Event management method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides an event management method, and the method comprises the steps: obtaining a relation graph and / or a current tracking image of an absent person when the absent person appears in a roll call event in a specific region; carrying out analysis processing on the absence of the absence personnel based on at least one of the relation graph and the current tracking image to obtain an absence reason of the absence personnel; and managing the roll call event based on the absence reason. The relation graph and the current chasing image are integrated as the analysis basis of the absence reason, the accurate absence reason can be obtained through analysis, and in the analysis process, at least one analysis basis can be adopted for analysis, the absence reason can be analyzed under the condition of no human interference, so that the analysis efficiency is improved, and the user experience is improved. And the management efficiency of the roll call event can be improved.

Description

Technical Field

[0001] The present invention relates to the field of management, and in particular to an event management method, device, electronic equipment and storage medium. Background Art

[0002] With the advancement of science and technology, it is possible to use intelligent technology in specific areas. In specific areas, traditional roll call events are usually conducted manually. If there are absentees during the roll call process, the roll call personnel will analyze the reasons for the absence of the absentees based on historical roll call experience or manually. The analysis efficiency is not high, resulting in low management efficiency of roll call events. Therefore, how to provide an event management method that can improve the analysis efficiency of absence reasons in roll call events and improve the management efficiency of roll call events has become an urgent problem to be solved. Summary of the invention

[0003] The embodiment of the present invention provides an event management method, which aims to solve the problem of low event management efficiency of the existing event management method. Since the relationship map and the current tracking image are combined as the analysis basis of the absence reason, the accurate absence reason can be analyzed, and in the analysis process, at least one analysis basis can be used for analysis, and the absence reason can be analyzed without human intervention, thereby improving its analysis efficiency, and further improving the management efficiency of the roll call event.

[0004] In a first aspect, an embodiment of the present invention provides an event management method, the method comprising the following steps:

[0005] When an absent person appears in a roll call event in a specific area, a relationship map and / or a current tracking image of the absent person is obtained;

[0006] Analyzing and processing the absence of the absent person based on at least one of the relationship graph and the current tracking image to obtain the absence reason of the absent person;

[0007] Based on the absence reason, the roll call event is managed.

[0008] Optionally, the absent person is a specific person in the specific area, and before obtaining the relationship graph of the absent person, the method further includes:

[0009] Acquire personnel file data of each of the specific persons in the specific area before being in the specific area and historical tracking images of each of the specific persons when being in the specific area;

[0010] Based on the personnel profile data, determining the basic relationship network of each of the specific personnel;

[0011] Based on the historical tracking images, determining a specific relationship network of each specific person, wherein the specific relationship network includes a relationship between the specific person and other specific persons;

[0012] Based on the basic relationship network and the specific relationship network of each specific person, a relationship graph of each specific person is constructed;

[0013] A search process is performed in the relationship graphs of all the specific personnel to obtain the relationship graph of the absent personnel.

[0014] Optionally, determining the basic relationship network of each specific person based on the personnel profile data includes:

[0015] Preprocessing the personnel file data to obtain preprocessed personnel file data, wherein the preprocessing includes cleaning and standardization, and the preprocessed personnel file data includes at least one category of sub-file data;

[0016] Based on the category of the sub-archive data, construct corresponding data nodes and associated labels;

[0017] An association relationship is established between the data node and the corresponding association tag to obtain the basic relationship network.

[0018] Optionally, determining the specific relationship network of each specific person based on the historical tracking images includes:

[0019] Performing behavior recognition processing based on the historical tracking images to obtain behavior interaction events between each of the specific persons and other specific persons;

[0020] Based on the behavioral interaction events, an event relationship network of each of the specific persons is constructed;

[0021] Based on the event relationship network, counting the relationship score between each of the specific persons and other specific persons;

[0022] Based on the relationship score, establishing an association relationship between each specific person and other specific persons;

[0023] Based on the association relationship, a specific relationship network of each specific person is constructed.

[0024] Optionally, the analyzing and processing the absence of the absentee based on at least one of the relationship graph and the current tracking image to obtain the absence reason of the absentee includes:

[0025] When there are multiple absent persons, determining the personnel relationships between the absent persons based on the relationship map;

[0026] The absence of the absent person is analyzed and processed based on at least one of the current tracking image and the personnel relationship to obtain the absence reason of the absent person.

[0027] Optionally, the personnel relationship includes a hostile relationship and a non-hostile relationship, and the analyzing and processing the absence of the absent person based on at least one of the current tracking image and the personnel relationship to obtain the absence reason of the absent person includes:

[0028] When the personnel relationship between the absent persons is the non-hostile relationship, a behavior recognition process is performed based on the current tracking image to obtain a behavior recognition result of the absent persons, and based on the behavior recognition result, the absence reason of the absent persons is determined;

[0029] When the personnel relationship between the absent personnel is the hostile relationship, absence alarm information corresponding to the absent personnel is generated based on the hostile relationship, and the absence alarm information is fed back to the management personnel of the specific area, so that the management personnel can upload the absence reasons of the absent personnel after managing the absent personnel.

