Method, device, electronic device and storage medium for combining and combining cases

Through facial recognition and relationship diagram structure, the problem of regional restrictions between case handling departments is solved, efficient correlation and concurrent case information is achieved, and the efficiency of handling a series of cases is improved.

CN114677242BActive Publication Date: 2024-11-29SHENZHEN INTELLIFUSION TECHNOLOGIES CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202110809110.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-07-16
Publication Date
2024-11-29
Estimated Expiration
2041-07-16

AI Technical Summary

Technical Problem

The regional restrictions and information differences between case handling departments lead to inefficient efficiency in case series, affecting the efficiency of handling series of cases.

Method used

By obtaining the face images and case information of the target person, using face recognition and relationship diagram structure to process the case in succession, breaking regional restrictions, sharing case information, and improving correlation efficiency.

Benefits of technology

Effectively link cases, improve the efficiency of case-by-case concurrent cases, and thus improve the efficiency of handling series of cases.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114677242B_ABST
    Figure CN114677242B_ABST
Patent Text Reader

Abstract

The embodiment of the present invention provides a method for serializing and combining cases, obtaining a search request for a case, the search request including a facial image of a target person and case information; performing face recognition on the target facial image to obtain the target person information corresponding to the facial image of the target person; searching in a historical case library based on the target person information to obtain a search result corresponding to the target person information, the historical case library storing historical case information corresponding to historical cases; if the search result includes historical case information, obtaining a first relationship graph structure corresponding to the historical case information; performing a first serialization and combination processing on the case based on the first relationship graph structure. By obtaining a search request including the facial image of the target person and case information, and using the relationship graph structure to serialize and combine the cases, the cases can be effectively associated, the efficiency of serialization and combination can be improved, and the processing efficiency of a series of cases can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data management, and in particular to a method, device, electronic device and storage medium for serializing and combining cases. Background Art

[0002] The processing of cases in series and in parallel is an important measure taken by the relevant departments in case handling. It mainly refers to connecting a series of cases corresponding to a person or a team, organizing special cases together, and conducting case handling from one case to another, and from one case to multiple cases. It can maximize the concentration of case handling information and narrow the scope of case handling, so as to achieve the purpose of quickly handling cases and saving case handling resources. However, in the process of organization, it is necessary to screen out related cases and find related clues in related cases before combining cases in series and in parallel. Due to the regionality between the relevant departments in case handling and the information gap between the case handlers, it is impossible to effectively associate the key information of the case, which makes the efficiency of combining cases in series and in parallel low, thus affecting the handling of series cases. Summary of the invention

[0003] An embodiment of the present invention provides a method for combining and linking cases. By obtaining a search request including a facial image of a target person and case information, historical case information corresponding to the target person is retrieved, and a relationship graph structure is used to combine and link cases. This method can break the geographical restrictions between case handling departments, share information between case handling personnel, effectively link cases, improve the efficiency of combining and linking cases, and further improve the handling efficiency of a series of cases.

[0004] In a first aspect, an embodiment of the present invention provides a method for combining and concatenating cases, the method comprising:

[0005] Obtaining a search request for a case, the search request including a face image of a target person and case information;

[0006] Performing face recognition on the target face image to obtain target person information corresponding to the target person's face image;

[0007] According to the target person information, searching is performed in a historical case database to obtain a search result corresponding to the target person information, wherein the historical case database stores historical case information corresponding to the historical case;

[0008] If the search result includes historical case information, determining whether the case information exists in the historical case information;

[0009] If the case information exists in the historical case information, obtaining a first relationship graph structure corresponding to the historical case information;

[0010] According to the first relationship graph structure, the cases are subjected to a first serial and combined case processing.

[0011] Optionally, the first relationship graph structure includes event nodes, personnel nodes, and relationship edges, and the nodes are connected by relationship edges. The first serial and combined case processing of the cases according to the first relationship graph structure includes:

[0012] Obtaining a target level of a person node corresponding to the target person in the first relationship graph structure; and

[0013] Obtaining a target relationship edge of the target person in the first relationship graph structure;

[0014] Calculating the participation degree of the target personnel according to the target level and the target relationship edge;

[0015] The first series of cases will be processed based on the participation level of the target persons.

[0016] Optionally, the first case consolidation process according to the participation degree of the target person includes:

[0017] Processing the case together with one or more historical cases in which the target person's participation degree is greater than a first preset participation degree; and / or

[0018] Linking the case with one or more historical cases whose participation degree of the target person is greater than a second preset participation degree;

[0019] Among them, the first preset participation degree is greater than the second preset participation degree.

[0020] Optionally, the method further includes:

[0021] If the search result does not include historical case information, the case information is made public;

[0022] If there are identical search requests during the publicity period, the cases will be processed as a second series of cases;

[0023] After the public disclosure period, the case information will be organized into a historical case database.

[0024] Optionally, the case information includes key case information, and after the public disclosure period, the method further includes:

[0025] According to the quantity and quality of the key case information, determining whether the case information meets the construction conditions of the second relationship graph structure;

[0026] If the case information satisfies the construction condition of the second relationship graph structure, constructing the corresponding second relationship graph structure according to the case key information;

[0027] If the case information does not meet the construction condition of the second relationship graph structure, the case information is matched for case similarity in the historical case database to obtain a matching result, wherein the matching result includes historical case information whose case similarity is greater than a preset threshold;

[0028] The key case information is sorted and added to the first relationship graph structure corresponding to the historical case information whose case similarity is greater than a preset threshold.

