A method and apparatus for determining a peer relationship

Abstract

The embodiment of the application discloses a same-traveling relationship judgment method and device, the method comprises the following steps: acquiring optional nodes and / or optional paths between a starting point and a destination of a target person; acquiring all shooting records of the target person appearing within a time threshold before and after the target person appearing, which are shot by cameras corresponding to the optional nodes and / or the optional paths; determining action information of other persons except the target person between the starting point and the destination, according to the all shooting records, the action information being about the other persons traveling between the starting point and the destination during the target person traveling between the starting point and the destination; and judging whether the other persons have the same-traveling relationship with the target person according to the action information of each of the other persons. Through the embodiment scheme, hidden same-traveling persons or hidden following persons in the period before and after the target person appearing are discovered.

Description

Technical Field

[0001] This article relates to monitoring technology, and in particular to a method and device for determining peer relationships. Background Technology

[0002] Currently, the determination of whether two people are following or traveling together is based on information such as the physical distance between them and the time they spend together. The more times two people collide in close proximity within a certain period of time, the higher the probability of them following or traveling together.

[0003] People with an anti-following mentality will deliberately distance themselves from others in space, such as by keeping a certain distance, avoiding obvious locations, or taking a detour to follow others, making it impossible to detect this hidden following or companionship relationship using conventional methods. Summary of the Invention

[0004] This application provides a method and apparatus for determining peer relationships, which can uncover hidden peers or followers.

[0005] This application provides a method for determining peer relationships, the method may include:

[0006] Obtain the optional nodes and / or optional paths between the target person's starting point and destination;

[0007] Obtain all shooting records captured by the cameras corresponding to the optional nodes and / or optional paths within a time threshold before and after the appearance of the target person;

[0008] Based on all the recorded footage, determine the movement information of other persons besides the target person between the starting point and the destination during the journey between the starting point and the destination;

[0009] Determine whether the target person is a colleague of another person based on the action information of each other person.

[0010] In an exemplary embodiment of this application, obtaining the optional nodes and / or optional paths between the starting point and destination of the target person may include:

[0011] Obtain a list of possible paths and / or a list of possible node sets between the starting point and the destination;

[0012] The optional node set list includes all the necessary areas between the starting point and the destination, and all the optional nodes contained in each necessary area; the necessary areas divide the path between the starting point and the destination into multiple paths;

[0013] The list of optional paths contains all possible paths between the starting point and the destination; each optional path includes: a first path with no alternative paths, and / or a second path with alternative paths.

[0014] In an exemplary embodiment of this application, for the first path, obtaining all shooting records captured by the camera corresponding to the optional path within a time threshold before and after the appearance of the target person may include:

[0015] Obtain all target person shooting records along the first path;

[0016] Identify the first camera corresponding to each target person's video recording;

[0017] Acquire all first-shot records of each first camera within the first time threshold before and after the corresponding target person's shooting record.

[0018] In an exemplary embodiment of this application, obtaining all recordings captured by the camera corresponding to the optional node within a time threshold before and after the appearance of the target person may include:

[0019] Determine all necessary routes between the starting point and the destination;

[0020] Determine the second camera corresponding to each optional node in each mandatory area;

[0021] When the target person passes through any selectable node in any necessary area, all second shooting records of all the second cameras in that necessary area within the second time threshold before and after the target person's passing time are obtained.

[0022] In an exemplary embodiment of this application, for the second path, when the second path includes a first sub-path and a second sub-path that are alternatives to each other, obtaining all shooting records within a time threshold before and after the appearance of the target person captured by the camera corresponding to the alternative path may include:

[0023] Obtain the first sub-path traversed by the target person, and the third camera corresponding to the first sub-path;

[0024] Obtain the second sub-path and the fourth camera corresponding to the second sub-path;

[0025] The third shooting record is obtained by acquiring all the shooting records of the third camera and the fourth camera within a third time threshold before and after the target person passes through the first path.