[0030] Optionally, managing the roll call event based on the absence reason includes:

[0031] Determine the impact of the absence reason on the specific area;

[0032] Based on the influence, determining the absence points of the absent personnel in the roll call event;

[0033] Based on the absence points and the relationship graph, generating an event report corresponding to the roll call event;

[0034] Based on the event report, the roll call event is managed.

[0035] In a second aspect, an embodiment of the present invention further provides an event management device, the event management device comprising:

[0036] A first acquisition module is used to acquire a relationship map and / or a current tracking image of an absent person when an absent person appears in a roll call event in a specific area;

[0037] a first analysis module, configured to analyze and process the absence of the absentee based on at least one of the relationship graph and the current tracking image, and obtain the absence reason of the absentee;

[0038] The first management module is used to manage the roll call event based on the absence reason.

[0039] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the event management method provided in the embodiment of the present invention when executing the computer program.

[0040] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the event management method provided in the embodiment of the invention are implemented.

[0041] In an embodiment of the present invention, when an absentee appears in a roll call event in a specific area, the relationship map and / or the current tracking image of the absentee are obtained; the absence of the absentee is analyzed and processed based on at least one of the relationship map and the current tracking image to obtain the absence reason of the absentee; and the roll call event is managed based on the absence reason. Since the relationship map and the current tracking image are combined as the analysis basis for the absence reason, the accurate absence reason can be analyzed, and in the analysis process, at least one analysis basis can be used for analysis, and the absence reason analysis can be realized without human intervention, thereby improving its analysis efficiency, and further improving the management efficiency of the roll call event. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 is a flow chart of an event management method provided by an embodiment of the present invention;

[0044] Figure 2 is a structural diagram of an event management device provided in an embodiment of the present invention;

[0045] Figure 3 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] like Figure 1 As shown, Figure 1 is a flow chart of an event management method provided by an embodiment of the present invention, including:

[0048] 101. When an absent person appears in a roll call event in a specific area, a relationship map and / or a current tracking image of the absent person is obtained.

[0049] In an embodiment of the present invention, the above event management method can be applied to an event management platform, which can be constructed by a server or a server cluster, and the server or server cluster can be any electronic device with functions such as data processing, data analysis, data storage, and data transmission. Specifically, when an absent person appears in a roll call event, the above event management platform can execute any event management method provided in the embodiment of the present invention to achieve the management of the roll call event.

[0050] The above-mentioned characteristic area can be an area where specific personnel are centrally managed, for example, it can be a closed school, a centralized management place for personnel who violate laws and regulations, etc. The above-mentioned current tracking image can be an image within a preset time period before and after the reference point on the timeline when the above-mentioned roll call event occurs as a reference point. For example, the time when the absent person appears in the roll call event is 21:00, then the above-mentioned current tracking image can be a tracking image of the absent person in the time period from 20:50 to 21:10. The above-mentioned preset time period can be set based on historical event management experience or obtained through a limited number of experiments.

[0051] It should be noted that the absent persons mentioned above are specific persons, and the specific persons mentioned above can be understood as persons who are subject to centralized management in the specific areas mentioned above. When the specific persons mentioned above enter the specific areas mentioned above to be subject to centralized management, the basic relationship network of the specific persons mentioned above can be constructed through the personnel file data of the specific persons before the specific areas mentioned above, and after the specific persons have been in the specific areas for the first time, the image acquisition device deployed in the specific areas can be used to acquire historical tracking images of the specific persons when they were in the specific areas, and the specific relationship network of the specific persons mentioned above can be constructed through the historical tracking images. The specific relationship network includes the relationship between the specific persons and other specific persons, and the relationship map mentioned above is constructed based on the basic relationship network and the specific relationship network. The relationship map mentioned above is stored in the event management platform, and when an absent person appears in a roll call event, the relationship map corresponding to the absent person can be extracted.

[0052] Furthermore, historical tracking images may be collected at regular intervals, and the specific relationship network may be reconstructed through the historical tracking images, and the relationship map may be updated according to the specific relationship network to ensure the accuracy and timeliness of the relationship map.

[0053] 102. Analyze and process the absence of the absent person based on at least one of the relationship graph and the current tracking image to obtain the absence reason of the absent person.

[0054] In an embodiment of the present invention, the absentee may be one or more than one. When there is only one absentee, the current tracking image of the absentee can be obtained, and behavior recognition processing can be performed based on the current tracking image to obtain the behavior recognition result of the absentee. According to the behavior recognition result, the reason for the absence of the absentee is determined. For example, if the category of the behavior recognition result is an abnormal behavior event, and the abnormal behavior event can specifically be a fight, then the reason for the absence can be that the absentee fought during the roll call event, resulting in absence. Alternatively, if the category of the behavior recognition result is a non-abnormal behavior event, and the non-abnormal behavior event can specifically be going to the toilet, then the reason for the absence can be that the absentee was going to the toilet during the roll call event, resulting in absence.