[0029] Optionally, the relationship edge includes a correlation relationship edge and a personnel relationship edge, the event node and the personnel node are connected via the correlation relationship edge, and each of the personnel nodes is connected via the personnel relationship edge, and a corresponding relationship graph structure is constructed according to the key case information, including:

[0030] Constructing an event node according to the case information of the case;

[0031] Construct a personnel node based on the target personnel information of the case;

[0032] Construct relevant relationship edges based on the key case information between the case and the target person;

[0033] Constructing personnel relationship edges according to the social relationships between the target personnel;

[0034] Based on the event node, the construction personnel node, the related relationship edge and the personnel relationship edge, a second relationship graph structure corresponding to the case is constructed.

[0035] Optionally, the historical cases include unfinished cases, and the method further includes:

[0036] Calculate the correlation between the second relationship graph structure corresponding to the case and the first relationship graph structure corresponding to the unfinished case to obtain the correlation between the case and the unfinished case;

[0037] If the correlation between the case and the unfinished case is greater than a preset correlation, the case and the unfinished case will be processed as a third series.

[0038] In a second aspect, an embodiment of the present invention provides a device for serializing and combining cases, the device comprising:

[0039] A first acquisition module is used to acquire a search request for a case, wherein the search request includes a face image of a target person and case information;

[0040] A recognition module, used to perform face recognition on the target face image to obtain target person information corresponding to the target person's face image;

[0041] A retrieval module, used to search in a historical case library according to the target person information to obtain a retrieval result corresponding to the target person information, wherein the historical case library stores historical case information corresponding to the historical case;

[0042] A judgment module, configured to judge whether the case information exists in the historical case information if the search result includes the historical case information;

[0043] A second acquisition module, configured to acquire a first relationship graph structure corresponding to the historical case information if the case information exists in the historical case information;

[0044] A processing module is used to perform a first serial and case-combining processing on the cases according to the first relationship graph structure.

[0045] 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 in the method for serializing and paralleling cases provided in an embodiment of the present invention when executing the computer program.

[0046] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the method for stringing and combining cases provided in the embodiment of the invention are implemented.

[0047] In an embodiment of the present invention, a search request for a case is obtained, the search request includes a facial image of a target person and case information; face recognition is performed on the target facial image to obtain the target person information corresponding to the facial image of the target person; based on the target person information, a search is performed in a historical case library to obtain a search result corresponding to the target person information, and the historical case library stores historical case information corresponding to historical cases; if the search result includes historical case information, a first relationship graph structure corresponding to the historical case information is obtained; based on the first relationship graph structure, a first serial case processing is performed on the case. The present invention obtains a search request including a facial image of a target person and case information, retrieves historical case information corresponding to the target person, and uses a relationship graph structure to perform serial case processing on the case, which can break the geographical restrictions between relevant departments in case handling, share information between case handlers, effectively associate cases, improve the efficiency of serial case processing, and thus improve the efficiency of handling a series of cases. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] 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.

[0049] Figure 1 is a flow chart of a method for combining and concatenating cases provided by an embodiment of the present invention;

[0050] Figure 2 is a flow chart of a first serial and combined case processing method provided by an embodiment of the present invention;

[0051] Figure 3 It is a structural schematic diagram of a case serial and parallel device provided by an embodiment of the present invention;

[0052] Figure 4 is a structural schematic diagram of a processing module provided by an embodiment of the present invention;

[0053] Figure 5 is a structural schematic diagram of a processing unit provided by an embodiment of the present invention;

[0054] Figure 6 It is a structural schematic diagram of another case serial and parallel device provided by an embodiment of the present invention;

[0055] Figure 7 It is a structural schematic diagram of another case serial and parallel device provided by an embodiment of the present invention;

[0056] Figure 8 It is a structural schematic diagram of another case serial and parallel device provided by an embodiment of the present invention;

[0057] Fig. 9 It is a structural schematic diagram of another case serial and parallel device provided by an embodiment of the present invention;

[0058] Fig.10 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] 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.

[0060] See also Figure 1 , Figure 1 is a flow chart of a method for combining and combining cases provided by an embodiment of the present invention. Figure 1 As shown, the case linking method includes the following steps:

[0061] 101. Obtain a search request for a case.

[0062] In the embodiment of the present invention, the search request includes the face image of the target person and the case information. The case refers to the case to be handled by the relevant department or case handler. The search request is initiated by the case handler through the user terminal. The case information includes the case name (the case name may also be a case code, case number, etc.) and the case description.

[0063] When the case handler initiates a search request through the user terminal, he can upload the face image of the target person and the case description. The face image of the target person is used for face recognition to identify the relevant information of the target person. The case description is mainly composed of key case information, which can be case clues.

[0064] 102. Perform face recognition on the face image of the target person to obtain target person information corresponding to the face image of the target person.

[0065] In an embodiment of the present invention, facial features are extracted from a target facial image using facial image technology to obtain facial features of a target person, and then the facial features are used to search for corresponding personnel information in a personnel information database, thereby determining the personnel information corresponding to the target person as the target personnel information, wherein the above-mentioned target person information may include the identity information, social activity information and retrieved information of the target person.