[0026] In an exemplary embodiment of this application, the action information may include: the total number of times other persons appear under the cameras corresponding to the optional nodes and / or optional paths;

[0027] Determining the movement information of other persons besides the target person between the starting point and the destination based on all the captured records may include:

[0028] Extract all other people captured in the entire shooting record;

[0029] Based on all the shooting records, determine the movement trajectory shooting records of each of the other persons among all the other persons;

[0030] The total number of times each other person appeared in the camera corresponding to the path information is determined based on the movement trajectory recordings of each other person.

[0031] In an exemplary embodiment of this application, after extracting all other people captured from the total number of shooting records, the method may further include:

[0032] Create a separate profile for each of the other characters, with each different character corresponding to a different profile.

[0033] The personal information, movement trajectory, and action information of each other person are recorded in the corresponding file of that other person.

[0034] In an exemplary embodiment of this application, determining the total number of times each other person appears under the camera corresponding to the path information based on the recorded movement trajectory of each other person may include:

[0035] The total number of times the other person enters the first camera, second camera, third camera, and / or fourth camera is counted, and this sum is taken as the total number of appearances.

[0036] Wherein, when the images captured by the first camera, the second camera, the third camera and / or the fourth camera have repeated scenes, the repeated scene is counted only once in the sum of the number of times.

[0037] In an exemplary embodiment of this application, determining whether the target person is a fellow traveler based on the action information of each other person may include:

[0038] Determine whether the total number of occurrences for each other character exceeds a preset threshold.

[0039] When the total number of occurrences exceeds the threshold, it is determined that the other person and the target person are colleagues.

[0040] When the total number of occurrences is less than or equal to the number of occurrences threshold, it is determined that the other person is not in the same industry as the target person.

[0041] This application also provides a peer relationship determination device, which may include a processor and a computer-readable storage medium. The computer-readable storage medium stores instructions that, when executed by the processor, implement any of the peer relationship determination methods described above.

[0042] Compared with related technologies, the embodiments of this application may include: acquiring optional nodes and / or optional paths between the starting point and destination of the target person; acquiring all shooting records captured by cameras corresponding to the optional nodes and / or optional paths within a time threshold before and after the appearance of the target person; determining, based on the all shooting records, the action information of other persons besides the target person between the starting point and the destination during the target person's journey; and determining, based on the action information of each other person, whether the other person and the target person are traveling together. This embodiment, by acquiring all shooting records captured by cameras corresponding to optional nodes and / or optional paths within a time threshold before and after the appearance of the target person, and determining the action information of other persons during the target person's journey between the starting point and the destination based on the all shooting records, can obtain the action information of other persons within the time period before and after the appearance of the target person, and determine the traveling relationship based on this action information, thereby realizing the discovery of hidden traveling companions or hidden followers within the time period before and after the appearance of the target person.

[0043] Other features and advantages of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the application. Other advantages of this application can be realized and obtained by means of the solutions described in the description and the accompanying drawings. Attached Figure Description

[0044] The accompanying drawings are used to provide an understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0045] Figure 1 This is a flowchart of the parallel relationship determination method according to an embodiment of this application;

[0046] Figure 2 This is a schematic diagram illustrating the optional paths and nodes between the origin and destination in an embodiment of this application.

[0047] Figure 3 This is a schematic diagram of the necessary areas and optional nodes in an embodiment of this application;

[0048] Figure 4 This is a schematic diagram of the optional path set in an embodiment of this application;

[0049] Figure 5 This is a schematic diagram of a method for determining parallel relationships according to an embodiment of this application;

[0050] Figure 6 This is a diagram illustrating the composition of the parallel relationship determination device according to an embodiment of this application. Detailed Implementation

[0051] This application describes several embodiments, but these descriptions are exemplary and not restrictive, and it will be apparent to those skilled in the art that many more embodiments and implementations are possible within the scope of the embodiments described herein. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment may be used in combination with, or may replace, any feature or element of any other embodiment.

