CN110659560B - Method and system for identifying associated object - Google Patents

Abstract

The invention is suitable for the technical field of data processing, and provides a method and a system for identifying a related object, wherein the method comprises the following steps: the information acquisition terminal acquires object information of a plurality of different types of target objects; the object information includes position information of the target object; the information analysis system receives the object information, extracts an object identifier of a target object from each object information, and generates an object data packet according to the object identifier and the object information; the data processing system receives the object data packet and constructs a moving track of the target object according to a plurality of pieces of position information of the same object identifier; and the data processing system establishes the association relation between the target objects of different types based on the movement track. According to the invention, the incidence relation of different target objects is established, and when certain type of object information cannot be acquired, the position tracking of the user can be continuously carried out by acquiring other types of object information, so that the application range of user information acquisition and the information acquisition efficiency can be improved.

Description

Method and system for identifying associated object

Technical Field

The invention belongs to the technical field of data processing, and particularly relates to a method and a system for identifying a related object.

Background

With the continuous development of information acquisition technology and information recognition technology, the position of the moving track of the target object can be tracked, for example, the face images of pedestrians are acquired through the camera modules distributed in each area, the face image of the target object is selected from the face images of the pedestrians, the moving track of the target object is created through the face images of the target object acquired at different moments, and therefore the position of the target object is tracked.

However, in the above-mentioned user information collecting technology, when the face is blocked, or when the target object gets on a related vehicle, etc., the face image of the user cannot be effectively collected, and thus the target object cannot be tracked. Therefore, in the existing user information acquisition technology, the acquisition mode is single, and operations such as user place value tracking and the like cannot be realized under the condition that key information cannot be acquired, so that the information acquisition efficiency and the application range are reduced.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method and a system for identifying a related object, so as to solve the problems that in the existing user information acquisition technology, an acquisition manner is single, and operations such as user bit value tracking cannot be implemented when critical information cannot be acquired, thereby reducing the efficiency of information acquisition and the application range.

A first aspect of an embodiment of the present invention provides an identification method for an associated object, which is applied to an identification system for an associated object, where the identification system for an associated object includes: the system comprises an information acquisition terminal, an information analysis system and a data processing system;

the identification method of the associated object comprises the following steps:

the information acquisition terminal acquires object information of a plurality of different types of target objects; the object information includes position information of the target object;

the information analysis system receives the object information sent by each information acquisition terminal, extracts an object identifier of the target object from each object information, and generates an object data packet according to the object identifier and the object information;

the data processing system receives the object data packet sent by the information analysis system, and constructs a moving track of the target object corresponding to the object identifier according to a plurality of position information of the same object identifier;

the data processing system establishes incidence relations among the target objects of different types based on the moving track; the different types of target objects with the incidence relation correspond to the same entity user.

A second aspect of the embodiments of the present invention provides an identification system of an associated object, where the identification system of an associated object includes: the system comprises an information acquisition terminal, an information analysis system and a data processing system;

the information acquisition terminal is used for acquiring object information of a plurality of different types of target objects; the object information includes location information of the target object;

the information analysis system is used for receiving the object information sent by each information acquisition terminal, extracting the object identifier of the target object from each object information, and generating an object data packet according to the object identifier and the object information;

the data processing system is configured to receive the object data packet sent by the information analysis system, and construct a movement trajectory of the target object corresponding to the object identifier according to a plurality of pieces of location information of the same object identifier;

the data processing system is used for establishing incidence relations among the target objects of different types based on the moving track; the different types of target objects with the incidence relation correspond to the same entity user.

The method and the system for identifying the associated object provided by the embodiment of the invention have the following beneficial effects:

the embodiment of the invention obtains the object information of different types of target objects through the information acquisition terminal, analyzes the object information through the information analysis system, extracts the object identifier of the target object, then obtains a plurality of object information of the same object identifier through the data processing system, establishes the moving track of the target object corresponding to the object identifier, and finally determines a plurality of different types of target objects with incidence relation through collision recognition of the plurality of moving tracks, thereby realizing automatic recognition of the incidence objects. Compared with the existing information acquisition technology, the information acquisition terminal can acquire different types of target objects, such as object information of the types of a mobile terminal, a user face, a license plate number and the like, establish moving tracks of different target objects, identify whether different target objects correspond to the same entity user or not through the similarity of the moving tracks of different target objects, establish the association relation of different target objects, and continuously track the position of the user through acquiring other types of object information when certain type of object information cannot be acquired, so that the application range of user information acquisition and the efficiency of information acquisition can be improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.

Fig. 1 is an interaction flowchart of a method for identifying an associated object according to a first embodiment of the present invention;

fig. 2 is a flowchart of a specific implementation of the method for identifying an associated object S104 according to a second embodiment of the present invention;

fig. 3 is a flowchart of a specific implementation of a method for identifying an associated object according to a third embodiment of the present invention;

fig. 4 is a flowchart of a detailed implementation of the method for identifying an associated object S102 according to a fourth embodiment of the present invention;

fig. 5 is a flowchart illustrating an implementation of the method S1023 for identifying an associated object according to a fifth embodiment of the present invention;

fig. 6 is a flowchart of a specific implementation of a method for identifying an associated object according to a sixth embodiment of the present invention;

FIG. 7 is a block diagram of a system for identifying related objects according to an embodiment of the present invention;

Fig. 8 is a block diagram of a system for identifying an associated object according to another embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.

The embodiment of the invention obtains the object information of different types of target objects through the information acquisition terminal, analyzes the object information through the information analysis system, extracts the object identifier of the target object, then obtains a plurality of object information of the same object identifier through the data processing system, establishes the moving track of the target object corresponding to the object identifier, and finally determines a plurality of different types of target objects with association relationship through collision recognition of the plurality of moving tracks, thereby realizing automatic identification of the associated objects.