[0055] When there are multiple absent persons, the relationship map of each absent person can be obtained, and the personnel relationships between the absent persons can be determined based on the relationship map of each absent person. The personnel relationships can include hostile relationships and non-hostile relationships. When the personnel relationships are hostile relationships, the hostile relationships of the absent persons can be directly fed back to the managers of the specific area, and the managers will manage the absent persons and then feedback the above-mentioned reasons for absence.

[0056] Alternatively, when the personnel relationship is non-hostile, the current tracking image of each absent person can be obtained, and behavior recognition processing can be performed based on the current tracking image to obtain the behavior recognition result of the absent person, and the reason for the absence of the absent person can be determined based on the behavior recognition result.

[0057] It should be noted that when there are multiple absentees and the absentees are hostile to each other, the probability of a dangerous incident is high. Therefore, the analysis process can be simplified and the management personnel can be notified in time to manage in order to avoid the occurrence of dangerous incidents. When the absentees are not hostile to each other, the behavior recognition processing can be directly performed based on the current tracking image to realize the analysis of the cause of absence.

[0058] 103. Manage roll call events based on the reason for absence.

[0059] In an embodiment of the present invention, an absence record table of the absent personnel may also be stored in the event management platform. The absence record table may record the number of absences and absence points of the absent personnel. The absence points may consist of abnormal absence points and non-abnormal absence points.

[0060] Specifically, when the absence is caused by an abnormal behavior event, the abnormal absence points of the absentee can be increased accordingly, and when the absence is caused by a non-abnormal behavior event, the non-abnormal absence points of the absentee can be increased accordingly to complete the above management.

[0061] Furthermore, in the absence record table of the absentee, when the number of absences meets the preset number threshold, and the abnormal absence score is greater than the preset first score threshold or the non-abnormal absence score is greater than the preset second score threshold, an alarm message can be sent to the managers of the specific area to remind the above-mentioned managers to pay attention to the management of the above-mentioned absentee.

[0062] In an embodiment of the present invention, when an absentee appears in a roll call event in a specific area, the relationship map and / or the current tracking image of the absentee are obtained; the absence of the absentee is analyzed and processed based on at least one of the relationship map and the current tracking image to obtain the absence reason of the absentee; and the roll call event is managed based on the absence reason. Since the relationship map and the current tracking image are combined as the analysis basis for the absence reason, the accurate absence reason can be analyzed, and in the analysis process, at least one analysis basis can be used for analysis, and the absence reason analysis can be realized without human intervention, thereby improving its analysis efficiency, and further improving the management efficiency of the roll call event.

[0063] It is understandable that in the specific implementation of the present application, related data such as relationship maps, current tracking images, historical tracking images, personnel file data, etc. are involved. When the embodiments in the present application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data and the construction, training and use of related models need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0064] It should be noted that the event management method provided in the embodiment of the present invention can be applied to devices such as computers and servers that can perform event management.

[0065] Optionally, before the step of obtaining the relationship map of absent persons, the personnel file data of each specific person in the specific area before being in the specific area and the historical tracking images of each specific person when being in the specific area can also be obtained; based on the personnel file data, the basic relationship network of each specific person is determined; based on the historical tracking images, the specific relationship network of each specific person is determined; based on the basic relationship network and the specific relationship network of each specific person, a relationship map of each specific person is constructed; and search processing is performed in the relationship maps of all specific persons to obtain the relationship map of the absent persons.

[0066] In an embodiment of the present invention, the absentee is a specific person in a specific area, and the specific relationship network includes the relationship between the specific person and other specific persons. The personnel profile data may include data such as name, age, gender, family members, work address before entering the specific area, home address, and room number after entering the specific area. Based on the above data, a basic relationship network for each specific person may be constructed, and the basic relationship network may include the kinship, work relationship, and residence relationship of the specific person.

[0067] The specific relationship network can be obtained by performing behavior recognition processing based on the historical tracking images to obtain behavior interaction events between specific persons, and according to the categories of the behavior interaction events, determine the personnel relationships (including hostile relationships or non-hostile relationships) between the specific persons. Based on the personnel relationships, a specific relationship network for each specific person is constructed.

[0068] After obtaining the above basic relationship network and the above specific relationship network, the above basic relationship network and the above specific relationship network can be combined to obtain the above relationship map. It should be noted that if there is an overlapping part between the above basic relationship network and the above specific relationship network, the weight of the overlapping part can be increased after the combination. In the relationship map, the higher the weight, the higher the credibility, and vice versa.

