A method and system for identifying potential risk personnel based on image and data fusion

Abstract

The present invention relates to a method and system for identifying potential risk personnel based on graph-data fusion, wherein the method comprises the following steps: collecting view data and business data of risk personnel with criminal records and fusing them to form graph-data fusion data, and creating a fusion information base based on the graph-data fusion data; extracting risk personnel and related personnel in the fusion information base as nodes, constructing a relationship network diagram, and collecting related personnel information to supplement the nodes in the relationship network diagram, and creating a node attribute list and a relationship list according to the fusion information base and the related personnel information; calculating the influence index between two adjacent nodes through the relationship between each node, selecting a candidate node, and constructing an influence relationship network of the candidate node with the selected candidate node as the center, and the node inserted in the influence relationship network of the candidate node is determined by the influence index and a set threshold; confirming whether the candidate node is a potential risk person through the influence relationship network of the candidate node.

Description

Technical Field

[0001] The present invention relates to a method and system for identifying potential risk personnel based on image-data fusion, belonging to the technical field of risk personnel identification. Background Art

[0002] For early warning or identification of risk personnel, currently, we mainly obtain multi-dimensional data such as calls, financial transactions, travel records, accommodation records, and Internet cafes of risk personnel, and build analysis and early warning models based on actual business logic, so as to apply them to risk-related intelligence analysis, risk control, personnel early warning and other businesses. It is mainly based on big data analysis applications of business data.

[0003] Currently, there are many application systems and methods that use big data analysis technologies such as machine learning and data mining to analyze criminal intelligence, including risk transactions, personnel flows, capital flows, etc. For example, the Chinese invention patent with patent number "CN201910917858.7" discloses a drug control intelligence analysis system based on big data, and the patent number "CN202210726928.2" discloses a drug monitoring, early warning and evaluation method and system.

[0004] The above solution is mainly based on business data, and is not combined with the large amount of fresh view data collected by the video image big data platform. It is unable to fully grasp the latest status characteristics, life trajectory, contacts, etc. of the target object. The analyzed results are not intuitive enough and often require manual review in combination with the video image big data platform. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention proposes a method and system for identifying potential risk personnel based on image and data fusion.

[0006] The technical solution of the present invention is as follows: On the one hand, the present invention proposes a method for identifying potential risk personnel based on image-data fusion, comprising the following steps: Collect the visual data and business data of risk personnel with criminal records and fuse them into graph-number fusion data, and create a fusion information database based on the graph-number fusion data; Extract risk personnel and related personnel in the fusion information database as nodes, build a relationship network diagram, and collect related personnel information to supplement the nodes in the relationship network diagram. Create a node attribute list and relationship list based on the fusion information database and related personnel information, and define the node corresponding to the risk personnel as a risk node, the node corresponding to the related personnel as a risk neighbor node, and other nodes as ordinary nodes; Through the relationship between each node, the influence index between two adjacent nodes is calculated, and any risk neighbor node or ordinary node is selected as a candidate node. With the selected candidate node as the center, an influence relationship network of the candidate node is constructed. The nodes inserted into the influence relationship network of the candidate node are determined by the influence index and the set threshold; Confirm whether the candidate node is a potential risk person through the influence relationship network of the candidate node.

[0007] As a preferred implementation, the step of calculating the influence index between two adjacent nodes through the relationship between the nodes includes: Select two adjacent nodes, and obtain the interaction behavior data of the two nodes through the attribute lists and relationship lists of the two nodes; Different behavior weights are assigned to interaction behaviors according to behavior categories. The impact index between two adjacent nodes is calculated based on the number of different interaction behaviors and the total number of interaction behaviors between the two nodes and the frequency of risk words in the interaction behavior data of the two nodes.