In the embodiment of the present invention, the execution subject of the flow is an identification system of the associated object, and the identification system of the associated object at least includes: the system comprises an information acquisition terminal, an information analysis system and a data processing system. Fig. 1 shows an interaction flowchart of a method for identifying an associated object according to a first embodiment of the present invention, which is detailed as follows:

in S101, the information acquisition terminal acquires object information of a plurality of different types of target objects; the object information includes position information of the target object.

In this embodiment, the identification system of the associated object may be configured with a plurality of information acquisition terminals, each of the information acquisition terminals may be fixedly disposed in a preset monitoring area, and acquire object information of a target object passing through the monitoring area, and optionally, the information acquisition terminal may move based on a preset driving path, for example, the information acquisition terminal may be a driving recorder on a vehicle, and acquire object information of the target object monitored by the driving path. The plurality of information acquisition terminals can form a distributed information acquisition system of the identification system of the associated object, the identification system of the associated object can mark each information acquisition terminal on a preset monitoring map interface, and generate a monitoring area of the identification system of the associated object according to an effective acquisition area of the information acquisition terminal, and a manager can perform point supplementation on the monitoring area marked by the monitoring map interface, namely, add the information acquisition terminal in the area needing to be monitored, thereby expanding the monitoring range and improving the management efficiency of the monitoring area.

In this embodiment, the object information collected by the information collection terminal is related to the device type of the information collection terminal. For example, if the information acquisition terminal is a camera module, the acquired object information is image information; if the information acquisition terminal is a base station device, the acquired object information is device information accessed to the base station device. Specifically, the information acquisition terminal may transmit a mobile local area network signal, which is used to cover mobile signals of all mobile terminals in a specific area, and if a certain mobile terminal enters the area, it needs to send a device registration instruction to the base station device, where the device registration instruction carries terminal information of the mobile terminal, such as a physical address of the mobile terminal, for example, an MAC address or a mobile IMEI code, and may also be a network address of the mobile terminal, for example, an IP address. The information acquisition terminal can also be a wireless access device which can transmit wireless local area network signals in an installation area, the mobile terminal can be accessed to the wireless local area network through a built-in wireless communication module, and then the wireless access device can acquire terminal information of the accessed mobile terminal.

In this embodiment, the information collecting terminal may be provided with an effective object list, where the effective object list is provided with a plurality of different types of object types of the object information to be collected, and if the information collecting terminal detects an object type recorded in the effective object list, the object is identified as a target object, and object information about the target object is collected. For example, the information collecting apparatus may set a vehicle and a human valid object type, recognize image information of the frame as object information if there is a vehicle or a human in the detection screen, and upload the object information to the information parsing system.

In this embodiment, the object information includes, in addition to information related to the target object, for example, image information of the target object or an identifier related to the target object is recorded, and also includes position information, and the position information may be obtained by feedback from the target object, for example, the target object is a mobile terminal, and in the process of acquiring the object information, the mobile terminal may send positioning information to an information acquisition terminal, so that the position information corresponding to the acquisition of the object information may be determined by the positioning information; if the information acquisition terminal is fixedly placed, the installation position of the information acquisition terminal can be used as the position information of the object information.

Optionally, in this embodiment, after acquiring the object information of the target object, the information acquisition terminal may perform preprocessing on the object information, and in particular, if the object information is an image data type, the information acquisition terminal may perform noise reduction on the image data and extract the image data through an image recognition algorithm to generate the object information. If the object information cannot be extracted by the image recognition algorithm, the object information can be recognized as invalid information and discarded, thereby reducing feedback of invalid data. Preferably, the identification system of the associated object further includes a middleware, and the information acquisition terminal may send the information type of the obtained object information to the middleware corresponding to the information type, preprocess the object information through the middleware, and send the preprocessed object information to the information analysis system, thereby improving the processing rate of the object information and reducing the transmission amount of invalid output.

In S102, the information analysis system receives the object information sent by each information acquisition terminal, extracts an object identifier of the target object from each object information, and generates an object data packet according to the object identifier and the object information.

In this embodiment, after receiving the object information, the information parsing system may identify an information type of the object information, and extract an object identifier included in the object information through a parsing algorithm corresponding to the information type, where the object identifier is used to uniquely identify the target object. For example, a face image of the physical user, a license plate number of the vehicle, a physical address of the mobile terminal, and the like. Specifically, the information analysis system detects a subject type of a subject included in the image information if it is detected that the object information is image information, extracts a region image corresponding to the subject if the subject type is in the valid object list, and extracts an object identifier of the target object from the region image by a corresponding recognition algorithm. For example, if the license plate number needs to be extracted, extracting character information from the vehicle region image through a character recognition algorithm, and recognizing the character information as the license plate number of the target vehicle; if the face information needs to be extracted, the face region in the human body region image is positioned through the preset face characteristic points, and therefore the face image corresponding to the target is extracted and obtained.

Optionally, in this embodiment, the information analysis system may include a plurality of information analysis nodes, and after receiving the object information sent by the information acquisition terminal, the information analysis system may perform cluster-distributed deployment on the object information in each information analysis node, so as to avoid a situation where data is lost due to downtime of each node, and improve data security.

Optionally, in this embodiment, different information analysis nodes in the information analysis system may be configured to process object information uploaded in different areas, and in this case, each information analysis node may be fixedly associated with at least one information acquisition terminal, and generate a forwarding routing table based on a correspondence between the information analysis node and the information acquisition terminal. After receiving the object information sent by the information acquisition terminal, the information analysis system queries the forwarding routing table to determine an information analysis node to which the information acquisition terminal belongs, and sends the object information to the information analysis node for analysis processing.