[0066] 103. According to the target person information, search is performed in the historical case database to obtain search results corresponding to the target person information.

[0067] In an embodiment of the present invention, the historical case database stores historical case information corresponding to the historical case. There is an index relationship between the historical case information in the historical case database and the personnel information in the personnel information database. The specific index relationship may be that one historical case information corresponds to one or more personnel information, and one personnel information may correspond to a null value or one or more historical case information.

[0068] According to the index relationship between the target person information and the historical case information, the corresponding historical case information can be indexed in the historical case library as a retrieval result.

[0069] There may be two types of the above search results, one is a result including historical case information, and the other is a result not including historical case information. The above result including historical case information can indicate that the target person has a relationship with the historical case, and the above result not including historical case information can be a null value, and the result not including historical case information can indicate that the target person has no historical case.

[0070] 104. If the search result includes historical case information, determine whether the case information exists in the historical case information.

[0071] In the embodiment of the present invention, if the search result includes historical case information, it means that the target person has a relationship with the retrieved historical case. Furthermore, if the case information exists in the historical case information, it means that the case is a historical case. If the case information does not exist in the historical case information, it means that the case is a new case.

[0072] When the search results include historical cases, a specific analysis can be conducted on the target person and the historical cases corresponding to the target person to determine whether there is a possibility of serial cases.

[0073] 105. If the case information exists in the historical case information, obtain the first relationship graph structure corresponding to the historical case information.

[0074] In an embodiment of the present invention, if the case information exists in the historical case information, it means that the case is a historical case. The above-mentioned first relationship graph structure can be stored in a relationship graph library, and the first relationship graph structure in the above-mentioned relationship graph library has an index relationship with the historical case information in the historical case library, and the specific index relationship is that one historical case information corresponds to one first relationship graph structure. The above-mentioned first relationship graph structure can include event nodes, personnel nodes and relationship edges, and each node is connected by a relationship edge.

[0075] In the above-mentioned first relationship graph structure, the above-mentioned event node can record the structured information corresponding to the case, and the structured information of the case can include the case name, case description, etc. The above-mentioned case description can be the case time, case location, case process, case result, etc. The personnel node can record the corresponding personnel structured information, and the above-mentioned personnel structured information can include the identity information of the personnel, case relationship information and retrieved information. The above-mentioned relationship edge represents the relationship between each node, and the above-mentioned relationship edge can be a weighted relationship edge, and the weight of the relationship edge can represent the degree of correlation between the two nodes. The larger the weight of the relationship edge between the personnel node and the event node, the more relevant the personnel node is to the event node, and further indicates that the personnel is more relevant to the case.

[0076] 106. Based on the first relationship diagram structure, the cases are processed in the first series.

[0077] In an embodiment of the present invention, when a target person corresponds to multiple historical case information, the target person also corresponds to multiple first relationship graph structures, and whether the multiple historical case information needs to be serialized and merged can be determined based on the correlation between the multiple first relationship graph structures.

[0078] In a possible embodiment, the first relationship graph structure includes multiple levels, the central level is the event node, the first level is the personnel node directly connected to the event node through one relationship edge, the second level is the personnel node whose shortest path to the event node is two relationship edges, and the remaining levels can be deduced accordingly.

[0079] The cases can be linked together according to the level of the target person in the first relationship graph structure. For example, if the level of the target person in five first relationship graph structures is the first level, the historical cases corresponding to the five first relationship graph structures can be linked together.

[0080] In a possible embodiment, the target person information, the search results and the results of the combined cases may be returned. The search results may also include the historical searcher, the unit where the historical searcher is located and the historical search time.

[0081] In an embodiment of the present invention, a search request for a case is obtained, the search request includes a facial image of a target person and case information; face recognition is performed on the target facial image to obtain the target person information corresponding to the facial image of the target person; based on the target person information, a search is performed in a historical case library to obtain a search result corresponding to the target person information, and the historical case library stores historical case information corresponding to historical cases; if the search result includes historical case information, a first relationship graph structure corresponding to the historical case information is obtained; based on the first relationship graph structure, a first serial case processing is performed on the case. The present invention obtains a search request including a facial image of a target person and case information, retrieves historical case information corresponding to the target person, and uses a relationship graph structure to perform serial case processing on the case, which can break the geographical restrictions between relevant departments in case handling, share information between case handlers, effectively associate cases, improve the efficiency of serial case processing, and thus improve the efficiency of handling a series of cases.

[0082] Optional, see Figure 2 , Figure 2 is a flow chart of a first serial and parallel case processing method provided by an embodiment of the present invention, such as Figure 2 As shown, the specific steps include:

[0083] 201. Obtain a target level of a personnel node corresponding to a target personnel in a first relationship graph structure.

[0084] In an embodiment of the present invention, in a first relationship graph structure, a target person corresponds to a person node, and the level of a person node in the first relationship graph structure can indicate the degree of relevance between the person corresponding to the person node and the case. The closer the level of the person node is to the central level where the event node is located, the higher the degree of relevance between the person corresponding to the person node and the case. It should be noted that the person node corresponding to the above-mentioned target person can also be referred to as a target person node.

