Intelligent multi-target tracking method and system

Abstract

The application is suitable for the technical field of multi-target tracking, and provides an intelligent multi-target tracking method and system. The application determines multiple tracking targets, remotely and uniformly monitors the multiple tracking targets, mutually cooperatively monitors the multiple corresponding tracking targets, analyzes deviation, selects suspicious targets, and calculates dynamic suspicious indexes. When the indexes exceed a preset threshold, multiple intervention targets are selected. The suspicious targets are monitored by multiple intervention, and it is identified and judged whether there is abnormal deviation. The suspicious targets deviating from the multiple intervention targets are selected by combining remote uniform monitoring with mutual cooperative monitoring, the dynamic suspicious indexes are calculated, when the indexes exceed the threshold, multiple intervention targets are selected for multiple intervention monitoring, and it is identified and judged whether there is abnormal deviation. The unmanned aerial vehicle targets concentrated by the unmanned aerial vehicles are accurately identified and judged whether the deviation is abnormal, and a basis is provided for subsequent cooperative flight control and stability of task execution of the unmanned aerial vehicle group.

Description

An intelligent multi-target tracking method and system Technical Field

[0001] This invention belongs to the field of multi-target tracking technology, and in particular relates to an intelligent multi-target tracking method and system. Background Technology

[0002] Multi-target tracking is one of the core research directions in the fields of computer vision and intelligent perception. It is used to simultaneously detect, identify and continuously track multiple targets in video sequences or continuous frames. It mainly achieves the preservation of target identity and estimation of dynamic state by analyzing the target's spatial position, motion trajectory and appearance features. It has wide applications in fields such as intelligent monitoring, autonomous driving, drone perception, traffic flow analysis and behavior recognition.

[0003] For multi-target tracking applications of drone swarms, drones may deviate in attitude and position due to environmental factors such as wind speed and direction during flight. Some deviations are often normal dynamic adjustments. However, existing technologies can only monitor drone swarms remotely, making it difficult to accurately identify and determine whether deviations are abnormal, which in turn affects the stability of subsequent coordinated flight control and mission execution of the drone swarm. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent multi-target tracking method and system, which aims to solve the technical problems existing in the prior art mentioned in the background.

[0005] The embodiments of the present invention are implemented as follows:

[0006] An intelligent multi-target tracking method, the method specifically includes the following steps:

[0007] Multiple tracking targets are identified, and the multiple tracking targets are remotely monitored in a unified manner to obtain unified monitoring data;

[0008] Perform collaborative stability analysis on multiple tracking targets, match multiple corresponding collaborative targets, perform mutual collaborative monitoring on multiple corresponding tracking targets, and obtain multiple collaborative monitoring data;

[0009] Deviation analysis is performed on the unified monitoring data and multiple collaborative monitoring data to select suspicious targets and calculate a dynamic suspicion index;

[0010] When the dynamic suspicion index exceeds the preset index threshold, select multiple intervention targets;

[0011] By using multiple intervention targets, the suspected target is monitored from multiple angles to obtain multiple intervention monitoring data, and to identify and determine whether there is any abnormal deviation.

[0012] As a further limitation of the technical solution of this embodiment of the invention, the step of determining multiple tracking targets and performing remote unified monitoring on the multiple tracking targets to obtain unified monitoring data specifically includes the following steps:

[0013] Receive multi-target tracking tasks;

[0014] The multi-target tracking task is performed to identify multiple tracking targets;

[0015] Based on the multi-target tracking task, plan the remote monitoring parameters;

[0016] According to the remote monitoring parameters, multiple tracking targets are remotely monitored in a unified manner to obtain unified monitoring data.

[0017] As a further limitation of the technical solution of this embodiment of the invention, the step of performing cooperative stability analysis on multiple tracking targets, matching multiple corresponding cooperative targets, performing mutual cooperative monitoring on multiple corresponding tracking targets, and obtaining multiple cooperative monitoring data specifically includes the following steps:

[0018] Based on the multi-target tracking task, plan dynamic routes for multiple tracking targets;

[0019] Based on multiple dynamic routes, a cooperative stability analysis is performed on multiple tracking targets, and multiple corresponding cooperative targets are matched from the multiple tracking targets;

[0020] Create multiple collaborative monitoring subtasks corresponding to the aforementioned collaborative objectives;

[0021] Based on the multiple collaborative monitoring sub-tasks, multiple corresponding tracking targets are monitored collaboratively to obtain multiple collaborative monitoring data.

[0022] As a further limitation of the technical solution of this invention embodiment, the deviation analysis of the unified monitoring data and multiple collaborative monitoring data, the selection of suspicious targets, and the calculation of the dynamic suspiciousness index specifically include the following steps:

[0023] Based on the unified monitoring data, attitude deviations of multiple tracking targets are identified, and multiple attitude deviation data are recorded.

[0024] Based on multiple collaborative monitoring data, distance deviation is identified for multiple corresponding tracking targets, and multiple distance deviation data are recorded.

[0025] Based on the multiple attitude deviation data and the multiple distance deviation data, a suspicious target is selected from the multiple tracking targets;

[0026] Based on multiple attitude deviation data and multiple distance deviation data, the dynamic suspicion index of the suspicious target is calculated.