Optionally, in this embodiment, different information analysis nodes in the information analysis system and the information acquisition terminals form a chain structure, that is, the information acquisition terminals of different information types correspond to different information analysis nodes, and the object information is uploaded to the information analysis node matched with the information type of the object information for analysis processing. If a new target object is added to the identification system of the associated object, and the information type of the object information of the target object is not matched with the existing information type, a corresponding information analysis node can be newly added in the information analysis system, and a new analysis chain is formed by the information analysis node and the newly added information acquisition terminal, so that the extension of the object type is realized. If the identification system of the associated object needs to delete a certain type of target object, the analysis chain associated with the information type is correspondingly deleted, and the analysis algorithm of the information analysis node corresponding to the analysis chain is configured to be the analysis algorithm of the rest of the existing other information types, so that the resource utilization rate of the identification system of the associated object can be improved.

Optionally, the information analysis system may be configured with a standby information analysis node, and if it is detected that the task amount of the current analysis task is greater than a preset first task threshold, the uploaded object information may be analyzed through the standby information analysis node; if the task quantity of the current analysis task is smaller than the preset second task threshold value, the information analysis node in use can be provided with a standby information node, and therefore the resource utilization rate is improved.

In this embodiment, the information analysis system may encapsulate the extracted object identifier and the object information to generate an object data packet, and certainly, after the source data is extracted, a part related to the source data in the object information may be deleted, and only the object identifier is reserved, so that the size of the object data packet may be reduced. For example, the information parsing system may delete the object image and retain only the extracted face image.

In S103, the data processing system receives the object data packet sent by the information analysis system, and constructs a movement trajectory of the target object corresponding to the object identifier according to the plurality of pieces of location information of the same object identifier.

In this embodiment, the data processing system may receive an object data packet sent by the information analysis system, where the object data packet includes an object identifier and location information corresponding to the acquisition time of the object data packet, and the data processing system may classify the data packets according to the object identifier included in the data packet, regard the object data packets belonging to the same object identifier as one data group, and draw a movement trajectory about the target object on a preset map interface according to the data group including a plurality of location information. Particularly, each object data packet has a timestamp, and the data processing system can determine the sending sequence of each object data packet by detecting the timestamps, so that each position information is sequentially connected based on the sequence, and a moving track with a moving direction is obtained.

In this embodiment, in the process of establishing the moving track by the data processing system, the information acquisition terminal may continuously acquire the object information of the target object, and send the object data packet of the object information through the information analysis system, and in this case, the data processing system may update the moving track in real time, thereby implementing real-time dynamic tracking of the target object.

Optionally, in this embodiment, the data processing system may be provided with a graphic database, and if it is detected that the object data packet includes image data, the image data in the object data packet is extracted, and then the image data is stored in the storage area associated with the object identifier carried in the object data packet, so as to create an image library corresponding to each target object, thereby facilitating subsequent query operations. Specifically, the data processing system can create a plurality of image databases according to the object types, and store the image data of different types of target objects into the image databases associated with the target objects, so that the efficiency of data storage and data query can be improved.

In S104, the data processing system establishes an association relationship between the target objects of different types based on the movement trajectory; the different types of target objects with the incidence relation correspond to the same entity user.

In this embodiment, after determining the movement trajectories of the different types of target objects, the data processing system may calculate the association degrees between the different types of target objects according to the movement areas and the movement times of the movement trajectories, and if the difference degrees between the movement areas of the movement trajectories of the two target objects are smaller and the movement times are the same, the probability that the two target objects correspond to the same entity user is higher, so that the matching factor between the two target objects may be determined according to the movement areas and the movement times, and the matching degree between the two target objects may be calculated based on the two matching factors. If the matching degree of the two target objects is greater than the preset matching degree threshold value, identifying that the two types of target objects correspond to the same entity user, for example, the target object A is a vehicle, the target object B is a mobile terminal, and if the matching degree of the two target objects is greater than the preset matching degree threshold value, determining that the vehicle and the mobile terminal belong to the same entity user. Based on this, the entity user can be tracked by acquiring the position information of the mobile terminal under the condition that the entity user does not drive the vehicle.

As can be seen from the above, in the identification method for associated objects provided in the embodiments of the present invention, the information acquisition terminal acquires object information of target objects of different types, the information analysis system analyzes the object information, the object identifier of the target object is extracted, the data processing system acquires a plurality of object information of the same object identifier, the moving track of the target object corresponding to the object identifier is established, and finally, collision identification is performed on the plurality of moving tracks, so that a plurality of target objects of different types having an association relationship can be determined, and automatic identification of the associated object is achieved. Compared with the existing information acquisition technology, the information acquisition terminal can acquire different types of target objects, such as object information of the types of a mobile terminal, a user face, a license plate number and the like, establish moving tracks of different target objects, identify whether different target objects correspond to the same entity user or not through the similarity of the moving tracks of different target objects, establish the association relation of different target objects, and continuously track the position of the user through acquiring other types of object information when certain type of object information cannot be acquired, so that the application range of user information acquisition and the efficiency of information acquisition can be improved.

Fig. 2 shows a flowchart of a specific implementation of the method for identifying an associated object S104 according to a second embodiment of the present invention. Referring to fig. 2, an execution subject of the embodiment of the present invention is the data processing system, and with respect to the embodiment shown in fig. 1, the method for identifying an associated object S104 provided in this embodiment includes: s1041 to S1043, which are described in detail below:

further, the object information further includes time information; the data processing system establishes incidence relations of the target objects of different types based on the moving track, and the incidence relations comprise:

in S1041, the data processing system determines the movement trigger time of the movement trajectory according to the time information corresponding to each piece of the position information in the movement trajectory.