[0085] 202. Obtain a target relationship edge of the target person in the first relationship graph structure.

[0086] In an embodiment of the present invention, in a first relationship graph structure, a person node may correspond to one or more relationship edges, further, a person node may correspond to at most one event node, and a person node may correspond to one or more person nodes. The target relationship edge refers to a relationship edge directly connected to the person node corresponding to the target person.

[0087] Furthermore, in the embodiment of the present invention, the above-mentioned relationship edge can be a weighted relationship edge. Different weight ranges can be set according to the connection relationship between different nodes. For example, the weight range of the relationship edge between personnel nodes can be set to 0-0.5, and the weight range of the relationship edge between personnel nodes and event nodes can be set to 0.5-1.0.

[0088] 203. Calculate the participation of the target personnel according to the target level and the target relationship edge.

[0089] In an embodiment of the present invention, the above-mentioned degree of participation can be understood as the degree to which the target person participates in the corresponding case. The closer the personnel node corresponding to the target person is to the event node of the case, the deeper the degree to which the target person participates in the corresponding case. The more connections the personnel node corresponding to the target person has with the remaining personnel nodes of the case, the deeper the degree to which the target person participates in the corresponding case.

[0090] Specifically, the participation degree of the above target personnel can be determined by the following formula:

[0091]

[0092]

[0093]

[0094]

[0095] Among them, the above C j represents the participation of the target person in the jth case. jrepresents the target level of the target person node in the first relationship graph structure corresponding to the jth case. j represents the relationship edge coefficient of the target person node in the first relationship graph structure corresponding to the jth case. j represents the number of target relationship edges between target person nodes and person nodes in the first relationship graph structure corresponding to the jth case, a ij represents the weight value of the relationship edge between the target person node and the i-th person node in the first relationship graph structure corresponding to the j-th case. jj represents the weight value of the relationship edge between the target person node and the event node in the first relationship graph structure corresponding to the jth case (if there is no relationship edge between the target person node and the event node, then d jj is 0), the above q ij represents the social relationship value between the target person node and the i-th person node in the first relationship graph structure corresponding to the j-th case. The social relationship value q ij The value of is greater than 0 and less than 1. The social relationship value is determined according to the closeness of the social relationship between two people. For example, the social relationship value of direct relatives can be 0.9, the social relationship value of collateral relatives can be 0.7, the social relationship value of colleagues can be 0.5, and the social relationship value of strangers can be 0.1. i represents the number of times the target person node and the i-th person node go together in the first relationship graph structure corresponding to the j-th case. The above l is a constant and can be set based on experience. The above t j represents the time the target person spends at the jth case scene. The above m j Represents the number of times the target person appears at the jth case scene.

[0096] According to the above formula, the degree of participation of the target person in each case can be calculated. Since the connection between the target person and the case itself and the connection between the target person and other people in the case are taken into account, the degree of participation of the target person in the case can be more accurately reflected.

[0097] 204. Carry out the first batch of combined cases based on the target personnel’s level of participation.

[0098] In the embodiment of the present invention, it can be determined whether to carry out case-linking or case-joint processing based on different participation levels.

[0099] Optionally, the above case may be one of the historical cases corresponding to the target person, or the above case may be a new case corresponding to the target person.

[0100] Furthermore, the above-mentioned case can be processed together with one or more historical cases in which the above-mentioned target person's participation degree is greater than the first preset participation degree; and / or the above-mentioned case can be processed in series with one or more historical cases in which the above-mentioned target person's participation degree is greater than the second preset participation degree; wherein the above-mentioned first preset participation degree is greater than the above-mentioned second preset participation degree. When the participation degree of the target person is greater than the first preset participation degree, it means that the target person is highly involved in the corresponding historical case, and the above-mentioned case has a very high similarity with the historical case, and the case can be processed in series. When the participation degree of the target person is greater than the second preset participation degree, it means that the target person has participated in the corresponding historical case, and the above-mentioned case has some similar clues with the historical case, and the case can be processed in series. When the participation degree of the target person is less than the second preset participation degree, it means that the target person's participation degree is low, and the connection between the above-mentioned case and the corresponding historical case has no clue value, and it is not necessary to process the case in series.

[0101] Optionally, in step 104 of the embodiment of the present invention, the search result may be a result that does not include historical case information. Further, if the above search result does not include historical case information, the above case information will be publicized; if there is an identical search request during the publicity period, the above case will be processed as a second serial case; after the publicity period, the above case information will be organized into a historical case library. If the search result does not include historical case information, it means that the target person is a new person, and the target person and the corresponding case can be publicized. Specifically, the above publicity can be to send the corresponding target person information and case information to the user terminal of other case handling personnel to cause a warning.

[0102] Furthermore, the above case information includes key case information. Generally speaking, if the case corresponding to the target person is a new case, there will be a large number of clues and high-quality clues; if the case corresponding to the target person is a historical case, this is mainly because when the historical case was created, there was no information about the target person in the clues, resulting in no information about the target person in the historical case information. It should be noted that there will be more known clues and high-quality clues in historical cases, while the target person will have fewer clues and low-quality clues. Therefore, the case corresponding to the target person can be judged and processed by the number of clues and the quality of the clues.