[0052] This application includes and contemplates combinations of features and elements known to those skilled in the art. The embodiments, features, and elements disclosed in this application may also be combined with any conventional features or elements to form a unique inventive scheme as defined by the claims. Any feature or element of any embodiment may also be combined with features or elements from other inventive schemes to form another unique inventive scheme as defined by the claims. Therefore, it should be understood that any feature shown and / or discussed in this application may be implemented individually or in any suitable combination. Therefore, the embodiments are not limited except by the limitations imposed by the appended claims and their equivalents. Furthermore, various modifications and changes may be made within the scope of the appended claims.

[0053] Furthermore, in describing representative embodiments, the specification may have presented methods and / or processes as a specific sequence of steps. However, the method or process should not be limited to the specific order of steps described herein, to the extent that it does not depend on such a specific order. As will be understood by those skilled in the art, other sequences of steps are also possible. Therefore, the specific order of steps set forth in the specification should not be construed as a limitation of the claims. Moreover, the claims concerning the method and / or process should not be limited to the steps performed in the written order, and those skilled in the art will readily understand that these orders can be varied and still remain within the spirit and scope of the embodiments of this application.

[0054] This application provides a method for determining peer relationships, such as... Figure 1 As shown, the method may include steps S101-S104:

[0055] S101. Obtain the optional nodes and / or optional paths between the target person's starting point and destination;

[0056] S102. Obtain all shooting records captured by the cameras corresponding to the optional nodes and / or optional paths within the time threshold before and after the appearance of the target person;

[0057] S103. Based on all the recorded footage, determine the movement information of other persons besides the target person between the starting point and the destination during the movement of the target person between the starting point and the destination.

[0058] S104. Determine whether the target person is a fellow traveler based on the action information of each other person.

[0059] In an exemplary embodiment of this application, two people or objects (such as drones), or one person and one object, each departing separately, separated by a certain distance and time, all starting from the same point and arriving at the same destination, can be considered as traveling together or following. People with anti-following awareness will deliberately space themselves apart, such as by maintaining a certain distance, avoiding obvious locations, or taking a detour to track, making it impossible for conventional methods to detect such following or traveling relationships. To overcome this anti-following, embodiments of this application propose a method for determining traveling relationships, which can uncover following or traveling relationships where there is a detour or a deliberate separation.

[0060] In an exemplary embodiment of this application, the solution of this application determines the optional nodes and optional paths traversed by the target person, and on the basis of the traditional method, adds the recording of the peer and trailing relationships of the optional nodes and optional paths, thereby discovering hidden following and peer people.

[0061] In an exemplary embodiment of this application, the person may include a human or an object, such as a drone or driverless car with following capabilities. In this embodiment, a human is used as an example to illustrate the implementation scheme of this application.

[0062] In an exemplary embodiment of this application, a target person A is determined to have a starting point (Source) and a destination (Destination). The target person A travels from the starting point (Source) to the destination (Destination), and may be accompanied or followed by one or more people along the way. The starting point (Source) and destination (Destination) selected by the target person may be different from those of the accompanying or following person. That is, the accompanying or following person may enter midway, accompany or follow for a period of time, and then withdraw midway.

[0063] In exemplary embodiments of this application, as Figure 2 As shown, target A passed through two key areas, Area 1 and Area 2, with path 1 as the optional path. The companion and follower also passed through Area 1 and Area 2, but through different entrances and exits, taking path 2 as the optional path. The companion and follower started traveling together and following each other at tracking point 1, and stopped at tracking point 2. The paths of the companion and follower are: PathX, Path2, PathY.

[0064] Traditional methods only calculate snapshots taken under the same lens, so the same-line and tail-line relationships at the entrances and exits of the two key areas Area1 and Area2 will be lost; similarly, the same-line and tail-line relationships on paths path1 and path2 will be lost.