In this embodiment, when the information acquisition terminal acquires the object information, the time information of the acquisition time may be encapsulated in the object information. The data processing system may extract time information from the object data packet and determine a corresponding acquisition time for the location information. The data processing system can arrange the position information based on the time sequence of the position information, and takes the earliest time information as the movement trigger time of the movement track.

Optionally, in this embodiment, the data processing system may be a server cluster formed based on a Spark system, and since the Spark system has high efficiency of joint query in time and region, multiple object data packets located in the same location region can be quickly extracted, and time stamps of the object data packets are classified, so as to establish moving tracks related to different target objects, perform collision analysis based on the moving tracks, determine matching degrees between the moving tracks, establish association relationships between different types of target objects, and improve identification efficiency and accuracy of the association relationships. In this case, the Spark cluster system may obtain load parameters of each server in the cluster, select a target server from each available server according to a load balancing algorithm, send a task of creating a movement trajectory to the target server, and generate the movement trajectory of the target object through the target server.

In S1042, the data processing system extracts a plurality of target trajectories with the same movement trigger time from all the movement trajectories, and calculates a coincidence rate between the plurality of target trajectories.

In this embodiment, after determining the movement trigger time of the movement tracks, the data processing system may select multiple movement tracks with the same movement trigger time from all the movement tracks, divide the extracted multiple movement tracks into the same movement track group, and perform a subsequent coincidence rate calculation process. If different types of target objects belong to the same entity user, the movement trigger time and the movement path of the two target objects should be completely overlapped, that is, two movement tracks with higher overlapping rate exist in the multiple target objects with the association relationship in the same time. Based on the above, the data processing system may select multiple target tracks that may have an association relationship according to the movement trigger time.

In this embodiment, the data processing system may perform repetition rate calculation on a plurality of target tracks, and the calculation process may be: selecting a target track in the same movement trigger area as a candidate track; drawing the candidate tracks on a preset map interface, calculating the coincidence length between any two moving tracks, and calculating the coincidence rate between the two moving tracks based on the total length and the coincidence length of the moving tracks.

In S1043, the data processing system selects the target track with the coincidence rate greater than a preset coincidence threshold as an association track, and establishes an association relationship between the target objects corresponding to the association track.

In this embodiment, if the coincidence rate of the two movement trajectories is high, it indicates that the probability that the two target objects correspond to the same entity user is higher. Based on this, the data processing system may identify a plurality of target tracks with coincidence rates greater than a preset coincidence threshold as associated tracks, and identify that the entity users corresponding to the target objects corresponding to the two associated tracks are the same, for example, if the coincidence rate of the moving track of a certain vehicle and the moving track of another mobile terminal is greater than the preset coincidence threshold, it may be determined that the owner of the vehicle and the affiliated user of the mobile terminal are the same person, and at this time, the data processing system may establish an association relationship between the two or more target objects with coincidence rates greater than the coincidence threshold. It should be noted that the maximum number may be set for different object types in the association relationship, that is, the number of each object type in the same association relationship cannot exceed the maximum number. For example, the maximum number corresponding to the face type is 1, and the maximum number corresponding to the mobile terminal may be 5, that is, the face images of the same entity user may only correspond to one, and one entity user may carry a plurality of mobile terminals, in this case, the number of the target objects that may include the mobile terminal type in the association relationship may be multiple.

Optionally, when determining that the coincidence rate between the moving trajectories of the two target objects is greater than a preset coincidence rate threshold, the data processing system may acquire the moving trajectories of the two target objects corresponding to other moving trigger times, calculate the coincidence rate between the moving trajectories corresponding to the other moving trigger times, and count the number of the moving trigger times at which the coincidence rate between the moving trajectories of the target objects is greater than the preset coincidence rate threshold, if the number of the moving trigger times is greater than the preset number threshold, identify that an association relationship exists between the two target objects, and establish the association relationship between the target objects.

In the embodiment of the invention, the target tracks can be selected by determining the movement trigger time of each movement track, so that a large amount of invalid coincidence rate calculation operations can be reduced, the identification efficiency of the association relation is improved, the association relation between the target objects can be accurately established by calculating the coincidence rate between different types of target objects, and the identification accuracy of the association relation is improved.

Fig. 3 shows a flowchart of a specific implementation of a method for identifying an associated object according to a third embodiment of the present invention. Referring to fig. 3, with respect to the embodiment described in fig. 1, an execution subject of the embodiment of the present invention is the data processing system, and the method for identifying a related object provided in this embodiment further includes: s301 to S305 are described in detail as follows:

Further, the data processing system comprises a data search server based on a distributed full-text retrieval ElasticSearch cluster; after the data processing system establishes the association relationship between the target objects of different types based on the movement trajectory, the method further includes:

in S301, the data search server receives an inquiry request sent by a user terminal; the query request comprises a target time period of a required query and a target user identification.

In this embodiment, the data processing system further includes an elastic search server cluster (ES server cluster), where the ES server cluster includes a plurality of data search servers, and the data search servers are used to undertake data search tasks initiated by users. Optionally, in this embodiment, the ES server cluster supports dynamic expansion, that is, multiple standby servers may be configured in the ES server cluster, and if the ES server detects that the load parameters of each current data search server are greater than a preset maximum operation threshold, it is identified that the current ES server cluster is in an overload state, and it is necessary to perform an expansion operation on the ES server cluster, at this time, the standby server may be started, and a data search task to be executed is sent to the enabled standby server for response.