[0103] Optionally, in an embodiment of the present invention, it can be determined whether the case information satisfies the construction conditions of the second relationship graph structure based on the quantity of the case key information and the quality of the case key information; if the case information satisfies the construction conditions of the second relationship graph structure, a corresponding second relationship graph structure is constructed based on the case key information; if the case information does not meet the construction conditions of the second relationship graph structure, the case information is matched for case similarity in the historical case library to obtain a matching result, which includes historical case information whose case similarity is greater than a preset threshold; the case key information is sorted and added to the first relationship graph structure corresponding to the historical case information whose case similarity is greater than the preset threshold.

[0104] The above-mentioned key information of the case may include the key information provider, the persons mentioned in the key information, the time when the key information was provided, the location where the key information was provided, etc. The quality of the above-mentioned key information of the case can be determined based on the key information provider, the persons mentioned in the key information, the time when the key information was provided, the location where the key information was provided, etc. Specifically, it can be determined based on the expert score of the key information provider, the number of persons mentioned in the key information, the interval between the time when the key information was provided and the time when the case occurred, and the distance between the location where the key information was provided and the location where the case occurred. The higher the expert score of the key information provider, the higher the quality of the key information of the case; the more the number of persons mentioned in the key information, the higher the quality of the key information of the case; the shorter the interval between the time when the key information was provided and the time when the case occurred, the higher the quality of the key information of the case; the closer the distance between the location where the key information was provided and the location where the case occurred, the higher the quality of the key information of the case.

[0105] For cases with a large amount of key case information and high quality of key case information in the case information, it can be considered that it is a new case corresponding to a new target person, and a second relationship graph structure can be constructed for the case. It should be noted that the above-mentioned second relationship graph structure is only to distinguish it from the first relationship graph structure corresponding to the historical case, and there is no other difference except the content of the relationship graph structure. Furthermore, the constructed second relationship graph structure can be stored as the first relationship graph structure in the relationship graph library, and the case information of the case is stored as historical case information in the historical case library, and an index of the case information of the case and the corresponding relationship graph structure is established.

[0106] Optionally, the above-mentioned relationship edges include correlation relationship edges and personnel relationship edges, the above-mentioned event nodes are connected to the above-mentioned personnel nodes via the above-mentioned correlation relationship edges, and the above-mentioned personnel nodes are connected via the above-mentioned personnel relationship edges. Specifically, the event node can be constructed based on the case information of the above-mentioned case; the personnel node can be constructed based on the target personnel information of the above-mentioned case; the correlation relationship edge can be constructed based on the case key information between the above-mentioned case and the target personnel; the personnel relationship edge can be constructed based on the social relationship between the above-mentioned target personnel; based on the above-mentioned event nodes, the above-mentioned constructed personnel nodes, the above-mentioned correlation relationship edges and the above-mentioned personnel relationship edges, a second relationship graph structure corresponding to the above-mentioned case is constructed. It should be noted that the above-mentioned correlation relationship edges have corresponding weight values, and the above-mentioned personnel relationship edges also have corresponding weight values. The weight value of the above-mentioned correlation relationship edges can be the weight value d of the relationship edge between the target personnel node and the event node. jj , the weight value of the above personnel relationship edge can be the social relationship value q between the target personnel node and the i-th personnel node ij .

[0107] Optionally, in an embodiment of the present invention, the historical cases stored in the historical case library include completed cases and uncompleted cases, and the first relationship graph structure stored in the relationship graph library may also include the first relationship graph structure corresponding to the completed cases and the first relationship graph structure corresponding to the uncompleted cases. Further, the second relationship graph structure corresponding to the case and the first relationship graph structure corresponding to the uncompleted case may be correlated to obtain the correlation between the case and the uncompleted case; if the correlation between the case and the uncompleted case is greater than a preset correlation, the case and the uncompleted case are processed as a third series.

[0108] Furthermore, the above correlation calculation is mainly performed using node similarity. Specifically, the intersection and union ratio of the first node set A in the first relationship graph structure and the second node set B in the second relationship graph structure can be used as the correlation S IOU , as shown in the following formula:

[0109]

[0110] In the embodiment of the present invention, the greater the correlation, the more similar the two cases are, and the greater the possibility of combining the cases. Specifically, the third combination of cases can be: combining one or more unfinished cases whose correlation with the above case is greater than a first preset correlation; and / or combining one or more unfinished cases whose correlation with the above case is greater than a second preset correlation; wherein the first preset correlation is greater than the second preset correlation.

[0111] The embodiments of the present invention can retrieve historical case information corresponding to the target person by obtaining a search request including a facial image of the target person and case information, and use a relationship graph structure to process the cases in series and in parallel. This can break the geographical restrictions between the relevant departments in case handling, share information between the case handlers, effectively associate cases, improve the efficiency of case connection and in parallel, and thus improve the handling efficiency of a series of cases.

[0112] It should be noted that the case stringing and merging method provided in the embodiment of the present invention can be applied to smart phones, computers, servers and other devices that can perform data queries.