[0008] As a preferred implementation, in the step of constructing a candidate node influence relationship network with the selected candidate node as the center, and determining the nodes inserted in the candidate node influence relationship network by the influence index and the set threshold: According to the set first threshold, second threshold, ..., kth threshold, from the neighbor node set, two-hop neighbor node set, ..., k-hop neighbor node set of the candidate node, select the neighbor node whose comprehensive influence index with the candidate node is greater than the corresponding threshold and insert it into the candidate node influence relationship network; Among them, in the neighbor node set of the candidate node, the comprehensive influence index between the neighbor node and the candidate node is the influence index between the two nodes; in the k-hop neighbor node set, the comprehensive influence index between the k-hop neighbor node and the candidate node is calculated by the influence index between all nodes on the relationship path from the k-hop neighbor node to the candidate node.

[0009] As a preferred implementation, the method for confirming whether a candidate node is a potential risk person through the influence relationship network of the candidate node is specifically as follows: Confirm whether the influence relationship network of each candidate node contains risk nodes, and exclude candidate nodes that do not contain risk nodes; For candidate nodes whose influence relationship network includes risk nodes, if the candidate node and the risk node are multi-hop neighbor nodes, the indirect influence index between the candidate node and the risk node is calculated through the relationship path between the candidate node and the risk node; And calculate the influence of the candidate node in the influence relationship network through the influence relationship network of the candidate node; Combined with the influence index or indirect influence index between the candidate node and the risk node, as well as the influence of the candidate node in the influence relationship network, the risk factor of the corresponding candidate node is calculated, and the risk factor and risk threshold are compared to determine whether the corresponding candidate node is a potential risk person.

[0010] As a preferred implementation, the view data includes personnel image data and vehicle image data; the business data includes identity data, call record data, financial data and registration data.

[0011] As a preferred embodiment, the method further comprises: Pre-process business data, including data cleaning and data standardization; Preprocess view data, including view data filtering, target extraction, and target matching.

[0012] As a preferred implementation, in the step of extracting risk personnel and related personnel in the fusion information database as nodes, and collecting related personnel information to supplement the nodes in the relationship network diagram: By fusing the graph and number data of risk personnel, we can identify the close contact personnel as related personnel; After determining the related persons, collect the graph fusion data of the related persons and determine the closely connected persons as supplementary nodes.

[0013] On the other hand, the present invention also proposes a potential risk personnel identification system based on image-data fusion, comprising: A graph-data fusion information base construction module is used to collect the view data and business data of risk personnel with criminal records and fuse them to form graph-data fusion data, and create a fusion information base based on the graph-data fusion data; The relationship network diagram construction module is used to extract risk personnel and related personnel in the fusion information database as nodes, construct a relationship network diagram, and collect related personnel information to supplement the nodes in the relationship network diagram. According to the fusion information database and related personnel information, the attribute list and relationship list of the nodes are created, and the nodes corresponding to risk personnel are defined as risk nodes, the nodes corresponding to related personnel are defined as risk neighbor nodes, and other nodes are defined as ordinary nodes. The influence network construction module is used to calculate the influence index between two adjacent nodes through the relationship between each node, select any risk neighbor node or ordinary node as a candidate node, and build an influence relationship network of the candidate node with the selected candidate node as the center. The node inserted in the influence relationship network of the candidate node is determined by the influence index and the set threshold; The potential risk personnel identification module is used to confirm whether the candidate node is a potential risk person through the influence relationship network of the candidate node.

[0014] On the other hand, the present invention further proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for identifying persons with potential risks based on image and data fusion as described in any embodiment of the present invention is implemented.

[0015] On the other hand, the present invention further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for identifying persons with potential risks based on image-data fusion as described in any embodiment of the present invention.

[0016] Additional aspects and advantages of the present invention will be set forth in the following description, and some of them will be apparent from the description, or may be understood by practicing the present invention. In addition, the various aspects and advantages of the present invention may be realized and obtained by the method steps and combinations particularly pointed out in the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of a method flow chart of Embodiment 1 of the present invention; Figure 2 A schematic diagram of a relationship network diagram shown in an embodiment of the present invention; Figure 3 This is a schematic diagram of an influence relationship network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0020] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0021] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0022] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0023] Embodiment 1: See also Figure 1 This embodiment provides a method for identifying potential risk personnel based on image-data fusion, which is applied to a public security big data system. The method specifically includes the following steps: S100. Collect the view data and business data of risk personnel with criminal records as the basic information database, including multi-dimensional data information such as call data, financial data, virtual identity data, video image data, registration data, etc.; view data and business data are used in different systems, so it is necessary to fuse these two types of data and apply them in one system. According to the public security big data standard specifications, access various types of business data and view data that need to be analyzed, and process, associate, analyze, compare, fuse and other operations on the data to fuse them into graph-number fusion data, and create a fusion information database based on the graph-number fusion data.