[0069] Specifically, an ArangoDB graph database can be deployed on the above-mentioned event management platform, and the above-mentioned relationship graphs of all specific personnel can be stored in the above-mentioned ArangoDB graph database, and a unique aid is set for each specific person to concatenate ArangoDB and es as a unique id. When there are absent personnel, the relationship graph of the absent personnel can be quickly retrieved through the above-mentioned unique id.

[0070] Optionally, in the step of determining the basic relationship network of each specific person based on the personnel file data, the personnel file data can also be preprocessed to obtain preprocessed personnel file data; based on the category of the sub-file data, corresponding data nodes and associated labels are constructed; and an association relationship is constructed between the data nodes and the corresponding associated labels to obtain a basic relationship network.

[0071] In an embodiment of the present invention, preprocessing includes cleaning processing and standardization processing, and the preprocessed personnel file data includes at least one category of sub-file data. For example, the above-mentioned name, age, gender, family members, work address before entering a specific area, home address, living room number after entering a specific area and other data, each data is a category of sub-file data.

[0072] When the above-mentioned specific personnel enter the above-mentioned specific areas for centralized management, they can comprehensively collect and input data such as clear facial images, home addresses, work addresses, relatives, and room numbers of the specific personnel living in the specific areas. The collected data is cleaned to remove duplicate, erroneous or invalid data. The data is standardized to unify the address format and relative relationship description to obtain pre-processed personnel file data. The above pre-processing can ensure the consistency of the data.

[0073] Specifically, you can use the ArangoDB graph database as a relational graph library to create nodes (Vertices): create a node for each specific person, specific person's relatives, home address, work address, room number living in a specific area, and store the corresponding label information.

[0074] More specifically, based on the pre-processed personnel profile data, relationships can be established between nodes, including the "residence" relationship between a specific person and his / her family residence, the "relative" relationship between a specific person and his / her relatives, the "work" relationship between a specific person and his / her work address, and the "belonging" relationship between a specific person and the room number in a specific area. Using ArangoDB's graph query function, all associated nodes and their relationships between specific persons can be easily retrieved. At the same time, a tag library is built for each category (family residence, work address, and relative relationship) based on the collected information. The tag library can be used to quickly retrieve and classify data. At the same time, in ArangoDB, each node is associated with the corresponding tag to support subsequent relationship graph query and analysis.

[0075] Optionally, in the step of determining the specific relationship network of each specific person based on historical tracking images, behavior recognition processing can also be performed based on the historical tracking images to obtain behavior interaction events between each specific person and other specific persons; based on the behavior interaction events, an event relationship network for each specific person is constructed; based on the event relationship network, the relationship points between each specific person and other specific persons are counted; based on the relationship points, an association relationship is established between each specific person and other specific persons; based on the association relationship, a specific relationship network for each specific person is constructed.

[0076] In an embodiment of the present invention, an image acquisition device (or a high-definition camera) may be deployed in each key area of ​​the above-mentioned specific area to ensure that the facial features of the person can be clearly captured. The above-mentioned image acquisition device is connected to the above-mentioned event management platform via a wired or wireless method, and the above-mentioned image acquisition device can acquire the above-mentioned historical tracking images and store the above-mentioned historical tracking images, or can be directly uploaded to the above-mentioned event management platform via the above-mentioned wired or wireless method and stored by the above-mentioned event management platform. Specifically, when the image acquisition device has an acquisition abnormality or is offline, the above-mentioned event management platform can prompt the management personnel of the above-mentioned specific area through SMS or web terminal to make timely repairs.

[0077] It is understandable that, based on the historical tracking images stored in the above-mentioned event management platform, the specific relationship network of each time period can be determined at regular intervals, and the relationship map can be updated according to the above-mentioned specific relationship network. After the update is completed, the historical tracking images of the previous time period can be deleted to reduce the required storage space.

[0078] The above-mentioned behavior recognition processing may include peer event recognition and fight event recognition, and the above-mentioned behavior interaction events may correspond to peer events and fight events. The above-mentioned behavior recognition processing may be implemented by a structured engine deployed on the above-mentioned event management platform. The above-mentioned structured engine uses a deep learning algorithm to build a feature value extraction model for face recognition, extracts face feature information and pushes it to the distributed stream message temporary storage platform kafka, and enters PostgreSQL for subsequent feature value similarity retrieval.

[0079] Furthermore, the event stream of aid corresponding to each captured face can be archived through the big data platform archiving engine, and the data mining engine can analyze peer events through the Azkaban scheduling spark task. The algorithm warehouse platform can identify the captured pictures through the long-tail algorithm to analyze the fighting events and combine the big data platform archiving engine to output the event information of the fighting personnel. After obtaining information such as behavioral events, fighting events, and aid, the above information can be output to ES (Elasticsearch), which can be understood as an engine with search, analysis and other functions. After receiving the above information, the above ES can analyze the reasons for absence.