[0113] Optional, see Figure 3 , Figure 3 is a schematic diagram of the structure of a case serial and parallel device provided by an embodiment of the present invention, such as Figure 3 As shown, the device comprises:

[0114] A first acquisition module 301 is used to acquire a search request for a case, wherein the search request includes a face image of a target person and case information;

[0115] The recognition module 302 is used to perform face recognition on the target face image to obtain target person information corresponding to the target person's face image;

[0116] A search module 303 is used to search in a historical case database according to the target person information to obtain a search result corresponding to the target person information, wherein the historical case database stores historical case information corresponding to the historical case;

[0117] A first judgment module 304, configured to judge whether the case information exists in the historical case information if the search result includes the historical case information;

[0118] A second acquisition module 305, configured to acquire a first relationship graph structure corresponding to the historical case information if the case information exists in the historical case information;

[0119] The first processing module 306 is used to perform a first serial and case processing on the cases according to the first relationship graph structure.

[0120] Optional, such as Figure 4 As shown, the processing module 306 includes:

[0121] A first acquisition unit 3061 is used to acquire a target level of a person node corresponding to the target person in the first relationship graph structure; and

[0122] A second acquisition unit 3062 is used to acquire a target relationship edge of the target person in the first relationship graph structure;

[0123] A calculation unit 3063, configured to calculate the participation degree of the target person according to the target level and the target relationship edge;

[0124] The processing unit 3064 is used to perform a first case consolidation process according to the participation degree of the target person.

[0125] Optional, such as Figure 5 As shown, the processing unit 3064 includes:

[0126] The first processing subunit 30641 is used to combine the case with one or more historical cases in which the participation degree of the target person is greater than a first preset participation degree; and / or

[0127] The second processing subunit 30642 is used to link the case with one or more historical cases whose participation degree of the target person is greater than a second preset participation degree;

[0128] Among them, the first preset participation degree is greater than the second preset participation degree.

[0129] Optional, such as Figure 6 As shown, the device also includes:

[0130] A publicizing module 307, for publicizing the case information if the search result does not include the historical case information;

[0131] The second processing module 308 is used to perform a second case concatenation process on the case if there is an identical search request during the public announcement period;

[0132] The sorting module 309 is used to sort the case information into the historical case database after the public announcement period.

[0133] Optional, such as Figure 7 As shown, the device also includes:

[0134] A second judgment module 310, configured to judge whether the case information satisfies the construction condition of the second relationship graph structure according to the quantity and quality of the case key information;

[0135] A first construction module 311 is used to construct a corresponding second relationship graph structure according to the case key information if the case information meets the construction condition of the second relationship graph structure;

[0136] A matching module 312 is used for, if the case information does not meet the construction condition of the second relationship graph structure, performing case similarity matching on the case information in the historical case database to obtain a matching result, wherein the matching result includes historical case information whose case similarity is greater than a preset threshold;

[0137] The adding module 313 is used to organize and add the case key information to the first relationship graph structure corresponding to the historical case information whose case similarity is greater than a preset threshold.

[0138] Optional, such as Figure 8 As shown, the device also includes:

[0139] A second construction module 314 is used to construct an event node according to the case information of the case;

[0140] The third construction module 315 is used to construct a personnel node according to the target personnel information of the case;

[0141] The fourth construction module 316 is used to construct a relevant relationship edge according to the case key information between the case and the target person;

[0142] A fifth construction module 317 is used to construct a personnel relationship edge according to the social relationship between each of the target personnel;

[0143] The sixth construction module 318 is used to construct a second relationship graph structure corresponding to the case based on the event node, the construction personnel node, the related relationship edge and the personnel relationship edge.

[0144] Optional, such as Fig. 9 As shown, the device also includes:

[0145] A calculation module 319, configured to calculate the correlation between the second relationship graph structure corresponding to the case and the first relationship graph structure corresponding to the unfinished case, to obtain the correlation between the case and the unfinished case;

[0146] The third processing module 320 is used to perform a third serial and combined case processing on the case and the unfinished case if the correlation between the case and the unfinished case is greater than a preset correlation.

[0147] It should be noted that the case serial and parallel device provided in the embodiment of the present invention can be applied to smart phones, computers, servers and other devices that can perform business analysis at the map level.

[0148] The case serial and parallel device provided in the embodiment of the present invention can realize each process of the case serial and parallel method in the above method embodiment, and can achieve the same beneficial effect. To avoid repetition, it will not be described here.

[0149] See also Fig.10 , Fig.10 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Fig.10 As shown, it includes: a memory 1002, a processor 1001, and a computer program of a method for serializing and combining cases stored in the memory 1002 and executable on the processor 1001, wherein:

[0150] The processor 1001 is used to call the computer program stored in the memory 1002 and execute the following steps:

[0151] Obtaining a search request for a case, the search request including a face image of a target person and case information;

[0152] Performing face recognition on the target face image to obtain target person information corresponding to the target person's face image;

[0153] According to the target person information, searching is performed in a historical case database to obtain a search result corresponding to the target person information, wherein the historical case database stores historical case information corresponding to the historical case;

[0154] If the search result includes historical case information, determining whether the case information exists in the historical case information;

[0155] If the case information exists in the historical case information, obtaining a first relationship graph structure corresponding to the historical case information;

[0156] According to the first relationship graph structure, the cases are subjected to a first serial and combined case processing.

