number of key frames of the monitoring videos within each regional scenario can be determined, and then abnormal intrusion recognition can be performed according to the number of key frames of the monitoring videos within each regional scenario.

[0078] S520. When the number of key frames of the monitoring video corresponding to the regional scenario is less than the key frame quantity threshold of the monitoring video, or the time span of the key frames of the monitoring video corresponding to the regional scenario is less than the preset time span threshold, delete the event node corresponding to the key frame of the monitoring video in the monitoring event graph.

[0079] When identifying abnormal intrusion, when the number of key frames of the monitoring video corresponding to the regional scene is less than the key frame number threshold of the monitoring video, it indicates that the number of key frames of the monitoring video in the regional scene does not meet the basic requirements for monitoring abnormal intrusion. At this time, in order to avoid introducing noise, the event nodes corresponding to the key frames of the monitoring video corresponding to the regional scene can be deleted in the monitoring event graph, so as to avoid the influence of abnormal intrusion caused by too few key frames of the monitoring video corresponding to the regional scene.

[0080] Similarly, when the time span of the key frames of the monitoring video corresponding to the regional scene is less than the preset time span threshold, it indicates that the time span of the key frames of the monitoring video corresponding to the regional scene does not meet the basic requirements for intrusion detection. Furthermore, the event nodes corresponding to the key frames of the monitoring video corresponding to the regional scene can be deleted in the monitoring event graph.

[0081] The beneficial effect of the above implementation method is to obtain the key frame number threshold and time span threshold of the monitoring video corresponding to each regional scene, and filter the key frames of the monitoring video corresponding to each regional scene through the key frame number threshold and time span threshold of the monitoring video corresponding to each regional scene, avoiding noise in abnormal intrusion detection and improving the accuracy of abnormal intrusion detection for target personnel.

[0082] In some implementation methods, the above method includes S610 to S620, and the following is a specific description of S610 to S620.

[0083] S610. Obtain the time periods to which the key frames of multiple monitoring videos belong, and obtain the time period adjustment factors corresponding to the time periods to which the key frames of multiple monitoring videos belong.

[0084] When detecting abnormal intrusion, the number of people in different time periods may be different, and the probability of possible abnormal intrusion may also be different. Therefore, the effect of abnormal intrusion detection can be adjusted according to the characteristics of the time period.

[0085] When adjusting the intrusion detection effect, the time periods to which the key frames of multiple monitoring videos belong can be obtained, and the time period adjustment factors corresponding to the time periods to which the key frames of multiple monitoring videos belong can be obtained. The time period adjustment factor characterizes the time adjustment characteristics of the time periods to which the key frames of multiple monitoring videos belong.

[0086] Exemplarily, the time period adjustment factor corresponding to the time period to which the key frame of the monitoring video belongs can be preset according to different time periods.

[0087] S620. Determine the product of the key frame number threshold and the time period adjustment factor for the surveillance video corresponding to each regional scenario to adjust the key frame number threshold. Determine the product of the time span threshold and the time period adjustment factor for the surveillance video corresponding to each regional scenario to adjust the time period adjustment factor.

[0088] After obtaining the time period adjustment factors corresponding to the time periods to which the key frames of multiple surveillance videos belong, it is possible to determine the product of the key frame number threshold and the time period adjustment factor for the surveillance video corresponding to each regional scenario to adjust the key frame number threshold, and thus be able to adjust the key frame number threshold according to the pedestrian flow characteristics of the belonging time period.

[0089] Similarly, it is possible to determine the product of the time span threshold and the time period adjustment factor for the surveillance video corresponding to each regional scenario to adjust the time period adjustment factor, and thus be able to adjust the time span threshold according to the pedestrian flow characteristics of the belonging time period.

[0090] The beneficial effect of the above implementation method is that the key frame number threshold is adjusted according to the pedestrian flow characteristics of the belonging time period, and the time span threshold is adjusted according to the pedestrian flow characteristics of the belonging time period, improving the accuracy of detecting abnormal intrusion of target personnel.