[0080] Specifically, the number of incidents of traveling together and the number and influence of fighting incidents between a specific person and other specific persons can be counted. When the number of incidents of traveling together between a specific person and another specific person reaches a preset threshold, a label of a close relationship can be attached to the specific person and another specific person.

[0081] Or when the number of fighting events reaches the above-preset number threshold, and at the same time, the influence of any fighting event reaches the preset influence threshold, the above-mentioned specific person and the above-mentioned other specific person may be labeled as having a hostile relationship. It should be noted that the above-mentioned influence may be a quantification of the severity of the above-mentioned fighting event. For example, the more people involved in the above-mentioned fighting event and the higher the degree of casualties, the higher the above-mentioned influence. On the contrary, the fewer people involved in the above-mentioned fighting event and the lower the degree of casualties, the lower the above-mentioned influence.

[0082] Alternatively, when the number of fighting events does not meet the preset number threshold or the influence does not meet the preset influence threshold, and the number of traveling events does not meet the preset number threshold, the above-mentioned specific person and one of the above-mentioned other specific persons are labeled as strangers.

[0083] The labels of the above-mentioned unfamiliar relationships, the labels of the above-mentioned close relationships, and the labels of the above-mentioned hostile relationships are used as the association relationships between the above-mentioned specific persons and other specific persons, and the above-mentioned specific relationship network is constructed according to the above-mentioned management relationship.

[0084] Optionally, in the step of analyzing and processing the absence of the absent person based on the relationship map and at least one of the current tracking image to obtain the reason for the absence of the absent person, it is also possible to determine the personnel relationship between the absent persons based on the relationship map when there are multiple absent persons; and analyze and process the absence of the absent person based on the current tracking image and at least one of the personnel relationship to obtain the reason for the absence of the absent person.

[0085] In an embodiment of the present invention, when there are multiple absent persons, the personnel relationships between the absent persons can be queried in the above-mentioned relationship graph based on the above-mentioned graph query function using ArangoDB, and the above-mentioned personnel relationships may include hostile relationships and non-hostile relationships.

[0086] When the personnel relationship is a non-hostile relationship, behavior recognition processing can be performed based on the current tracking image to obtain a corresponding behavior recognition result, and the absence reason of the absent person can be determined based on the behavior recognition result.

[0087] When the personnel relationship is a hostile one, corresponding alarm information can be generated based on the above personnel relationship and the name of the absent person and the room number where the person lives in the specific area, and the alarm information can be sent to the management personnel of the specific area, so that the management personnel can manage the absent persons and upload the reasons for the absence to avoid the occurrence of dangerous incidents.

[0088] Optionally, in the step of analyzing and processing the absence of the absent personnel based on at least one of the current tracking image and the personnel relationship to obtain the reason for the absence of the absent personnel, it is also possible to perform behavior recognition processing based on the current tracking image to obtain the behavior recognition result of the absent personnel when the personnel relationship between the absent personnel is a non-hostile relationship, and determine the reason for the absence of the absent personnel based on the behavior recognition result; when the personnel relationship between the absent personnel is a hostile relationship, generate absence alarm information corresponding to the absent personnel based on the hostile relationship, and feed back the absence alarm information to the management personnel of the specific area, so that the management personnel can upload the reason for the absence of the absent personnel after managing the absent personnel.

[0089] In an embodiment of the present invention, personnel relationships include hostile relationships and non-hostile relationships, and the above-mentioned non-hostile relationships may include close relationships and unfamiliar relationships. The above-mentioned behavior recognition processing includes any one of the behavior recognitions such as peer behavior recognition, fighting behavior recognition, toilet behavior recognition, and eating behavior recognition, and the corresponding above-mentioned behavior recognition results may include peer events, fighting events, eating events, toilet events, and recognition failures. It can be understood that the above-mentioned recognition failure may be a recognition failure caused by the inability to obtain the current tracking image of the absent person, or it may be a recognition failure caused by the poor image quality of the current tracking image of the absent person. The above-mentioned behavior recognition processing can also be implemented by the above-mentioned structured engine.

[0090] For example, when the relationship between the absent persons is a close one, it can be immediately determined whether the behavior recognition result of each absent person is a recognition failure. When the behavior recognition results of all absent persons are recognition failures, a pop-up reminder and text message reminder can be immediately sent to the manager of the specific area to immediately trigger a search of the rooms where the absent persons live in the specific area, and the manager needs to submit the reason for the failure to name the person. When the behavior recognition result of any absent person is a recognition failure, the absent person can be immediately reminded of the information of the known location of the person (that is, the location information of the known absent person), and the manager can immediately go to communicate the location information of the close persons for subsequent management. When the behavior recognition results of all absent persons are not recognition failures, the reason for the absence of the absent person can be determined based on the behavior recognition results.