[0157] Optionally, the first relationship graph structure includes event nodes, personnel nodes, and relationship edges, and the nodes are connected by relationship edges. The processor 1001 performs a first case concatenation and concatenation processing on the cases according to the first relationship graph structure, including:

[0158] Obtaining a target level of a person node corresponding to the target person in the first relationship graph structure; and

[0159] Obtaining a target relationship edge of the target person in the first relationship graph structure;

[0160] Calculating the participation degree of the target personnel according to the target level and the target relationship edge;

[0161] The first series of cases will be processed based on the participation level of the target persons.

[0162] Optionally, the processor 1001 performs a first case concatenation process according to the participation degree of the target person, including:

[0163] Processing the case together with one or more historical cases in which the target person's participation degree is greater than a first preset participation degree; and / or

[0164] Linking the case with one or more historical cases whose participation degree of the target person is greater than a second preset participation degree;

[0165] Among them, the first preset participation degree is greater than the second preset participation degree.

[0166] Optionally, the processor 1001 executes the method further comprising:

[0167] If the search result does not include historical case information, the case information is made public;

[0168] If there are identical search requests during the publicity period, the cases will be processed as a second series of cases;

[0169] After the public disclosure period, the case information will be organized into a historical case database.

[0170] Optionally, the case information includes key case information, and after the public disclosure period, the processor 1001 executes the method further comprising:

[0171] According to the quantity and quality of the key case information, determining whether the case information meets the construction conditions of the second relationship graph structure;

[0172] If the case information satisfies the construction condition of the second relationship graph structure, constructing the corresponding second relationship graph structure according to the case key information;

[0173] If the case information does not meet the construction condition of the second relationship graph structure, the case information is matched for case similarity in the historical case database to obtain a matching result, wherein the matching result includes historical case information whose case similarity is greater than a preset threshold;

[0174] The key case information is sorted and added to the first relationship graph structure corresponding to the historical case information whose case similarity is greater than a preset threshold.

[0175] Optionally, the relationship edge includes a correlation relationship edge and a personnel relationship edge, the event node and the personnel node are connected through the correlation relationship edge, and each of the personnel nodes is connected through the personnel relationship edge, and the processor 1001 executes to construct a corresponding relationship graph structure according to the case key information, including:

[0176] Constructing an event node according to the case information of the case;

[0177] Construct a personnel node based on the target personnel information of the case;

[0178] Construct relevant relationship edges based on the key case information between the case and the target person;

[0179] Constructing personnel relationship edges according to the social relationships between the target personnel;

[0180] Based on the event node, the construction personnel node, the related relationship edge and the personnel relationship edge, a second relationship graph structure corresponding to the case is constructed.

[0181] Optionally, the historical cases include unfinished cases, and the processor 1001 executes the method further comprising:

[0182] Calculate the correlation between the second relationship graph structure corresponding to the case and the first relationship graph structure corresponding to the unfinished case to obtain the correlation between the case and the unfinished case;

[0183] If the correlation between the case and the unfinished case is greater than a preset correlation, the case and the unfinished case will be processed as a third series.

[0184] It should be noted that the electronic device provided by the embodiment of the present invention can be applied to smart phones, computers, servers and other devices that can serialize and parallel cases.

[0185] The electronic device provided by the embodiment of the present invention can implement each process of the method for serializing and paralleling cases in the above method embodiment, and can achieve the same beneficial effect. To avoid repetition, it will not be described here.

[0186] An embodiment of the present invention further 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 case stringing and paralleling method or the application-side case stringing and paralleling method provided in the embodiment of the present invention, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0187] 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 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. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0188] 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. A method for combining and combining cases, characterized in that: The following steps are involved: Obtaining a search request for a case, the search request including a face image of a target person and case information; Performing face recognition on the face image of the target person to obtain target person information corresponding to the face image of the target person; According to the target person information, searching is performed in a historical case database to obtain a search result corresponding to the target person information, wherein the historical case database stores historical case information corresponding to the historical case; If the search result includes historical case information, determining whether the case information exists in the historical case information; If the case information exists in the historical case information, a first relationship graph structure corresponding to the historical case information is obtained; the first relationship graph structure includes event nodes, personnel nodes, and relationship edges, and each node is connected by a relationship edge; Obtaining a target level of a personnel node corresponding to the target personnel in the first relationship graph structure; as well as Obtaining a target relationship edge of the target person in the first relationship graph structure; The participation degree of the target person is calculated according to the target level and the target relationship edge; the participation degree of the target person is determined by the following formula: Among them, the C j represents the target person's participation in the jth case, and the k j represents the target level of the target person node in the first relationship graph structure corresponding to the jth case, and the b j represents the relationship edge coefficient of the target person node in the first relationship graph structure corresponding to the jth case, the N j represents the number of target relationship edges between target person nodes and person nodes in the first relationship graph structure corresponding to the jth case, a ij represents the weight value of the relationship edge between the target person node and the i-th person node in the first relationship graph structure corresponding to the j-th case, wherein d j represents the weight value of the relationship edge between the target person node and the event node in the first relationship graph structure corresponding to the jth case. If there is no relationship edge between the target person node and the event node, then d j is 0, the q ij represents the social relationship value between the target person node and the i-th person node in the first relationship graph structure corresponding to the j-th case. The social relationship value q ij The value of is greater than 0 and less than 1, and the social relationship value is determined according to the intimacy of the social relationship between two persons; ij represents the number of times the target person node and the i-th person node go together in the first relationship graph structure corresponding to the j-th case, where l is a constant, which is set based on experience, and t j represents the time the target person spends at the jth case scene, and the m j represents the number of times the target person appears at the jth case scene; Processing the case together with one or more historical cases in which the target person's participation degree is greater than a first preset participation degree; and / or Linking the case with one or more historical cases whose participation degree of the target person is greater than a second preset participation degree; Among them, the first preset participation degree is greater than the second preset participation degree.