[0091] Figure 6 This is a flowchart of the sixth method for abnormal intrusion monitoring in a smart community provided by an embodiment of the present application. As Figure 6 shown, the above method includes S710 to S720, and the following will specifically describe S710 to S720.

[0092] S710. Determine multiple associated persons of the target person and construct surveillance event graphs respectively corresponding to the multiple associated persons.

[0093] When performing abnormal intrusion detection, it is possible to detect a group of abnormal intrusions to improve the detection effect of abnormal intrusion, and at the same time avoid insufficient accuracy in detecting abnormal intrusion of a single abnormal intrusion person.

[0094] When performing group abnormal intrusion detection, it is possible to determine multiple associated persons of the target person and construct surveillance event graphs respectively corresponding to the multiple associated persons. The surveillance event graphs respectively corresponding to the multiple associated persons are surveillance event graphs respectively corresponding to multiple associated persons related to the target person.

[0095] S720. Based on the monitoring event graphs respectively corresponding to the target person and multiple associated persons, determine the group abnormal intrusion probability values corresponding to the target person and the multiple associated persons through an abnormal intrusion detection model. When the group abnormal intrusion probability values corresponding to the target person and the multiple associated persons are greater than or equal to a preset group abnormal intrusion probability value, send out a group abnormal intrusion prompt message corresponding to the target person and the multiple associated persons.

[0096] When detecting the group abnormal intrusion probability values, through the abnormal intrusion detection model, based on the monitoring event graphs respectively corresponding to the target person and the multiple associated persons, determine the group abnormal intrusion probability values corresponding to the target person and the multiple associated persons. The group abnormal intrusion probability values represent the probability that the target person and the multiple associated persons are determined as abnormal intrusions as a whole.

[0097] Exemplarily, the group abnormal intrusion detection model can be trained through the monitoring event graphs respectively corresponding to the marked target person and the multiple associated persons, and the group intrusion probability value data.

[0098] After obtaining the group intrusion probability values, when the group abnormal intrusion probability values corresponding to the target person and the multiple associated persons are greater than or equal to the preset group abnormal intrusion probability value, it indicates that the group intrusion probability of the target person and the multiple associated persons is too high. At this time, a group abnormal intrusion prompt message corresponding to the target person and the multiple associated persons can be sent out.

[0099] The beneficial effect of the above implementation method is that by identifying the abnormal intrusion probability values respectively corresponding to the target person and the multiple associated persons through the monitoring event graphs respectively corresponding to the target person and the multiple associated persons, the group intrusion probability values corresponding to the target person and the multiple associated persons can be accurately identified, improving the monitoring accuracy of group abnormal intrusion.

[0100] In some implementation methods, in the above S710, determining the multiple associated persons of the target person includes S711 and S712.

[0101] S711. Construct the monitoring event graphs respectively corresponding to multiple persons, and determine the event nodes respectively corresponding to the same area scenarios in the monitoring event graphs of each person and the target person; determine the similarity of the event nodes respectively corresponding to each person and the target person in the same area scenario, and determine the sum of the similarities of all event nodes of each person and the target person as the person association degree.

[0102] When identifying the associated persons of a target person, multiple monitoring event graphs corresponding to different persons can be constructed, and the event nodes corresponding to the same area scenarios in the monitoring event graphs of each person and the target person are determined. The event nodes corresponding to the same area scenarios of the target person and the associated persons indicate that the target person and the associated persons appear in the same area at the same time. Therefore, there may be a possibility of group abnormal intrusion.

[0103] When identifying the associated persons of a target person, the similarity of the event nodes corresponding to the same area scenarios of each person and the target person can be determined, and the sum of the similarities of all the event nodes of each person and the target person is determined as the person association degree. The person association degree characterizes the possibility of group abnormal intrusion of each person and the target person in different areas.

[0104] S712. When the person association degree of the first person is greater than or equal to the preset person association degree, the first person is taken as the associated person of the target person.

[0105] After obtaining the person association degree of each person, when the person association degree of the first person is greater than or equal to the preset person association degree, the first person is taken as the associated person of the target person, and then the group abnormal intrusion probability of the associated persons can be identified.