In this embodiment, the ES server cluster may create a corresponding data index for each target object according to the object identifier of each target object, divide the corresponding data storage space within the server cluster, and store all object data packets related to the object identifier in the corresponding data storage space. When receiving a query task sent by a user terminal, the data storage space corresponding to the corresponding object identifier can be determined according to the data index table, and a query operation is responded.

In this embodiment, the user terminal may send, through a locally installed client, an inquiry request to the ES server cluster, where the inquiry request includes a user identifier of a target user to be inquired and a target time period to be inquired. It should be noted that the target time period may be configured by default, and if the target time period is empty, it is identified that the user needs to query the movement track of the target user at the current time.

In S302, the data search server obtains the association relationship corresponding to the target user identifier, and determines a plurality of target objects of different types associated with the target user identifier based on the association relationship.

In this embodiment, after receiving the query request, the data search server extracts the target user identifier included in the query request, and queries an association relationship corresponding to the target user identifier in the association relationship list, where the association relationship records a plurality of target objects of different types belonging to the target user, such as a target face, a target vehicle, and a target terminal.

In S303, the data search server extracts, from all the object information uploaded by the information acquisition terminal, the object information whose acquisition time is within the target time period as candidate information.

In this embodiment, the data search server performs screening on all object information uploaded by the information acquisition terminal according to a target time period carried by the query information, and selects the object information whose acquisition time is within the target time period as candidate information, so that a movement track of the target user at any time can be constructed.

It should be noted that the object information may be stored in each data search server in the ES server cluster, the data search servers may perform concurrent search in the server cluster for the time period required for query, and each data search server feeds back the matched object information to the data search server initiating the search requirement, thereby achieving the purpose of searching in the whole network.

In S304, the data search server extracts, from the candidate information, candidate information about any type of the target object associated with the target user identifier as target information.

In this embodiment, the data search server detects whether the fed back candidate information includes object information of a target object associated with the target user identifier, and if the fed back candidate information includes the object information of the target object associated with the target user identifier, the candidate information is identified as the target information, so that a real-time dynamic trajectory of the target user can be constructed by extracting the obtained target information about the target user.

In S305, the data search server marks the target information on a preset map interface, and generates a real-time dynamic trajectory of the target user identifier according to the sequence of the acquisition times of the target information.

In this embodiment, the data search server may mark the extracted target information on the map interface, establish a real-time dynamic trajectory having a moving direction based on the sequence of the acquisition time according to each target information, and display the query result to the querying user through the client of the user terminal.

In the embodiment of the invention, the ES server cluster formed by a plurality of data search servers responds to the data query request initiated by the user terminal, and because the ES server cluster is a highly-telescopic open-source full-text search and analysis engine, the ES server cluster can rapidly store, search and analyze a large amount of data in a near real-time manner; meanwhile, the ElasticSearch provides a latitude and longitude search-based function, provides query operation for time/region search, and improves search efficiency and dynamic track display effect.

Fig. 4 shows a flowchart of a specific implementation of the method for identifying an associated object S102 according to a fourth embodiment of the present invention. Referring to fig. 4, an execution subject of the embodiment of the present invention is the information analysis system, and with respect to any one of the embodiments in fig. 1 to 3, in the method for identifying a related object provided in this embodiment, S102 includes: s1021 to S1024 are described in detail as follows:

furthermore, the information analysis system comprises a task forwarding terminal and an information analysis node;

the information analysis system receives the object information sent by each information acquisition terminal, extracts the object identifier of the target object from each object information, and generates an object data packet according to the object identifier and the object information, and the method comprises the following steps:

in S1021, the task forwarding terminal receives the object information sent by the information acquisition terminal, and acquires a load parameter of each information analysis node at the time when the object information is received.

In this embodiment, the information analysis system includes a task forwarding terminal and an information analysis node. After receiving the object information sent by the information acquisition terminal, the task forwarding terminal, that is, the middleware for forwarding the object information, may store the object information in a message queue, where the message queue may be a Kafka message queue, and since the Kafka message queue is deployed in a cluster manner, data loss caused by downtime of a single node is prevented.

In this embodiment, the information analysis node needs to perform a registration operation in the node center when being online, so as to declare that the information analysis node is available, and when the information analysis node is abnormal or shutdown and other situations need to perform offline, a logout operation also needs to be performed in the node center. Therefore, through the registration and the logout operation of the information analysis nodes, the node center can determine the running state of each information analysis node, thereby realizing the dynamic adjustment of the analysis tasks.

In this embodiment, the task forwarding terminal may obtain a load parameter of each information analysis node in the information analysis system, and preferably, if the information analysis system corresponds to one node center, the task forwarding terminal may directly obtain load conditions of all information analysis nodes in the cluster through the node center.

In S1022, the task forwarding terminal imports the load parameter into a preset forwarding priority conversion algorithm, and determines a forwarding priority of each information analysis node.

In this embodiment, the task forwarding terminal may import the collected load parameters into the forwarding priority conversion algorithm, so as to calculate the forwarding priority of each information analysis node at the current time, and determine a target analysis node responding to the analysis task based on the forwarding priority.

Optionally, in this embodiment, the forwarding priority conversion algorithm may be: 1) a polling algorithm, namely, sequentially sending the requests to each information analysis node according to a polling mode; 2) a random algorithm, namely generating a random number by a preset random number generation algorithm, and sending the object information to an information analysis node corresponding to the random number; 3) according to a weight distribution algorithm, namely according to the distributed weight proportion, sending the object information with the quantity corresponding to the weight proportion to an information analysis node; 4) each object information of the Ipthash is distributed according to the hash result of the access ip, so that each information acquisition terminal is fixedly corresponding to one information analysis node; 5) and a minimum connection number distribution algorithm, and sending the new request to the server with the minimum connection number. The invention can configure the weight information for the information analysis node according to the weight rule, for example, when a certain micro service is not receiving the request, the weight of the micro service is set to be 0; and the task forwarding terminal monitors the performance of the information analysis node in real time, dynamically allocates corresponding weight to the information analysis node, if the performance of the information analysis node is better, high weight is allocated, otherwise, the task forwarding terminal manages and distributes tasks to continuously monitor the performance of the information analysis node in the process, and the weight configuration of the information analysis node is adjusted in real time.