[0024] S200. Extract risk personnel and related personnel in the fusion information database as nodes, build a relationship network diagram, and collect related personnel information to supplement the nodes in the relationship network diagram. Create a node attribute list and a relationship list based on the fusion information database and related personnel information. At the same time, define the nodes corresponding to risk personnel as risk nodes, the nodes corresponding to related personnel as risk neighbor nodes, and other nodes as ordinary nodes (other nodes are nodes supplemented by collecting related personnel information).

[0025] In this embodiment, those persons who are closely connected with the risk persons are identified by analyzing the graph-number fusion data of the risk persons, and these persons are determined as associated persons; after the associated persons are successfully determined, the graph-number fusion data of these associated persons continue to be collected, and based on these data, other persons who are closely connected with the associated persons are further determined, and these persons are added to the relationship network diagram as supplementary nodes.

[0026] S300, through the relationship between each node, the influence index between any two adjacent nodes can be calculated; based on the influence index, any risk neighbor node or ordinary node can be selected as a candidate node, and an influence relationship network corresponding to the selected candidate node can be constructed with the selected candidate node as the center. In this influence relationship network, some nodes will be inserted, and the inserted nodes can be risk nodes, risk neighbor nodes or other nodes. Whether a node is inserted into the inserted influence relationship network is determined according to the influence index between nodes and the set threshold.

[0027] S400: confirm whether the candidate node is a potential risk person through the influence relationship network of the candidate node.

[0028] S500: If the candidate node is confirmed as a potential risk person, the attribute of the candidate node is upgraded from a risk neighbor node or other node to a potential risk node, and the relationship network diagram is updated. At the same time, the graph-data fusion data of the potential risk person is re-added to the fusion information database for subsequent analysis and monitoring.

[0029] The public security big data system that applies this method will continuously monitor the activities of all risk nodes and potential risk nodes. If any abnormal behavior or new related personnel are found, the re-evaluation process will be automatically triggered to ensure the accuracy and timeliness of risk personnel identification.

[0030] In one embodiment, in step S300, the step of calculating the influence index between two adjacent nodes according to the relationship between the nodes includes: S301. Select two adjacent nodes in the network graph. The two nodes can be any two directly connected nodes. Once the two nodes are selected, the next step is to obtain the interaction behavior data between them by consulting the relationship list between the two nodes. The interaction behavior data includes a list of behaviors between the two nodes, such as financial transactions, social media interactions, social (such as relatives, friends, classmates, colleagues), communication behaviors, peer behaviors, etc., as well as interaction characteristics, such as the number of transactions, transaction amount, social media interaction frequency, communication frequency, number of peers, etc.

[0031] S302: assign different behavior weights to the interaction behaviors according to the behavior categories, and calculate the influence index between two adjacent nodes according to the number of different interaction behaviors and the total number of interaction behaviors between the two nodes and the frequency of risk words in the interaction behavior data of the two nodes.

[0032] In one embodiment, the specific steps for calculating the impact index are as follows: First, calculate the interaction behavior weights of the two nodes: ; in, is the interaction behavior weight of the two nodes, Indicates the behavior category, is the total number of behavior categories, Indicates The behavior weight of each behavior category, Indicates The number of behavior categories, Indicates the total number of interactive behaviors.

[0033] Then, based on the interaction behavior weights of the two nodes and the frequency of risk words, combined with TF-IDF mining technology, the influence index between two adjacent nodes is calculated: ; in, Representation Node and nodes The impact index between Representation Node and nodes The interaction behavior weights between Indicated in The frequency of risky words in the interaction behavior data, Represents the inverse text frequency index of risk words.