[0065] In an exemplary embodiment of this application, by determining the optional nodes and optional paths traversed by the target person A, and on the basis of the traditional method, the method adds recording of the peer and trailing relationships of the optional nodes and optional paths, without missing the peer and trailing relationships of the key areas Area1, Area2, and Path2, thereby discovering hidden followers and peers.

[0066] In an exemplary embodiment of this application, obtaining the optional nodes and / or optional paths between the starting point and destination of the target person may include:

[0067] Obtain a list of possible paths and / or a list of possible node sets between the starting point and the destination;

[0068] The optional node set list includes all the necessary areas between the starting point and the destination, and all the optional nodes contained in each necessary area; the necessary areas divide the path between the starting point and the destination into multiple paths;

[0069] The list of optional paths contains all possible paths between the starting point and the destination; each optional path may include: a first path with no alternative paths and / or a second path with alternative paths.

[0070] In an exemplary embodiment of this application, the path information of the target person A from the starting point Source to the destination Destination can be analyzed to determine the special areas that must be passed through (must-pass areas), thereby determining the optional nodes NodeSet that must be passed through in each must-pass area, and finally determining the set of all optional nodes NodeSet (optional node set), and can further generate a list of the set (i.e., the list of optional node sets) NodeSetList.

[0071] In the exemplary embodiments of this application, an optional node refers to a specific node that can be passed through to enter a particular region. All optional nodes within a mandatory region can constitute a set NodeSet. A set of optional nodes can contain one or more optional nodes. Different particular regions (mandatory regions) can correspond to different sets of optional nodes.

[0072] In exemplary embodiments of this application, as Figure 3 As shown, to enter Area1, one can pass through Entrance 1, Entrance 2, and Entrance 3. These three optional nodes, Entrance 1, Entrance 2, and Entrance 3, constitute an optional node set NodeSet1.

[0073] In an exemplary embodiment of this application, the path of target person A from the starting point Source to the destination Destination can be analyzed to determine the optional paths PathSet along the way, and finally determine the set of all optional paths, i.e., the optional path set. Based on the optional path set, an optional path list PathSetList can be generated.

[0074] In an exemplary embodiment of this application, an optional path refers to a route from one location to another that can involve multiple paths. These paths may be mandatory paths with no alternative paths (i.e., the first path) or paths with alternative paths (i.e., the second path). All such paths (including the first and second paths) can be called optional paths, and all optional paths between the starting point and the destination can constitute an optional path set (PathSet). There can be one or more optional paths. Different starting points and destinations can correspond to different sets of optional paths (i.e., optional path sets).

[0075] In exemplary embodiments of this application, as Figure 4As shown, the path from location Place1 to location Place2 can be taken through either path 1 or path 2. Therefore, path 1 and path 2 constitute the optional path set PathSet1.

[0076] In an exemplary embodiment of this application, the person traveling with or following the target person A cannot be too far away from the target person A in order to avoid losing them; any path within the visible distance can be an optional path. For example, the target person A is on the left side of the road, and the person traveling with or following the target person A is on the right side of the road.

[0077] In an exemplary embodiment of this application, optional nodes and optional paths to reach the destination are determined, thereby discovering optional nodes and optional paths that cannot be bypassed by following or traveling together, thereby further calculating the following and traveling together relationships.

[0078] In an exemplary embodiment of this application, when the optional path is the first path, obtaining all shooting records captured by the camera corresponding to the optional path within a time threshold before and after the appearance of the target person may include:

[0079] Obtain all first shooting records of all first cameras that captured the target person on the first path within a first time threshold before and after capturing the target person.

[0080] In an exemplary embodiment of this application, obtaining all first shooting records from all first cameras that captured the target person along the first path within a first pre- and post-capture time threshold may include:

[0081] Obtain all target person shooting records along the first path;

[0082] Identify the first camera corresponding to each target person's video recording;

[0083] Acquire all first-shot records of each first camera within the first time threshold before and after the corresponding target person's shooting record.