In S1023, the task forwarding terminal selects a target parsing node from the information parsing nodes based on the forwarding priority, and sends the received object information to the target parsing node.

In this embodiment, after calculating the forwarding priority of each information analysis node, the task forwarding terminal may select one information analysis node with the highest forwarding priority as a target analysis node, and may further select any one of the information analysis nodes with the forwarding priority greater than the forwarding threshold as the target analysis node according to a preset forwarding threshold, and send the object information to the target analysis node for analysis operation.

In S1024, the target parsing node extracts an object identifier of the target object from each object information, and generates an object data packet according to the object identifier and the object information.

In this embodiment, the target analysis node extracts the object identifier of the object information through an analysis algorithm, and a specific implementation process is completely the same as the operation of S102, and specific description may refer to related explanation of S102, which is not described herein again.

In the embodiment of the invention, the object information is cached through the task forwarding terminal, and the analysis tasks are sequentially sent to each information analysis node through the task forwarding terminal, so that the dynamic adjustment of the analysis tasks can be realized, the response efficiency of the analysis tasks is improved, and the load balance of the information analysis system is improved.

Fig. 5 is a flowchart illustrating a specific implementation of the method for identifying an associated object S1023 according to a fifth embodiment of the present invention. Referring to fig. 5, with respect to the embodiment described in fig. 4, the identification method S1023 for the associated object provided in this embodiment includes: S501-S502, the details are as follows:

further, the task forwarding terminal selects a target parsing node from the information parsing nodes based on the forwarding priority, and sends the received object information to the target parsing node, including:

in S501, if the task forwarding terminal detects that all the forwarding priorities are smaller than a preset priority threshold, a capacity expansion analysis node is created in the information analysis system, and an initialization configuration operation is performed on the capacity expansion analysis node based on configuration information of the information analysis node.

In this embodiment, if the task forwarding terminal detects that the forwarding priority of each information analysis node is smaller than the preset priority threshold, it identifies that each information analysis node is in an overload or full load state, and at this time, it needs to perform capacity expansion operation on the information analysis system. In this case, the information analysis system may select one of the standby servers as a capacity expansion analysis node, or create a virtual server in the information analysis system, and identify the virtual server as a capacity expansion analysis node.

In this embodiment, the information analysis system may obtain configuration information of an information analysis node in use, and perform an initialization configuration operation on the capacity expansion analysis node based on the configuration information, so as to add the capacity expansion analysis node into the information analysis system, where of course, the information analysis system needs to configure a unique identifier for the capacity expansion analysis node.

In S502, the task forwarding terminal identifies the configured expansion analysis node as the target analysis node, and sends the received object information to the target analysis node.

In this embodiment, the task forwarding terminal may identify the capacity expansion analysis node as a target analysis node, and respond to the analysis task of the object information through the capacity expansion analysis node after capacity expansion.

In the embodiment of the invention, dynamic capacity expansion is carried out when the information analysis system is detected to need capacity expansion, and a capacity expansion analysis node is created, so that the data processing capacity of the information analysis system can be improved, and the information analysis efficiency is improved.

Fig. 6 is a flowchart illustrating a specific implementation of a method for identifying an associated object according to a sixth embodiment of the present invention. Referring to fig. 6, with respect to any one of the embodiments shown in fig. 1 to 3, the method for identifying a related object provided in this embodiment further includes: s601 to S602 are specifically described as follows:

Further, the identification system of the associated object also comprises an image storage system based on Hbase cluster;

the identification method of the associated object further comprises the following steps:

in S601, if the information analysis system detects that the information type of the object information is an image information type, the information analysis system extracts an object identification area image of the target object from the object information.

In this embodiment, the identification system of the associated object further includes an image storage system for storing image data included in the object information. Based on this, if the information analysis system detects that the information type of the object information is the image information type, the information analysis system can determine the object identifier through the analysis algorithm and extract the area image where the object identifier is located, that is, the object identification area image.

In S602, the image storage system receives the object identification area image uploaded by the information analysis system, selects an image storage node with the highest storage priority from the image storage system as a target storage node, and stores the object identification image via the target storage node.

In this embodiment, the information analysis system may send the object identification area image to an image storage system, where the image storage system is specifically an image storage system composed of a plurality of image data servers constructed based on an HBASE, that is, a hadoop database architecture, and the image storage system may dynamically extend storage nodes, for example, perform capacity expansion operation on the data storage system by adding a standby image server or a virtual storage node.

In this embodiment, after receiving the object identification area image fed back by the information analysis system, the image storage system may determine the storage priority of each image storage node in the image storage system, and select a target storage node according to the storage priority, thereby implementing load balancing of image storage.

In the embodiment of the invention, the image storage system is configured to store the image data in the object information, so that the efficiency of image storage can be improved.

It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

Fig. 7 is a block diagram illustrating a structure of an associated object recognition system according to an embodiment of the present invention, where the associated object recognition system includes: the system comprises an information acquisition terminal, an information analysis system and a data processing system, wherein the information acquisition terminal, the information analysis system and the data processing system are used for executing all steps in the embodiment corresponding to the figure 1. Please refer to fig. 1 and fig. 1 for the corresponding description of the embodiment. For convenience of explanation, only the portions related to the present embodiment are shown.