[0034] In one embodiment, behavior categories can be directly distinguished by specific categories. For example, financial transactions, social media interactions, communication behaviors, peer behaviors, etc. are each regarded as a behavior category, and different behavior weights are assigned to different behavior categories. For example, the behavior weight of financial transactions is 0.3, the behavior weight of social media interactions is 0.05, the behavior weight of communication behaviors is 0.1, the behavior weight of peer behaviors is 0.15, and so on.

[0035] In another embodiment, the behavior categories can also be divided according to the risk level, for example, into serious risk behavior categories, general risk behavior categories and ordinary behavior categories. Public security personnel can determine which risk level behavior category each interactive behavior data belongs to based on case handling experience or logical reasoning, for example: fund transactions are classified into serious risk behavior categories, communication behaviors and peer behaviors are classified into general risk behavior categories, and social media interactions are classified into ordinary behavior categories. Different behavior weights are assigned to different behavior categories, for example, the behavior weight of the serious risk behavior category is 0.5, the behavior weight of the general risk behavior category is 0.35, and the behavior weight of the ordinary behavior category is 0.15.

[0036] In one embodiment, in step S300, a candidate node influence relationship network is constructed with the selected candidate node as the center, and the node inserted in the candidate node influence relationship network is determined by the influence index and the set threshold: According to the set first threshold, second threshold, ..., kth threshold, from the neighbor node set, two-hop neighbor node set, ..., k-hop neighbor node set of the candidate node, select the neighbor node whose comprehensive influence index with the candidate node is greater than the corresponding threshold and insert it into the candidate node influence relationship network; Among them, in the neighbor node set of the candidate node, the comprehensive influence index between the neighbor node and the candidate node is the influence index between the two nodes; in the k-hop neighbor node set, the comprehensive influence index between the k-hop neighbor node and the candidate node is calculated by the influence index between all nodes on the relationship path from the k-hop neighbor node to the candidate node.

[0037] See Figure 2 , Figure 2 The candidate node selected is Zhang San, Zhang San’s neighbor node set includes [Li Si, Li Wu, Li Liu, Li Qi, Li Ba], Zhang San’s two-hop neighbor node set includes [Zhao San, Zhao Si, Zhao Wu], and Zhang San’s three-hop neighbor node set includes [Chen San].

[0038] In this embodiment, the first threshold, the second threshold and the third threshold are set to 0.7, 0.65 and 0.5 respectively. In the neighbor node set of Zhang San, the influence index between Li Wu, Li Qi and Li Ba and Zhang San is greater than 0.7, so Li Wu, Li Qi and Li Ba are inserted into Zhang San's influence relationship network. The three nodes are also neighbor nodes, and Li Si and Li Liu are removed.

[0039] In one embodiment, the comprehensive influence index between the k-hop neighbor node and the candidate node is calculated by the average value of the influence index between all nodes on the relationship path. For example, among Zhang San's two-hop neighbor nodes, the relationship path between Zhao San and Zhang San is Zhao San-Li Qi-Zhang San, and the average value of the influence index of the relationship path is (0.79+0.78) / 2=0.785, which is greater than the second threshold value. In this case, Zhao San is inserted into Zhang San's influence relationship network, and the relationship paths of Zhao San and Zhang San remain the same. Among Zhang San's three-hop neighbor nodes, the relationship path between Chen San and Zhang San is Chen San-Zhao San-Li Qi-Zhang San, and the average value of the influence index of the relationship path is (0.79+0.78+0.95) / 3=0.84, which is greater than the third threshold value. In this case, Chen San is inserted into Zhang San's influence relationship network, and the relationship paths of Chen San and Zhang San remain the same. The influence relationship network of Zhang San formed according to this embodiment is as follows: Figure 3 shown.

[0040] In another embodiment, the comprehensive influence index between the k-hop neighbor node and the candidate node is calculated by weighted summing the influence indexes between all nodes on the relationship path. This calculation method takes into account the contribution of each node on the path to the comprehensive influence index, wherein the influence index of each node is assigned different weights according to its position and importance in the path. The determination of these weights can be based on the empirical judgment of experts, or it can be learned and optimized through a machine learning model. In practical applications, the setting of weights is a key step, which directly affects the accuracy of the final comprehensive influence index. Although there are various methods for setting weights, the specific implementation details will not be described in detail in this embodiment.