[0084] In an exemplary embodiment of this application, after determining all first shooting records on the first path, a first shooting record list can be generated based on all the first shooting records. The first shooting record list may contain different first paths, and all shooting records of the first camera ListCam1 corresponding to the different first paths within a first time threshold before and after the time when the target person passes by.

[0085] In an exemplary embodiment of this application, cameras may be installed from the starting point (Source) to the destination (Destination) for real-time shooting or snapshots to obtain shooting records.

[0086] In exemplary embodiments of this application, as Figure 2 , Figure 5 As shown, based on the path Path0 (which is the first path with no alternative paths) from the starting point Source to the destination Destination of the target person A, the shooting records of the target person A on the path Path0 are filtered out.

[0087] In an exemplary embodiment of this application, according to the above scheme, all shooting records of each shooting record can be filtered out from the first camera ListCam1, and all shooting records within the time range of threshold A (first before and after time threshold) before and after the shooting time can be generated into a first shooting record list RecList1 (i.e., the first shooting record list). Furthermore, the shooting records can include the times when the target person A enters the first camera ListCam1 and leaves the first camera ListCam1, thus expanding the search time range to the time of entry into the first camera - threshold A (first before and after time threshold) and the time of leaving the first camera + threshold A, allowing for more precise filtering of the desired shooting records.

[0088] In an exemplary embodiment of this application, based on the path of target person A from the starting point Source to the destination Destination, the optional nodes and optional paths to reach the destination area can be determined, and the corresponding NodeSetList and PathSetList can be further generated.

[0089] In an exemplary embodiment of this application, obtaining all recordings captured by the camera corresponding to the optional node within a time threshold before and after the appearance of the target person may include:

[0090] When the target person passes through any selectable node in a must-pass area, obtain all the second shooting records of the second camera ListCam2 corresponding to all selectable nodes in the must-pass area within the second time threshold before and after the target person passes through.

[0091] In an exemplary embodiment of this application, obtaining all second shooting records of the second camera ListCam2 corresponding to all optional nodes in a necessary passage area when the target person passes through any optional node in the necessary passage area within a second time threshold before and after the time the target person passes through may include:

[0092] Determine all necessary routes between the starting point and the destination;

[0093] Determine the second camera ListCam2 corresponding to each optional node contained in each mandatory area;

[0094] When the target person passes through any selectable node in any necessary area, all second shooting records of the second cameras ListCam2 within the second time threshold before and after the target person's passing time are obtained.

[0095] In an exemplary embodiment of this application, after obtaining all second shooting records of one or more necessary passage areas, a second shooting record list can be generated; the second shooting record list may contain different necessary passage areas, all optional nodes contained in different necessary passage areas, and all shooting records of the second camera ListCam2 corresponding to each optional node within the second time threshold before and after the time when the target person passes through.

[0096] In exemplary embodiments of this application, as Figure 5 As shown, according to the above scheme, the shooting records of the second camera ListCam2 corresponding to NodeSetList within the time period before and after the time when target person A passes through NodeSetList (threshold B, the second time threshold) can be filtered out, generating a second shooting record list RecList2 (i.e., the second shooting record list). Furthermore, the shooting records can include the times when target person A enters and leaves the second camera ListCam2, thus expanding the search time range to the time of entering the second camera ListCam2 minus threshold B (the second time threshold) and the time of leaving the second camera ListCam2 plus threshold B, allowing for more precise filtering of the desired shooting records.

[0097] In an exemplary embodiment of this application, when the optional path is the second path, and when the second path includes a first sub-path and a second sub-path that are mutually alternatives, obtaining all shooting records captured by the camera corresponding to the optional path within a time threshold before and after the appearance of the target person may include:

[0098] Obtain all third-shot records of the third camera ListCam3 corresponding to the first sub-path and the fourth camera ListCam4 corresponding to the second sub-path within the third time threshold before and after the target person passes through the first sub-path.