[0091] When the relationship between the absent persons is hostile, an alarm page message and a text message may be generated to remind the management personnel that the absent person may cooperate with other absent persons in a fight and the management personnel may be dispatched to deal with the incident immediately.

[0092] When the absent persons are strangers to each other, the determination of the reason for their absence is consistent with the above-mentioned determination of close relationships, and the effect is the same.

[0093] Optionally, in the step of managing roll call events based on absence reasons, the influence of absence reasons on specific areas can also be determined; based on the influence, the absence points of absent persons in the roll call event are determined; based on the absence points and the relationship map, an event report corresponding to the roll call event is generated; and based on the event report, the roll call event is managed.

[0094] In the embodiment of the present invention, the absence points may include abnormal absence points and non-abnormal absence points, and the influence may be a quantification of the degree of adverse impact on the management, peace and harmony of a specific area.

[0095] Specifically, when the absence is caused by an abnormal behavior event, the abnormal absence score can be increased accordingly, and when the absence is caused by a non-abnormal behavior event, the non-abnormal absence score can be increased accordingly. It can be understood that initially, the abnormal absence score and the non-abnormal absence score are both 0.

[0096] For example, in the tth absence event, the abnormal absence score of the absentee is n, and the non-abnormal absence score is also n. If in the t+1th absence event, the absence is caused by an abnormal behavior event, then the abnormal absence score can be increased by n based on the above influence. At this time, the abnormal absence score of the absentee is n+1, and the non-abnormal absence score remains unchanged at n.

[0097] More specifically, after determining the absence score, the personnel relationship between the absent persons can be determined according to the relationship map. The event report is generated according to the personnel relationship and the absence score. Specifically, the event report can be implemented by any large language model.

[0098] After the above-mentioned event report is generated, the above-mentioned event report can be fed back to the management personnel of the above-mentioned specific area, so that the management personnel can conduct internal discussions and isolate the basis for the management of absent personnel.

[0099] It should be noted that the above-mentioned managers can manually upload the abnormal behavior events and non-abnormal behavior events of any specific person. After receiving the above-mentioned abnormal behavior events and non-abnormal behavior events, the event management platform can also adjust the above-mentioned abnormal absence points and / or the above-mentioned non-abnormal absence points based on the above-mentioned abnormal behavior events and non-abnormal behavior events.

[0100] like Figure 2 As shown, the embodiment of the present invention also provides an event management device, including:

[0101] The first acquisition module 201 is used to acquire the relationship map and / or the current tracking image of the absent person when there is an absent person in the roll call event in the specific area;

[0102] A first analysis module 202, configured to analyze and process the absence of the absentee based on at least one of the relationship graph and the current tracking image, and obtain the absence reason of the absentee;

[0103] The first management module 203 is used to manage the roll call event based on the absence reason.

[0104] Optionally, the absent person is a specific person in the specific area, and the event management device further includes:

[0105] A second acquisition module is used to acquire personnel file data of each of the specific persons in the specific area before they are in the specific area and historical tracking images of each of the specific persons when they are in the specific area;

[0106] A first determination module, configured to determine a basic relationship network of each of the specific personnel based on the personnel profile data;

[0107] A second determination module is used to determine a specific relationship network of each specific person based on the historical tracking images, wherein the specific relationship network includes a relationship between the specific person and other specific persons;

[0108] A first construction module is used to construct a relationship graph of each specific person based on the basic relationship network and the specific relationship network of each specific person;

[0109] The first retrieval module is used to perform retrieval processing in the relationship graphs of all the specific personnel to obtain the relationship graphs of the absent personnel.

[0110] Optionally, the first determining module includes:

[0111] A preprocessing submodule, used for preprocessing the personnel file data to obtain preprocessed personnel file data, wherein the preprocessing includes cleaning and standardization, and the preprocessed personnel file data includes at least one category of sub-file data;

[0112] A first construction submodule is used to construct corresponding data nodes and associated labels based on the categories of the sub-archive data;

[0113] The second construction submodule is used to construct an association relationship between the data node and the corresponding association tag to obtain the basic relationship network.

[0114] Optionally, the second determining module includes:

[0115] A first recognition submodule is used to perform behavior recognition processing based on the historical tracking images to obtain behavior interaction events between each of the specific persons and other specific persons;

[0116] A third construction submodule is used to construct an event relationship network of each of the specific persons based on the behavior interaction events;

[0117] A first statistical submodule, for counting the relationship score between each of the specific persons and other specific persons based on the event relationship network;

[0118] A first establishing submodule is used to establish an association relationship between each of the specific persons and other specific persons based on the relationship score;

[0119] The fourth construction submodule is used to construct a specific relationship network of each specific person based on the association relationship.