[0041] In one embodiment, in step S400, the method of confirming whether a candidate node is a potential risk person through the influence relationship network of the candidate node is specifically: S401: confirm whether the influence relationship network of each candidate node contains a risk node, and exclude candidate nodes that do not contain a risk node.

[0042] S402: For a candidate node whose influence relationship network includes a risk node, if the candidate node and the risk node are multi-hop neighbor nodes, the indirect influence index between the candidate node and the risk node is calculated through the relationship path between the candidate node and the risk node; in this embodiment, the calculation formula of the indirect influence index is: ; in, Represents a candidate node With risk nodes The indirect impact index, Represents a candidate node Candidate nodes on the relationship path The influence index between 1-hop neighbor nodes, Represents candidate nodes in the influence relationship network The average influence index between its neighbor nodes, Represents candidate nodes on the relationship path The influence index between the 1-hop neighbor node and the 2-hop neighbor node, Represents candidate nodes in the influence relationship network The average influence index between the 1-hop neighbor node and its neighbor nodes; Represents candidate nodes on the relationship path of Hop neighbor nodes and risk nodes The impact index between Represents candidate nodes on the relationship path of The average influence index between the hop neighbor node and its neighbor nodes; S403, calculating the influence of the candidate node in the influence relationship network through the influence relationship network of the candidate node; specifically, obtaining all interactive behavior data in the influence relationship network, determining the total number of interactive behaviors and the number of active interactive behaviors actively initiated by the candidate node and the number of passive behaviors passively accepted by the candidate node, and calculating the influence of the candidate node in the influence relationship network according to the following formula: ; in, Candidate node Influence in the network of influence relationships, is the preset behavior weight, Candidate node The number of proactive interactions initiated. Candidate node The number of passive behaviors of passive acceptance, is the total number of interactive behaviors; Based on the above scheme, this embodiment comprehensively considers the active interaction behavior and passive interaction behavior of the candidate node, and uses the preset behavior weights To balance the importance of both. A larger value indicates that the system focuses more on the active behavior of the candidate node; on the contrary, The smaller the value, the more emphasis is placed on the passive behavior of the candidate node. This calculation method can more comprehensively evaluate the actual influence of the candidate node in the influence relationship network.

[0043] S404, combining the influence index or indirect influence index between the candidate node and the risk node, and the influence of the candidate node in the influence relationship network, calculating the risk factor of the corresponding candidate node, and comparing the risk factor with the risk threshold to determine whether the corresponding candidate node is a potential risk person; The formula for calculating the risk factor is: .

[0044] in, Represents a candidate node risk factors; To dynamically adjust the weight, according to the candidate nodes The number of risk nodes in the impact relationship network is adjusted. Represents a candidate node The total number of risk nodes in the impact relationship network.

[0045] After calculating the risk factor of the candidate node, the risk factor is compared with the preset risk threshold. If the risk factor is greater than or equal to the risk threshold, the candidate node is determined to be a potential risk person; otherwise, it is determined to be a non-potential risk person. The value of the candidate node The number of risk nodes in the influence relationship network is positively correlated, because the candidate nodes The more risk nodes there are in the influence relationship network, the more candidate nodes The greater the probability of being exposed to or affected by risk nodes.

[0046] In one embodiment, the view data includes personnel image data and vehicle image data; the business data includes identity data, call record data, financial data and registration data.

[0047] This implementation also includes preprocessing of business data, specifically: For business data, the data mainly consists of structured data such as form data and flow data. The processing of this type of data includes data cleaning, filtering, conversion, etc.

[0048] For example, the bank's fund flow data, the data processing process is as follows: 1) Key data acquisition: Obtain fund flow data from the database, check key data such as fund transaction time, transaction amount, transaction account, counterparty account, transaction type, transaction location, etc., and obtain complete resources.

[0049] 2) Data cleaning: Data deduplication: Check and delete duplicate transaction records to avoid interference with analysis results.