[0099] In an exemplary embodiment of this application, obtaining all third-shot records from the third camera ListCam3 corresponding to the first sub-path and the fourth camera ListCam4 corresponding to the second sub-path within a third time threshold before and after the target person passes through the first sub-path may include:

[0100] Obtain the first sub-path traversed by the target person, and the third camera ListCam3 corresponding to the first sub-path;

[0101] Obtain the second sub-path and the fourth camera ListCam4 corresponding to the second sub-path;

[0102] The third shooting record is obtained by acquiring all the shooting records of the third camera ListCam3 and the fourth camera ListCam4 within the third time threshold before and after the target person passes through the first path.

[0103] In an exemplary embodiment of this application, after determining all the third shooting records corresponding to the second path, a list of third shooting records is generated; the list of third shooting records may contain different second paths, all the sub-paths contained in different second paths, and all shooting records of the cameras (e.g., ListCam3 and ListCam4) corresponding to each sub-path within the second time threshold before and after the moment when the target person passes through any sub-path.

[0104] In exemplary embodiments of this application, as Figure 5 As shown, the above method can filter out the capture records of the corresponding cameras (e.g., ListCam3 and ListCam4) within the time period before and after the target person A passes through PathSetList, within a threshold C (the third time threshold), generating a capture record list RecList3 (i.e., the third capture record list). Furthermore, since the capture records include the times when the target person A enters and leaves the camera, the search area is expanded to the time of entering the camera – threshold C and the time of leaving the camera + threshold C, allowing for more precise filtering of the desired capture records.

[0105] In an exemplary embodiment of this application, after obtaining the shooting records of various cameras (such as the first camera ListCam1, the second camera ListCam2, the third camera ListCam3, the fourth camera ListCam4, etc.) through the above scheme, RecList, RecList2, and RecList3 can be merged to obtain all shooting records of the optional path and / or optional nodes between the origin and the destination. All shooting records can generate a fourth shooting record list RecList4.

[0106] In an exemplary embodiment of this application, the action information may include: the total number of times other person B appears under the camera corresponding to the optional node and / or optional path;

[0107] Determining the movement information of other persons besides the target person between the starting point and the destination based on all the captured records may include:

[0108] Extract all other people B captured in the entire shooting record;

[0109] Based on all the shooting records, determine the movement trajectory shooting records of each of the other B persons among all the other B persons;

[0110] The total number of times each other person B appears under the camera corresponding to the optional node and / or optional path is determined based on the action trajectory recording of each other person B.

[0111] In an exemplary embodiment of this application, determining the total number of times each other person appears under the camera corresponding to the path information based on the recorded movement trajectory of each other person may include:

[0112] The total number of times the other person enters the first camera, second camera, and / or third camera is counted, and this sum is taken as the total number of appearances.

[0113] Wherein, when the images captured by the first camera, the second camera and / or the third camera have repeated scenes, the repeated scene is counted only once in the sum of the number of times.

[0114] In an exemplary embodiment of this application, if the camera corresponding to a certain shooting record of another person B is ListCam1 and is not in ListCam2, ListCam3, or ListCam4, then for a single activity under that camera ListCam1, the number of times to determine whether to follow or accompany can be incremented by 1 (specifically, it can be a preset counter incremented by 1).

[0115] In an exemplary embodiment of this application, if another person B lingers under the same camera (this can be determined through continuous shooting or by processing multiple consecutive captured images, such as image stitching and prediction, to predict the trajectory of the other person B and analyze the lingering behavior), it may be captured or photographed multiple times. For such repetitive scenarios, the lingering behavior can be counted only once. If another person B leaves the camera and then returns, the count continues to accumulate. Furthermore, considering that the target person A may linger, causing other person B to be photographed multiple times and the corresponding lingering behavior to be counted multiple times, the lingering behavior during the lingering period is counted only once; that is, this also falls under a repetitive scenario and is counted only once.