[0120] Optionally, the first analysis module 202 includes:

[0121] A first determination submodule is used to determine the personnel relationship between the absent personnel based on the relationship map when there are multiple absent personnel;

[0122] The first analysis submodule is used to analyze and process the absence of the absent person based on at least one of the current tracking image and the personnel relationship to obtain the absence reason of the absent person.

[0123] Optionally, the personnel relationship includes a hostile relationship and a non-hostile relationship, and the first analysis submodule includes:

[0124] a first processing unit, configured to, when the personnel relationship between the absent persons is the non-hostile relationship, perform behavior recognition processing based on the current tracking image to obtain a behavior recognition result of the absent persons, and determine the absence reason of the absent persons based on the behavior recognition result;

[0125] The second processing unit is used to generate absence alarm information corresponding to the absent personnel based on the hostile relationship when the personnel relationship between the absent personnel is the hostile relationship, and to feed back the absence alarm information to the management personnel of the specific area, so that the management personnel can upload the absence reasons of the absent personnel after managing the absent personnel.

[0126] Optionally, the first management module 203 includes:

[0127] A second determination submodule is used to determine the influence of the absence reason on the specific area;

[0128] A third determination submodule is used to determine the absence points of the absent personnel in the roll call event based on the influence;

[0129] A first generating submodule, configured to generate an event report corresponding to the roll call event based on the absence score and the relationship map;

[0130] The first management submodule is used to manage the roll call event based on the event report.

[0131] like Figure 3 As shown, an embodiment of the present invention further provides an electronic device, characterized in that it includes a processor, and the processor can execute any one of the above event management methods.

[0132] Specifically, it includes a processor 301 and a memory 302, and a computer program for executing the event management method stored in the memory 302 and capable of running on the processor 301, wherein:

[0133] The processor 301 runs the computer program of the event management method stored in the memory 302 to perform the following steps:

[0134] When an absent person appears in a roll call event in a specific area, a relationship map and / or a current tracking image of the absent person is obtained;

[0135] Analyzing and processing the absence of the absent person based on at least one of the relationship graph and the current tracking image to obtain the absence reason of the absent person;

[0136] Based on the absence reason, the roll call event is managed.

[0137] Optionally, the absent person is a specific person in the specific area, and before obtaining the relationship graph of the absent person, the method executed by the processor 301 further includes:

[0138] Acquire personnel file data of each of the specific persons in the specific area before being in the specific area and historical tracking images of each of the specific persons when being in the specific area;

[0139] Based on the personnel profile data, determining the basic relationship network of each of the specific personnel;

[0140] Based on the historical tracking images, determining a specific relationship network of each specific person, wherein the specific relationship network includes a relationship between the specific person and other specific persons;

[0141] Based on the basic relationship network and the specific relationship network of each specific person, a relationship graph of each specific person is constructed;

[0142] A search process is performed in the relationship graphs of all the specific personnel to obtain the relationship graph of the absent personnel.

[0143] Optionally, the determining of the basic relationship network of each specific person based on the personnel profile data performed by the processor 301 includes:

[0144] Preprocessing the personnel file data to obtain preprocessed personnel file data, wherein the preprocessing includes cleaning and standardization, and the preprocessed personnel file data includes at least one category of sub-file data;

[0145] Based on the category of the sub-archive data, construct corresponding data nodes and associated labels;

[0146] An association relationship is established between the data node and the corresponding association tag to obtain the basic relationship network.

[0147] Optionally, the determining of the specific relationship network of each specific person based on the historical tracking images performed by the processor 301 includes:

[0148] Performing behavior recognition processing based on the historical tracking images to obtain behavior interaction events between each of the specific persons and other specific persons;

[0149] Based on the behavioral interaction events, an event relationship network of each of the specific persons is constructed;

[0150] Based on the event relationship network, counting the relationship score between each of the specific persons and other specific persons;

[0151] Based on the relationship score, establishing an association relationship between each specific person and other specific persons;

[0152] Based on the association relationship, a specific relationship network of each specific person is constructed.

[0153] Optionally, the processor 301 performs the analyzing and processing of the absence of the absentee based on at least one of the relationship map and the current tracking image to obtain the absence reason of the absentee, including:

[0154] When there are multiple absent persons, determining the personnel relationships between the absent persons based on the relationship map;

[0155] The absence of the absent person is analyzed and processed based on at least one of the current tracking image and the personnel relationship to obtain the absence reason of the absent person.

[0156] Optionally, the personnel relationship includes a hostile relationship and a non-hostile relationship, and the processor 301 performs the analysis and processing of the absence of the absent person based on at least one of the current tracking image and the personnel relationship to obtain the absence reason of the absent person, including:

[0157] When the personnel relationship between the absent persons is the non-hostile relationship, a behavior recognition process is performed based on the current tracking image to obtain a behavior recognition result of the absent persons, and based on the behavior recognition result, the absence reason of the absent persons is determined;

[0158] When the personnel relationship between the absent personnel is the hostile relationship, absence alarm information corresponding to the absent personnel is generated based on the hostile relationship, and the absence alarm information is fed back to the management personnel of the specific area, so that the management personnel can upload the absence reasons of the absent personnel after managing the absent personnel.