2. The method according to claim 1, characterized in that The method further comprises: If the search result does not include historical case information, the case information is made public; If there are identical search requests during the publicity period, the cases will be processed as a second series of cases; After the public disclosure period, the case information will be organized into a historical case database.

3. The method according to claim 2, characterized in that The case information includes key case information. After the public disclosure period, the method further includes: According to the quantity and quality of the key case information, determining whether the case information meets the construction conditions of the second relationship graph structure; If the case information satisfies the construction condition of the second relationship graph structure, constructing the corresponding second relationship graph structure according to the case key information; If the case information does not meet the construction condition of the second relationship graph structure, the case information is matched for case similarity in the historical case database to obtain a matching result, wherein the matching result includes historical case information whose case similarity is greater than a preset threshold; The key case information is sorted and added to the first relationship graph structure corresponding to the historical case information whose case similarity is greater than a preset threshold.

4. The method according to claim 3, characterized in that The relationship edge includes a related relationship edge and a personnel relationship edge. The event node and the personnel node are connected through the related relationship edge, and each of the personnel nodes is connected through the personnel relationship edge. A corresponding relationship graph structure is constructed according to the key case information, including: Constructing an event node according to the case information of the case; Construct a personnel node based on the target personnel information of the case; Construct relevant relationship edges based on the key case information between the case and the target person; Constructing personnel relationship edges according to the social relationships between the target personnel; Based on the event node, the construction personnel node, the related relationship edge and the personnel relationship edge, a second relationship graph structure corresponding to the case is constructed.

5. The method according to claim 4, characterized in that The historical cases include unfinished cases, and the method further includes: Calculate the correlation between the second relationship graph structure corresponding to the case and the first relationship graph structure corresponding to the unfinished case to obtain the correlation between the case and the unfinished case; If the correlation between the case and the unfinished case is greater than a preset correlation, the case and the unfinished case will be processed as a third series.

6. A device for combining and combining cases, characterized in that: The device comprises: A first acquisition module is used to acquire a search request for a case, wherein the search request includes a face image of a target person and case information; A recognition module, used to perform face recognition on the face image of the target person to obtain target person information corresponding to the face image of the target person; A retrieval module, used to search in a historical case library according to the target person information to obtain a retrieval result corresponding to the target person information, wherein the historical case library stores historical case information corresponding to the historical case; A judgment module, configured to judge whether the case information exists in the historical case information if the search result includes the historical case information; A second acquisition module is used to acquire a first relationship graph structure corresponding to the historical case information if the case information exists in the historical case information; the first relationship graph structure includes event nodes, personnel nodes and relationship edges, and each node is connected by a relationship edge; A processing module is used to obtain the target level of the personnel node corresponding to the target personnel in the first relationship graph structure; and obtain the target relationship edge of the target personnel in the first relationship graph structure; according to the target level and the target relationship edge, calculate the participation degree of the target personnel; the participation degree of the target personnel is determined by the following formula: Among them, the C j represents the target person's participation in the jth case, and the k j represents the target level of the target person node in the first relationship graph structure corresponding to the jth case, and the b j represents the relationship edge coefficient of the target person node in the first relationship graph structure corresponding to the jth case, the N j represents the number of target relationship edges between target person nodes and person nodes in the first relationship graph structure corresponding to the jth case, a ij represents the weight value of the relationship edge between the target person node and the i-th person node in the first relationship graph structure corresponding to the j-th case, wherein d j represents the weight value of the relationship edge between the target person node and the event node in the first relationship graph structure corresponding to the jth case. If there is no relationship edge between the target person node and the event node, then d j is 0, the q ij represents the social relationship value between the target person node and the i-th person node in the first relationship graph structure corresponding to the j-th case. The social relationship value q ij The value of is greater than 0 and less than 1, and the social relationship value is determined according to the intimacy of the social relationship between two persons; ij represents the number of times the target person node and the i-th person node go together in the first relationship graph structure corresponding to the j-th case, where l is a constant, which is set based on experience, and t j represents the time the target person spends at the jth case scene, and the m j Indicates the number of times the target person appears at the jth case scene; combines the case with one or more historical cases in which the target person's participation degree is greater than a first preset participation degree; and / or combines the case with one or more historical cases in which the target person's participation degree is greater than a second preset participation degree; wherein the first preset participation degree is greater than the second preset participation degree.

7. 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 in the method for serializing and combining cases as described in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the method for combining and concatenating cases as described in any one of claims 1 to 5.

Citation Information

Patent Citations

  • Method and system for intelligent retrieval of series public security cases

    CN104346355A

  • Relational network construction method, device and electronic equipment

    CN107729466A

  • A case intelligent serial-parallel method based on face recognition

    CN109190498A

  • Risk information generation method and device, computer equipment and storage medium

    CN110458399A

  • Case studying and judging method and device based on deep learning

    CN110619064A