[0116] In an exemplary embodiment of this application, if a camera corresponding to a certain shooting record of another person B is in ListCam2, it is determined that the other person B has a single activity under the second camera ListCam2, and the number of times they travel together or follow is incremented by 1; the counting rules for repeated scenes are the same as above.

[0117] In an exemplary embodiment of this application, if a camera corresponding to a certain shooting record of another person B is in ListCam3 or ListCam4, then under the same path, the number of times a person travels together or follows another person is counted once for each single trip. That is, there may be multiple cameras under the same path, but the capture record under that path is only counted once for each person traveling together. If the person leaves the path and returns, the count continues to accumulate. In addition, considering that there are multiple capture situations under the same path, it can also be treated as a repeated scenario. However, this repeated scenario is special. It is possible that the target person A moves back and forth on the first sub-path in order to get rid of other people or for other reasons, and other people B also make corresponding back and forth movements. In this case, the count can be calculated according to the following rules: a) The number of times the target person A shoots under that path is used as the number of times a person travels together or follows another person (this can be a preferred solution, because if multiple shooting records of other people B can still be captured under this situation, it can be determined that other people B are indeed following the target person B); b) The number of times other people B shoots under that path is used as the number of times a person travels together or follows another person.

[0118] In an exemplary embodiment of this application, for all the shooting records in the aforementioned scheme (the fourth shooting record list RecList4), the movement trajectories can be categorized for different people (mainly other people B). Specifically, one file can be created for each person, and the records in RecList4 can be categorized and classified under different personnel files.

[0119] In an exemplary embodiment of this application, after extracting all other people B captured from the total number of shooting records, the method may further include:

[0120] Create a separate profile for each other character B, with each different other character B corresponding to a different profile;

[0121] Record the character information, movement trajectory, and action information of each other character B into the corresponding file of that other character.

[0122] In an exemplary embodiment of this application, determining whether the target person is a fellow traveler based on the action information of each other person may include:

[0123] Determine whether the total number of occurrences for each other character exceeds a preset threshold.

[0124] When the total number of occurrences exceeds the threshold, it is determined that the other person and the target person are colleagues.

[0125] When the total number of occurrences is less than or equal to the number of occurrences threshold, it is determined that the other person is not in the same industry as the target person.

[0126] In an exemplary embodiment of this application, since there are many pedestrians on the road, a large number of pedestrians occasionally collide in space. In order to reduce collision noise, it is assumed that if the number of times they walk together or follow each other exceeds (greater than) a threshold D (i.e., the number threshold), then they are considered to have a walking or following relationship; if it is less than or equal to the threshold D, then they are considered to be occasional collisions and not a walking or following relationship.

[0127] In an exemplary embodiment of this application, the embodiments of this application determine the optional nodes and optional paths that target person A passes through. Based on the traditional method, the embodiments add the recording of the companion and tail relationships of optional nodes and optional paths, which can discover the following and companion relationships of those who deliberately keep a certain distance or take a detour to follow, thereby realizing the discovery of hidden followers and companions.

[0128] In exemplary embodiments of this application, the advantages of the present application's solutions compared with conventional solutions can be shown in Table 1:

[0129] Table 1

[0130]

[0131] This application also provides a parallel relationship determination device 1, such as... Figure 6 As shown, it may include a processor 11 and a computer-readable storage medium 12, wherein the computer-readable storage medium 12 stores instructions that, when executed by the processor 11, implement the peer relationship determination method described in any of the above-mentioned embodiments.

[0132] In the exemplary embodiments of this application, any of the embodiments in the above method embodiments can be applied to the device embodiments, and will not be described in detail here.