[0106] The beneficial effect of the above implementation method is that the similarity of the event nodes corresponding to the same area scenarios of each person and the target person is determined, and the sum of the similarities of all the event nodes of each person and the target person is determined as the person association degree. When the person association degree of the first person is greater than or equal to the preset person association degree, the first person is taken as the associated person of the target person, which can identify the group abnormal intrusion probability of the associated persons.

[0107] In some implementation methods, the above method includes: obtaining the similarity adjustment factor corresponding to each area scenario, and determining the product of the similarity of the event nodes corresponding to the same area scenarios of each person and the target person and the similarity adjustment factor as the person association degree.

[0108] When determining the person association degree, since the characteristics of the flow of people in different areas are different. For example, people at the entrance of a community are more likely to gather, while people in the area around the community wall are not likely to gather. Therefore, different similarity adjustment factors can be set for different areas, and the similarity adjustment factor corresponding to each area scenario is obtained, and the product of the similarity of the event nodes corresponding to the same area scenarios of each person and the target person and the similarity adjustment factor is determined as the person association degree, so that the person association degree can be adjusted according to the area characteristics, improving the accuracy of identifying the associated persons of the target person and the accuracy of group abnormal intrusion.

[0109] The beneficial effects of the above implementation method are that the personnel correlation degree can be adjusted according to the regional characteristics, improving the recognition accuracy of the associated personnel of the target personnel and the accuracy of group abnormal intrusion.

[0110] The embodiment of the present application also provides an abnormal intrusion monitoring system for a smart community, including a unit for executing the method described in any one of the above.

[0111] Figure 7 It is a schematic logical structure diagram of an abnormal intrusion monitoring system for a smart community provided by an embodiment of the present application. As Figure 7 shown, the system 1 of this embodiment includes a processing unit 11, a storage unit 12, and a transceiver unit 13. The processing unit 11 is used to process data, the storage unit 12 is used to store data, and the transceiver unit 13 is used to transmit and receive data. The processing unit 11, the storage unit 12, and the transceiver unit 13 cooperate with each other to implement the above method. The beneficial effects of the embodiment of the present application have been described in the above method and will not be elaborated here.

[0112] It should be noted that for the information interaction, execution process, etc. between the above devices / units, since they are based on the same concept as the method embodiment of the present application, their specific functions and the technical effects brought can be specifically referred to the method embodiment part and will not be elaborated here.

[0113] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the above division of each functional unit and module is used for illustration. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working process of the units and modules in the above system can refer to the corresponding process in the foregoing method embodiment and will not be elaborated here.

[0114] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of this application, a computer program can be used to instruct relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunication signal.

[0115] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0116] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0117] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or units can be in an electrical, mechanical, or other form.

[0118] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0119] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An abnormal intrusion monitoring method for an intelligent community, characterized in that, The method includes: Obtain multiple surveillance videos corresponding to a target person captured by multiple cameras respectively, and determine the key frames of the multiple surveillance videos corresponding to the target person, the time sequence of the key frames of the multiple surveillance videos, and the time span between the key frames of the multiple surveillance videos; through an event recognition model, determine the event types and event monitoring metrics respectively corresponding to the key frames of the multiple surveillance videos according to the key frames of the multiple surveillance videos. Construct a surveillance event graph corresponding to the target person according to the event types and event monitoring metrics respectively corresponding to the key frames of the multiple surveillance videos; wherein, the surveillance event graph includes event nodes and edges connecting the event nodes, the attributes of the event nodes include event types and event monitoring metrics, the edges connecting the event nodes connect the event nodes in the time sequence of the key frames of the multiple surveillance videos, and the attributes of the edges connecting the event nodes include the time span between the event nodes corresponding to the key frames of the surveillance videos. Through an abnormal intrusion detection model, determine the abnormal intrusion probability value of the target person according to the surveillance event graph. When the abnormal intrusion probability value of the target person is greater than or equal to a preset abnormal intrusion probability value, send out an abnormal intrusion prompt message corresponding to the target person; wherein, the abnormal intrusion detection model is a graph neural network model.