[0050] Format conversion: Ensure that all date and time fields are in the same format, such as "YYYY-MM-DD HH:MM:SS". For amount fields, ensure that the number of digits after the decimal point is consistent.

[0051] Remove abnormal characters: In text fields (such as transaction notes), remove possible illegal characters or garbled characters.

[0052] 3) Data conversion: Unify the amount unit: If there are amounts in different currency units in the data, they need to be converted into the same currency unit (such as RMB).

[0053] Timestamp conversion: Convert timestamps into a format that is easy to analyze, such as extracting information such as the hour and day of the week when the transaction occurred, so that time-related patterns can be analyzed later.

[0054] Data filtering: Filter transfer transactions: Filter records of the "transfer" type from all transaction types.

[0055] Filter small transactions: Based on business experience, small transactions are usually unlikely to be abnormal transactions. Filter data with transaction amounts below 500 yuan.

[0056] Preprocess the view class data, specifically: The data processing process is as follows: 1) Data collection: Access the image data of various types of people, vehicles, etc. collected by the road area capture device.

[0057] 2) Image modeling: Using the view analysis algorithm deployed locally by the public security department, target features are extracted and modeled, such as facial contours, eyes, mouth, nose, etc. of people, license plates, vehicle brands, license plate numbers, license plate colors, vehicle colors, etc. of vehicles.

[0058] 3) Data filtering: According to the data results after modeling and analysis, invalid data is filtered out, such as no-target pictures, blurred pictures, no license plate pictures, incomplete target capture pictures, etc., to ensure the validity of the data.

[0059] 4) Model matching: The captured image data after modeling and analysis is matched with the local personnel list library. If the model similarity between the captured data and the local data reaches the set threshold, the identity information label of the local person is added to the data. If there are multiple matching data, the identity information in the list with the highest similarity shall prevail.

[0060] 5) Data Archiving: The data after modeling and processing is stored and archived in a locally deployed distributed storage system for storage, and efficient data retrieval and query are provided.

[0061] After completing data preprocessing, data association is still required. For business data, the key fields of various business data are used to associate the data. For example, for fund flow data, the transfer account is used as the key field for data association to form a penetrable fund data chain; for call record data, the communication number is used as the key field for data association to form call association relationship data; for accommodation data, the registered identity information of the accommodation personnel is used as the key field to form the accommodation record association data of the personnel.

[0062] For view data, it is mainly associated with the identity information of the person and the license plate information of the vehicle. For example, for personnel image data, the identity information tag added by matching the local list is used as the personnel identity ID for association. The personnel images are aggregated according to the identity ID to form personnel file data with the person as the individual unit; the vehicle image data is associated with the license plate number after analysis as the ID to form vehicle file data with the license plate number as the individual unit.

[0063] After preprocessing and data association, the business data and view data are fused with graph and data. The identity ID of the personnel is used as the key field, and the business data and view data are fused with the identity ID to form a graph-data fused personnel-specific database, including various types of personnel's call list data, accommodation data, financial data, Internet data, snapshot data, vehicle data, etc., which can all be comprehensively queried through an identity ID.

[0064] In one embodiment, a method for further determining persons who are closely connected with risk persons or associated persons based on the graph-data fusion data is as follows: You can input the real-name identity information of risk persons or related persons, information of related companions, travel location information, travel time information, and travel companion snapshot data information, set analysis conditions for the number of travels, number of days traveled, number of companions, and travel locations, and output the identity information of the companions of risk persons or related persons who meet the analysis conditions.

[0065] You can input the mobile phone communication data of risk persons or related persons, including call time, call duration, the other party's number, the operator they belong to, and the call location area. You can set analysis conditions such as call duration, call days, number of calls, and call location to determine the closeness of communication with risk persons or related persons, and output information on close communications with risk persons or related persons who meet the conditions.

[0066] You can input the train booking data, air ticket booking data, and online car-hailing travel data of risk persons or related persons, including travel time, travel type, departure location, arrival location, people on the same train, and people on the same plane, set analysis conditions such as travel time period, travel location, and people traveling together, and output information on people traveling with risk persons or related persons who meet the conditions.