[0133] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

Claims

1. A method for determining peer relationships, characterized in that, The method includes: Obtain the optional nodes and / or optional paths between the target person's starting point and destination; Obtain all shooting records captured by the cameras corresponding to the optional nodes and / or optional paths within a time threshold before and after the appearance of the target person; Based on all the recorded footage, determine the movement information of other persons besides the target person between the starting point and the destination during the journey between the starting point and the destination; Determine whether the target person is a colleague of another person based on the action information of each other person. The step of obtaining the optional nodes and / or optional paths between the target person's starting point and destination includes: Obtain a list of possible paths and / or a list of possible node sets between the starting point and the destination; The optional node set list includes all the necessary areas between the starting point and the destination, and all the optional nodes contained in each necessary area; the necessary areas divide the path between the starting point and the destination into multiple paths; The list of optional routes contains all possible routes between the starting point and the destination; the optional routes include: a first route with no alternative routes and / or a second route with alternative routes; The action information includes: the total number of times other characters appear under the cameras corresponding to the optional nodes and / or optional paths; The step of determining, based on all the captured footage, the movement information of other persons besides the target person during their walk on the first path between the starting point and the destination includes: Extract all other people captured in the entire shooting record; Based on all the shooting records, determine the movement trajectory shooting records of each of the other persons among all the other persons; The total number of times each other person appears under the camera corresponding to the optional node and / or optional path is determined based on the action trajectory recordings of each other person.

2. The method for determining peer relationships according to claim 1, characterized in that, For the first path, obtain all shooting records captured by the camera corresponding to the optional path within a time threshold before and after the appearance of the target person, including: Obtain all target person shooting records along the first path; Identify the first camera corresponding to each target person's video recording; Acquire all first-shot records of each first camera within the first time threshold before and after the corresponding target person's shooting record.

3. The method for determining peer relationships according to claim 1, characterized in that, Obtain all captured images from the camera corresponding to the optional node within a time threshold before and after the appearance of the target person, including: Determine all necessary routes between the starting point and the destination; Determine the second camera corresponding to each optional node in each mandatory area; When the target person passes through any selectable node in any necessary area, all second shooting records of all the second cameras in that necessary area within the second time threshold before and after the target person's passing time are obtained.

4. The method for determining peer relationships according to claim 1, characterized in that, For the second path, when the second path includes a first sub-path and a second sub-path that are mutually alternatives, all shooting records captured by the camera corresponding to the alternative path within a time threshold before and after the appearance of the target person are obtained, including: Obtain the first sub-path traversed by the target person, and the third camera corresponding to the first sub-path; Obtain the second sub-path and the fourth camera corresponding to the second sub-path; The third and fourth cameras acquire all the shooting records within a third time threshold before and after the target person passes through the first path as the third shooting record.

5. The method for determining peer relationships according to claim 1, characterized in that, After extracting all other people captured from the entire recording, the method further includes: Create a separate profile for each of the other characters, with each different character corresponding to a different profile. The personal information, movement trajectory, and action information of each other person are recorded in the corresponding file of that other person.

6. The method for determining peer relationships according to claim 1, characterized in that, The step of determining the total number of times each other person appears in the camera corresponding to the path information based on the recorded movement trajectory of each other person includes: The total number of times the other person enters the first camera, second camera, third camera, and / or fourth camera is counted, and this sum is taken as the total number of appearances. Wherein, when the images captured by the first camera, the second camera, the third camera and / or the fourth camera have repeated scenes, the repeated scene is counted only once in the sum of the number of times.

7. The method for determining peer relationships according to claim 1, characterized in that, The step of determining whether a target person is a fellow traveler based on the action information of each other person includes: Determine whether the total number of occurrences for each other character exceeds a preset threshold. When the total number of occurrences exceeds the threshold, it is determined that the other person and the target person are colleagues. When the total number of occurrences is less than or equal to the number of occurrences threshold, it is determined that the other person is not in the same industry as the target person.

8. A device for determining peer relationships, characterized in that, The device includes a processor and a computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed by the processor, implement the peer relationship determination method as described in any one of claims 1-7.

Images