2. The method according to claim 1, characterized in that The method includes: Obtain the regional scenes respectively corresponding to the key frames of the multiple surveillance videos, and obtain the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of the multiple surveillance videos adjacent in time sequence, and add the maximum movement time, minimum movement time, and normal movement time between the regional scenes corresponding to the key frames of the multiple surveillance videos to the attributes of the edges connecting the event nodes. Through an abnormal intrusion detection model, determine the abnormal intrusion probability value of the target person according to the surveillance event graph.

3. The method according to claim 2, characterized in that, The method includes: Determine the surveillance event graph at the moments respectively corresponding to the key frames of the multiple surveillance videos, through an abnormal intrusion detection model, determine the abnormal intrusion probability value of the target person corresponding to each surveillance event graph, and determine the cumulative value of the abnormal intrusion probability values. When the cumulative value of the abnormal intrusion probability values of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability values, send out an abnormal intrusion prompt message corresponding to the target person.

4. The method according to claim 3, characterized in that, The method includes: Obtain the regional scenes respectively belonging to the key frames of the multiple surveillance videos, and obtain the intrusion probability weight value corresponding to each regional scene, and determine the sum of the products of the abnormal intrusion probability values and the intrusion probability weight values corresponding to the regional scenes respectively belonging to the key frames of the multiple surveillance videos as the cumulative value of the abnormal intrusion probability values. When the cumulative value of the abnormal intrusion probability values of the target person is greater than or equal to the cumulative value of the preset abnormal intrusion probability values, send out an abnormal intrusion prompt message corresponding to the target person.

5. The method according to claim 4, wherein The method includes: Obtain the regional scenes respectively belonging to the key frames of the multiple surveillance videos, and obtain the key frame quantity threshold and time span threshold of the surveillance videos corresponding to each regional scene; determine the number of key frames of the surveillance videos within each regional scene. When the number of key frames of the surveillance video corresponding to the regional scene is less than the key frame number threshold of the surveillance video, or when the time span of the key frames of the surveillance video corresponding to the regional scene is less than the preset time span threshold, delete the event nodes corresponding to the key frames of the surveillance video corresponding to the regional scene in the surveillance event graph.

6. The method according to claim 5, wherein The method includes: Obtain the time periods to which the key frames of multiple surveillance videos belong, and obtain the time period adjustment factors corresponding to the time periods to which the key frames of multiple surveillance videos belong; Determine the product of the key frame number threshold and the time period adjustment factor corresponding to the surveillance video of each regional scene to adjust the key frame number threshold; determine the product of the time span threshold and the time period adjustment factor corresponding to the surveillance video of each regional scene to adjust the time period adjustment factor.

7. The method according to claim 6, wherein The method includes: Determine multiple associated persons of the target person, and construct surveillance event graphs corresponding to the multiple associated persons respectively; Through the abnormal intrusion detection model, according to the surveillance event graphs corresponding to the target person and the multiple associated persons respectively, determine the group abnormal intrusion probability values corresponding to the target person and the multiple associated persons. When the group abnormal intrusion probability value corresponding to the target person and the multiple associated persons is greater than or equal to the preset group abnormal intrusion probability value, send out the group abnormal intrusion prompt information corresponding to the target person and the multiple associated persons.

8. The method according to claim 7, characterized in that Determining multiple associated persons of the target person includes: Construct surveillance event graphs corresponding to multiple persons respectively, and determine the event nodes corresponding to the same regional scenes in the surveillance event graphs of each person and the target person; determine the similarity of the event nodes corresponding to the same regional scenes of each person and the target person, and determine the sum of the similarities of all the event nodes of each person and the target person as the person association degree; When the person association degree of the first person is greater than or equal to the preset person association degree, take the first person as an associated person of the target person.

9. The method according to claim 8, wherein The method includes: Obtain the similarity adjustment factor corresponding to each regional scene, and determine the product of the similarity of the event nodes corresponding to the same regional scenes of each person and the target person and the similarity adjustment factor as the person association degree.

10. An abnormal intrusion monitoring system for a smart community, characterized in that, It includes a unit for executing the method according to any one of claims 1 to 9.

Images