Referring to fig. 7, the identification system of the associated object includes an information acquisition terminal 71, an information analysis system 72, and a data processing system 73;

the information acquisition terminal 71 is configured to acquire object information of a plurality of different types of target objects; the object information includes location information of the target object;

the information analysis system 72 is configured to receive the object information sent by each information acquisition terminal 71, extract an object identifier of the target object from each object information, and generate an object data packet according to the object identifier and the object information;

the data processing system 73 is configured to receive the object data packet sent by the information analysis system 72, and construct a moving track of the target object corresponding to the object identifier according to a plurality of pieces of position information of the same object identifier;

the data processing system 73 is configured to establish an association relationship between the target objects of different types based on the movement trajectory; the different types of target objects with the incidence relation correspond to the same entity user.

Optionally, the object information further includes time information; the data processing system 73 is configured to establish association relationships between different types of the target objects based on the movement trajectory, and includes:

The data processing system 73 is configured to determine a movement trigger time of the movement track according to the time information corresponding to each piece of the position information in the movement track;

the data processing system 73 is configured to extract multiple target tracks with the same movement trigger time from all the movement tracks, and calculate a coincidence rate between the multiple target tracks;

the data processing system 73 is configured to select the target track with the coincidence rate greater than a preset coincidence threshold as an association track, and establish an association relationship between the target objects corresponding to the association track.

Optionally, the data processing system 73 comprises a data search server based on a distributed full-text retrieval ElasticSearch cluster;

the data search server is used for receiving a query request sent by a user terminal; the query request comprises a target time period required to be queried and a target user identifier;

the data search server is used for acquiring the incidence relation corresponding to the target user identifier and determining a plurality of target objects of different types associated with the target user identifier based on the incidence relation;

The data search server is configured to extract the object information whose acquisition time is within the target time period from all the object information uploaded by the information acquisition terminal 71, and use the object information as candidate information;

the data search server is used for extracting candidate information of any type of target object related to the target user identification from the candidate information as target information;

and the data search server is used for marking the target information on a preset map interface and generating a real-time dynamic track of the target user identifier according to the sequence of the acquisition time of the target information.

Optionally, the information analysis system 72 includes a task forwarding terminal and an information analysis node;

the information analysis system 72 is configured to receive the object information sent by each information acquisition terminal 71, extract an object identifier of the target object from each object information, and generate an object data packet according to the object identifier and the object information, and includes:

the task forwarding terminal is configured to receive the object information sent by the information acquisition terminal 71, and acquire a load parameter of each information analysis node at a time when the object information is received;

The task forwarding terminal is used for importing the load parameters into a preset forwarding priority conversion algorithm and determining the forwarding priority of each information analysis node;

the task forwarding terminal is used for selecting a target analysis node from the information analysis nodes based on the forwarding priority and sending the received object information to the target analysis node;

the target analysis node is used for extracting the object identifier of the target object from each object information and generating an object data packet according to the object identifier and the object information.

Optionally, the task forwarding terminal is configured to select a target parsing node from the information parsing nodes based on the forwarding priority, and send the received object information to the target parsing node, and includes:

the task forwarding terminal is configured to create a capacity expansion analysis node in the information analysis system if it is detected that all the forwarding priorities are smaller than a preset priority threshold, and perform an initialization configuration operation on the capacity expansion analysis node based on configuration information of the information analysis node;

and the task forwarding terminal is used for identifying the configured capacity expansion analysis node as the target analysis node and sending the received object information to the target analysis node.

The identification system of the associated object further comprises an image storage system based on Hbase cluster;

the information analysis system 72 is configured to, if it is detected that the information type of the object information is an image information type, extract an object identification area image of the target object from the object information;

the image storage system is configured to receive the object identification area image uploaded by the information analysis system 72, select an image storage node with the highest storage priority from the image storage system as a target storage node, and store the object identification image through the target storage node

Therefore, the identification system of the associated object provided by the embodiment of the present invention can also acquire different types of target objects, for example, different types of object information such as a mobile terminal, a user face, a license plate number, and the like, and establish moving tracks of different target objects, and then, through similarities of the moving tracks of different target objects, can identify whether different target objects correspond to the same entity user, and then establish an association relationship of different target objects, and when some type of object information cannot be acquired, position tracking can be continuously performed on the user by acquiring other types of object information, so that an application range of user information acquisition and efficiency of information acquisition can be improved.

Fig. 8 shows a block diagram of a system for identifying an associated object according to another embodiment of the present invention. As shown in fig. 8, the data acquisition terminal may include a camera module and a wireless access device, and acquire object information of different types of target objects, and send the object information to an information analysis system, where the information analysis system includes a task forwarding terminal and an information analysis node, the task forwarding terminal receives the object information sent by each information acquisition terminal, and sends the object information to an effective information analysis node registered in a Zookeeper center, the Zookeeper may cache the object information based on a kafka queue, the information analysis node sends a processed object data packet to a data processing system, and the data processing system includes an ES server cluster for implementing display trajectory information, a Spark server cluster for establishing an association relationship, and an hbsase server cluster for storing images.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit.

The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.