[0067] You can input vehicle travel data of risk persons or related persons, including vehicle capture records, license plate data, vehicle passenger data, and vehicle travel time, set analysis conditions such as vehicle travel time, vehicle travel location, vehicle travel route, number of vehicle rides, number of vehicle rides, and number of days the vehicle rides together, and output information on passengers traveling in the same vehicle for risk persons or related persons who meet the conditions.

[0068] You can input the accommodation data of risk persons or related persons, including accommodation time, accommodation hotel, accommodation floor, accommodation room, and people staying together, set analysis conditions such as accommodation in the same hotel, accommodation in the same room, number of stays, number of days of stay, and accommodation location, and output the information of people staying in the same hotel or room as the risk persons or related persons who meet the conditions.

[0069] Based on the above implementation plan, this embodiment can filter out multi-dimensional related information of risk personnel from massive business data and view data, and through the method of constructing relationship diagrams and influence diagrams, use multi-dimensional data to warn the related personnel of risk personnel and discover potential risk personnel. It can analyze potential risk personnel more accurately from more dimensions.

[0070] Embodiment 2: This embodiment provides a potential risk personnel identification system based on image and data fusion, which is characterized by comprising: A graph-data fusion information base construction module is used to collect the view data and business data of risk personnel with criminal records and fuse them to form graph-data fusion data, and create a fusion information base based on the graph-data fusion data; this module is used to implement the function of step S100 in the first embodiment, and will not be repeated here; A relationship network diagram construction module is used to extract risk personnel and related personnel in the fusion information library as nodes, construct a relationship network diagram, and collect related personnel information to supplement the nodes in the relationship network diagram. According to the fusion information library and related personnel information, a node attribute list and a relationship list are created, and the node corresponding to the risk personnel is defined as a risk node, the node corresponding to the related personnel is defined as a risk neighbor node, and other nodes are defined as ordinary nodes. This module is used to implement the function of step S200 in the first embodiment, and will not be repeated here. The influence network construction module is used to calculate the influence index between two adjacent nodes through the relationship between each node, select any risk neighbor node or ordinary node as a candidate node, and build an influence relationship network of the candidate node with the selected candidate node as the center. The node inserted in the influence relationship network of the candidate node is determined by the influence index and the set threshold; this module is used to implement the function of step S300 in the first embodiment, and will not be repeated here; The potential risk personnel identification module is used to confirm whether the candidate node is a potential risk person through the influence relationship network of the candidate node; this module is used to implement the function of step S400 in the first embodiment, and will not be repeated here.

[0071] Embodiment three: This embodiment proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for identifying persons with potential risks based on image-data fusion as described in any embodiment of the present invention is implemented.

[0072] Embodiment 4: This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for identifying persons with potential risks based on image-data fusion as described in any embodiment of the present invention is implemented.

[0073] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0074] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0075] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0076] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0077] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for identifying potential risk personnel based on image-data fusion, characterized in that: The following steps are involved: Collect the visual data and business data of risk personnel with criminal records and fuse them into graph-number fusion data, and create a fusion information database based on the graph-number fusion data; Extract risk personnel and related personnel in the fusion information database as nodes, build a relationship network diagram, and collect related personnel information to supplement the nodes in the relationship network diagram. Create a node attribute list and relationship list based on the fusion information database and related personnel information, and define the node corresponding to the risk personnel as a risk node, the node corresponding to the related personnel as a risk neighbor node, and other nodes as ordinary nodes; Through the relationship between each node, the influence index between two adjacent nodes is calculated, and any risk neighbor node or ordinary node is selected as a candidate node. With the selected candidate node as the center, an influence relationship network of the candidate node is constructed. The nodes inserted into the influence relationship network of the candidate node are determined by the influence index and the set threshold; Confirm whether the candidate node is a potential risk person through the influence relationship network of the candidate node.

2. According to claim 1, a method for identifying potential risk personnel based on image-data fusion is characterized in that: The step of calculating the influence index between two adjacent nodes through the relationship between the nodes includes: Select two adjacent nodes, and obtain the interaction behavior data of the two nodes through the attribute lists and relationship lists of the two nodes; Different behavior weights are assigned to interaction behaviors according to behavior categories. The impact index between two adjacent nodes is calculated based on the number of different interaction behaviors and the total number of interaction behaviors between the two nodes and the frequency of risk words in the interaction behavior data of the two nodes.