[0159] Optionally, the managing of the roll call event based on the absence reason performed by the processor 301 includes:

[0160] Determine the impact of the absence reason on the specific area;

[0161] Based on the influence, determining the absence points of the absent personnel in the roll call event;

[0162] Based on the absence points and the relationship graph, generating an event report corresponding to the roll call event;

[0163] Based on the event report, the roll call event is managed.

[0164] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the computer program implements the various processes of the event management method or the application-side event management method provided by the embodiment of the present invention, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0165] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the above-mentioned computer program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the above-mentioned computer-readable storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0166] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. An event management method, characterized in that: The method comprises the following steps: When an absent person appears in a roll call event in a specific area, a relationship map and / or a current tracking image of the absent person is obtained; Analyzing and processing the absence of the absent person based on at least one of the relationship graph and the current tracking image to obtain the absence reason of the absent person; Based on the absence reason, the roll call event is managed.

2. The event management method according to claim 1, characterized in that: The absent person is a specific person in the specific area. Before obtaining the relationship graph of the absent person, the method further includes: Acquire personnel file data of each of the specific persons in the specific area before being in the specific area and historical tracking images of each of the specific persons when being in the specific area; Based on the personnel profile data, determining the basic relationship network of each of the specific personnel; Based on the historical tracking images, determining a specific relationship network of each specific person, wherein the specific relationship network includes a relationship between the specific person and other specific persons; Based on the basic relationship network and the specific relationship network of each specific person, a relationship graph of each specific person is constructed; A search process is performed in the relationship graphs of all the specific personnel to obtain the relationship graph of the absent personnel.

3. The event management method according to claim 2, characterized in that: Determining the basic relationship network of each specific person based on the personnel profile data includes: Preprocessing the personnel file data to obtain preprocessed personnel file data, wherein the preprocessing includes cleaning and standardization, and the preprocessed personnel file data includes at least one category of sub-file data; Based on the category of the sub-archive data, construct corresponding data nodes and associated labels; An association relationship is established between the data node and the corresponding association tag to obtain the basic relationship network.

4. The event management method according to claim 2, characterized in that: The determining of the specific relationship network of each specific person based on the historical tracking images includes: Performing behavior recognition processing based on the historical tracking images to obtain behavior interaction events between each of the specific persons and other specific persons; Based on the behavioral interaction events, an event relationship network of each of the specific persons is constructed; Based on the event relationship network, counting the relationship score between each of the specific persons and other specific persons; Based on the relationship score, establishing an association relationship between each specific person and other specific persons; Based on the association relationship, a specific relationship network of each specific person is constructed.

5. The event management method according to any one of claims 1 to 4, characterized in that: The analyzing and processing the absence of the absentee based on at least one of the relationship graph and the current tracking image to obtain the absence reason of the absentee includes: When there are multiple absent persons, determining the personnel relationships between the absent persons based on the relationship map; The absence of the absent person is analyzed and processed based on at least one of the current tracking image and the personnel relationship to obtain the absence reason of the absent person.

6. The event management method according to claim 5, characterized in that: The personnel relationship includes a hostile relationship and a non-hostile relationship, and the analyzing and processing the absence of the absent person based on at least one of the current tracking image and the personnel relationship to obtain the absence reason of the absent person includes: When the personnel relationship between the absent persons is the non-hostile relationship, a behavior recognition process is performed based on the current tracking image to obtain a behavior recognition result of the absent persons, and based on the behavior recognition result, the absence reason of the absent persons is determined; When the personnel relationship between the absent personnel is the hostile relationship, absence alarm information corresponding to the absent personnel is generated based on the hostile relationship, and the absence alarm information is fed back to the management personnel of the specific area, so that the management personnel can upload the absence reasons of the absent personnel after managing the absent personnel.

7. The event management method according to any one of claims 1 to 4, characterized in that: The managing of the roll call event based on the absence reason includes: Determine the impact of the absence reason on the specific area; Based on the influence, determining the absence points of the absent personnel in the roll call event; Based on the absence points and the relationship graph, generating an event report corresponding to the roll call event; Based on the event report, the roll call event is managed.

8. An event management device, characterized in that: The event management device comprises: A first acquisition module is used to acquire a relationship map and / or a current tracking image of an absent person when an absent person appears in a roll call event in a specific area; a first analysis module, configured to analyze and process the absence of the absentee based on at least one of the relationship graph and the current tracking image, and obtain the absence reason of the absentee; The first management module is used to manage the roll call event based on the absence reason.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the event management method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the event management method according to any one of claims 1 to 7 are implemented.

Images