Claims (8)

1. A method for identifying a related object is applied to a system for identifying a related object, and is characterized in that the system for identifying a related object comprises: the system comprises an information acquisition terminal, an information analysis system and a data processing system;

the identification method of the associated object comprises the following steps:

the information acquisition terminal acquires object information of a plurality of different types of target objects; the object information includes location information of the target object;

the information analysis system comprises a task forwarding terminal and an information analysis node; the task forwarding terminal receives the object information sent by the information acquisition terminal and acquires load parameters of each information analysis node at the moment of receiving the object information;

The task forwarding terminal imports the load parameters into a preset forwarding priority conversion algorithm and determines the forwarding priority of each information analysis node;

the task forwarding terminal selects a target analysis node from the information analysis nodes based on the forwarding priority, and sends the received object information to the target analysis node;

the target analysis node extracts an object identifier of the target object from each object information and generates an object data packet according to the object identifier and the object information;

the data processing system receives the object data packet sent by the information analysis system, and constructs a moving track of the target object corresponding to the object identifier according to a plurality of position information of the same object identifier;

the data processing system establishes association relations among the target objects of different types based on the moving track; the different types of target objects with the incidence relation correspond to the same entity user.

2. The identification method according to claim 1, characterized in that the object information further includes time information; the data processing system establishes association relations of the target objects of different types based on the movement track, and the association relations comprise:

The data processing system determines the movement trigger time of the movement track according to the time information corresponding to each position information in the movement track;

the data processing system extracts a plurality of target tracks with the same movement trigger time from all the movement tracks and calculates the coincidence rate of the target tracks;

and the data processing system selects the target track with the coincidence rate larger than a preset coincidence threshold value as an associated track, and establishes an association relation between the target objects corresponding to the associated track.

3. The identification method according to claim 1, wherein the data processing system comprises a data search server based on a distributed full text retrieval ElasticSearch cluster;

the identification method of the associated object further comprises the following steps:

the data search server receives a query request sent by a user terminal; the query request comprises a target time period required to be queried and a target user identifier;

the data search server acquires the incidence relation corresponding to the target user identification, and determines a plurality of target objects of different types associated with the target user identification based on the incidence relation;

The data search server extracts the object information with the acquisition time in the target time period from all the object information uploaded by the information acquisition terminal as candidate information;

the data search server extracts candidate information of any type of the target object related to the target user identification from the candidate information as target information;

and the data search server marks the target information on a preset map interface and generates a real-time dynamic track of the target user identifier according to the sequence of the acquisition time of the target information.

4. The identification method according to claim 1, wherein the task forwarding terminal selects a target resolution node from the information resolution nodes based on the forwarding priority, and sends the received object information to the target resolution node, and the method includes:

if the task forwarding terminal detects that all the forwarding priorities are smaller than a preset priority threshold, creating a capacity expansion analysis node in the information analysis system, and performing initialization configuration operation on the capacity expansion analysis node based on configuration information of the information analysis node;

And the task forwarding terminal identifies the configured capacity expansion analysis node as the target analysis node and sends the received object information to the target analysis node.

5. The identification method according to any of claims 1 to 3, wherein the identification system of the associated object further comprises an image storage system based on Hbase clustering;

the identification method of the associated object further comprises the following steps:

if the information analysis system detects that the information type of the object information is the image information type, extracting an object identification area image of the target object from the object information;

and the image storage system receives the object identification area image uploaded by the information analysis system, selects an image storage node with the highest storage priority from the image storage system as a target storage node, and stores the object identification image through the target storage node.

6. An associated object recognition system, comprising: the system comprises an information acquisition terminal, an information analysis system and a data processing system;

the information acquisition terminal is used for acquiring object information of a plurality of different types of target objects; the object information includes location information of the target object;

The information analysis system comprises a task forwarding terminal and an information analysis node; the information analysis system is configured to receive the object information sent by each information acquisition terminal, extract an object identifier of the target object from each object information, and generate an object data packet according to the object identifier and the object information, and includes:

the task forwarding terminal is used for receiving the object information sent by the information acquisition terminal and acquiring the load parameters of each information analysis node at the moment when the object information is received;

the task forwarding terminal is used for importing the load parameters into a preset forwarding priority conversion algorithm and determining the forwarding priority of each information analysis node;

the task forwarding terminal is used for selecting a target analysis node from the information analysis nodes based on the forwarding priority and sending the received object information to the target analysis node;

the target analysis node is used for extracting an object identifier of the target object from each object information and generating an object data packet according to the object identifier and the object information;

The data processing system is configured to receive the object data packet sent by the information analysis system, and construct a movement trajectory of the target object corresponding to the object identifier according to a plurality of pieces of location information of the same object identifier;

the data processing system is used for establishing incidence relations among the target objects of different types based on the moving track; the different types of target objects with the incidence relation correspond to the same entity user.

7. The identification system of claim 6, wherein the object information further includes time information; the data processing system is configured to establish association relationships between the different types of target objects based on the movement trajectory, and includes:

the data processing system is used for determining the movement trigger time of the movement track according to the time information corresponding to each position information in the movement track;

the data processing system is used for extracting a plurality of target tracks with the same movement trigger time from all the movement tracks and calculating the coincidence rate of the target tracks;

and the data processing system is used for selecting the target track with the coincidence rate larger than a preset coincidence threshold value as an associated track and establishing an association relation between the target objects corresponding to the associated track.

8. The identification system of claim 6, wherein said data processing system comprises a data search server based on a distributed full text search cluster;

the data search server is used for receiving a query request sent by a user terminal; the query request comprises a target time period required to be queried and a target user identifier;

the data search server is used for acquiring the incidence relation corresponding to the target user identifier and determining a plurality of target objects of different types associated with the target user identifier based on the incidence relation;

the data search server is used for extracting the object information with the acquisition time in the target time period from all the object information uploaded by the information acquisition terminal as candidate information;

the data search server is used for extracting candidate information of any type of target object related to the target user identification from the candidate information as target information;

and the data search server is used for marking the target information on a preset map interface and generating a real-time dynamic track of the target user identifier according to the sequence of the acquisition time of the target information.

Images