3. The method for identifying potential risk personnel based on image-data fusion according to claim 1 is characterized in that: In the step of constructing a candidate node influence relationship network with the selected candidate node as the center, and the node inserted in the candidate node influence relationship network is determined by the influence index and the set threshold: According to the set first threshold, second threshold, ..., kth threshold, from the neighbor node set, two-hop neighbor node set, ..., k-hop neighbor node set of the candidate node, select the neighbor node whose comprehensive influence index with the candidate node is greater than the corresponding threshold and insert it into the candidate node influence relationship network; Among them, in the neighbor node set of the candidate node, the comprehensive influence index between the neighbor node and the candidate node is the influence index between the two nodes; in the k-hop neighbor node set, the comprehensive influence index between the k-hop neighbor node and the candidate node is calculated by the influence index between all nodes on the relationship path from the k-hop neighbor node to the candidate node.

4. The method for identifying potential risk personnel based on image-data fusion according to claim 3 is characterized in that: The method for confirming whether a candidate node is a potential risk person through the influence relationship network of the candidate node is specifically as follows: Confirm whether the influence relationship network of each candidate node contains risk nodes, and exclude candidate nodes that do not contain risk nodes; For candidate nodes whose influence relationship network includes risk nodes, if the candidate node and the risk node are multi-hop neighbor nodes, the indirect influence index between the candidate node and the risk node is calculated through the relationship path between the candidate node and the risk node; And calculate the influence of the candidate node in the influence relationship network through the influence relationship network of the candidate node; Combined with the influence index or indirect influence index between the candidate node and the risk node, as well as the influence of the candidate node in the influence relationship network, the risk factor of the corresponding candidate node is calculated, and the risk factor and risk threshold are compared to determine whether the corresponding candidate node is a potential risk person.

5. The method for identifying potential risk personnel based on image-data fusion according to claim 1 is characterized by: The view data includes personnel image data and vehicle image data; the business data includes identity data, call record data, capital data and registration data.

6. The method for identifying potential risk personnel based on image-data fusion according to claim 5 is characterized in that: The method further comprises: Pre-process business data, including data cleaning and data standardization; Preprocess view data, including view data filtering, target extraction, and target matching.

7. The method for identifying potential risk personnel based on image-data fusion according to claim 1 is characterized in that: In the step of extracting risk personnel and related personnel in the fusion information database as nodes, and collecting related personnel information to supplement the nodes in the relationship network diagram: By fusing the graph and number data of risk personnel, we can identify the close contact personnel as related personnel; After determining the related persons, collect the graph fusion data of the related persons and determine the closely connected persons as supplementary nodes.

8. A potential risk personnel identification system based on image and data fusion, characterized in that: include: A graph-data fusion information base construction module is used to collect the view data and business data of risk personnel with criminal records and fuse them to form graph-data fusion data, and create a fusion information base based on the graph-data fusion data; The relationship network diagram construction module is used to extract risk personnel and related personnel in the fusion information database as nodes, construct a relationship network diagram, and collect related personnel information to supplement the nodes in the relationship network diagram. According to the fusion information database and related personnel information, the attribute list and relationship list of the nodes are created, and the nodes corresponding to risk personnel are defined as risk nodes, the nodes corresponding to related personnel are defined as risk neighbor nodes, and other nodes are defined as ordinary nodes. The influence network construction module is used to calculate the influence index between two adjacent nodes through the relationship between each node, select any risk neighbor node or ordinary node as a candidate node, and build an influence relationship network of the candidate node with the selected candidate node as the center. The node inserted in the influence relationship network of the candidate node is determined by the influence index and the set threshold; The potential risk personnel identification module is used to confirm whether the candidate node is a potential risk person through the influence relationship network of the candidate node.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for identifying persons with potential risks based on image and data fusion as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for identifying persons with potential risks based on image-data fusion as described in any one of claims 1 to 7 is implemented.

Images