[0027] As a further limitation of the technical solution of this embodiment of the invention, the calculation formula of the dynamic suspicion index is as follows:

[0028] ;

[0029] ;

[0030] in, This is a dynamic suspicion index; For the time interval of deviation; for Attitude deviation value at any given time; for Distance deviation at any given time; This is the preset correction factor.

[0031] As a further limitation of the technical solution of this invention, the step of selecting multiple intervention targets when the dynamic suspicion index exceeds a preset index threshold specifically includes the following steps:

[0032] The dynamic suspicion index is compared with a preset index threshold;

[0033] When the dynamic suspicion index is greater than the index threshold, the index difference between the dynamic suspicion index and the index threshold is calculated;

[0034] The number of interventions is matched based on the index difference;

[0035] From the multiple tracking targets, select the number of targets to be intervened.

[0036] As a further limitation of the technical solution of this embodiment of the invention, the step of performing multi-point intervention monitoring on the suspected target through multiple intervention targets, obtaining multiple intervention monitoring data, and identifying and determining whether there is any abnormal deviation specifically includes the following steps:

[0037] Create multiple intervention monitoring subtasks for the aforementioned intervention targets;

[0038] Based on the multiple intervention monitoring sub-tasks, multi-point intervention monitoring and control are performed on the multiple intervention targets targeting the suspected targets.

[0039] Acquire multiple intervention monitoring data;

[0040] The multiple intervention monitoring data are comprehensively identified to determine whether there are any abnormal deviations.

[0041] An intelligent multi-target tracking system includes a remote unified monitoring module, a mutually collaborative monitoring module, a suspicion index calculation module, an intervention target selection module, and a multi-position intervention monitoring module, wherein:

[0042] The remote unified monitoring module is used to identify multiple tracking targets and perform remote unified monitoring on the multiple tracking targets to obtain unified monitoring data;

[0043] The mutual collaborative monitoring module is used to perform collaborative stability analysis on multiple tracking targets, match multiple corresponding collaborative targets, perform mutual collaborative monitoring on multiple corresponding tracking targets, and obtain multiple collaborative monitoring data.

[0044] The suspicion index calculation module is used to perform deviation analysis on the unified monitoring data and multiple collaborative monitoring data, select suspicious targets, and calculate the dynamic suspicion index;

[0045] The intervention target selection module is used to select multiple intervention targets when the dynamic suspicion index exceeds a preset index threshold;

[0046] The multi-position intervention monitoring module is used to perform multi-position intervention monitoring on the suspected target through multiple intervention targets, acquire multiple intervention monitoring data, and identify and determine whether there is any abnormal deviation.

[0047] As a further limitation of the technical solution of this embodiment of the invention, the mutually coordinated monitoring module specifically includes:

[0048] A dynamic route planning unit is used to plan dynamic routes for multiple tracking targets based on the multi-target tracking task.

[0049] The cooperative stability analysis unit is used to perform cooperative stability analysis on multiple tracking targets based on multiple dynamic routes, and to match multiple corresponding cooperative targets from the multiple tracking targets;

[0050] The subtask creation unit is used to create multiple collaborative monitoring subtasks corresponding to the collaborative objectives.

[0051] The mutual collaborative monitoring unit is used to perform mutual collaborative monitoring on multiple corresponding tracking targets according to multiple collaborative monitoring sub-tasks, and to obtain multiple collaborative monitoring data.

[0052] As a further limitation of the technical solution of this embodiment of the invention, the suspiciousness index calculation module specifically includes:

[0053] The attitude deviation recognition unit is used to identify attitude deviations of multiple tracking targets based on the unified monitoring data and record multiple attitude deviation data.

[0054] The distance deviation identification unit is used to identify distance deviations from multiple corresponding tracking targets based on multiple collaborative monitoring data, and to record multiple distance deviation data.

[0055] A suspicious target selection unit is configured to select a suspicious target from a plurality of tracked targets based on a plurality of attitude deviation data and a plurality of distance deviation data;

[0056] The dynamic suspicion index calculation unit is used to calculate the dynamic suspicion index of the suspicious target based on multiple attitude deviation data and multiple distance deviation data.

[0057] Compared with the prior art, the beneficial effects of the present invention are:

[0058] This invention identifies multiple tracking targets and performs remote unified monitoring of these targets; it also performs collaborative monitoring of these corresponding targets; it conducts deviation analysis, selects suspicious targets, and calculates a dynamic suspicion index; when the index exceeds a preset threshold, it selects multiple intervention targets; and it performs multi-position intervention monitoring of the suspicious targets to identify and determine whether abnormal deviations exist. By combining remote unified monitoring with collaborative monitoring, it selects suspicious targets that deviate, calculates a dynamic suspicion index, and selects multiple intervention targets for multi-position intervention monitoring when the index exceeds a threshold, thus identifying and determining whether abnormal deviations exist. This allows for accurate identification and determination of whether deviations from concentrated UAV targets are abnormal, providing a foundation for the subsequent coordinated flight control and mission execution stability of UAV swarms. Attached Figure Description

[0059] Figure 1 shows a flowchart of the intelligent multi-target tracking method provided in an embodiment of the present invention;

[0060] Figure 2 shows the application architecture diagram of the intelligent multi-target tracking system provided in an embodiment of the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be 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 and not intended to limit the invention.

[0062] It is understandable that in the application of multi-target tracking of drone swarms, the drones in the swarm may deviate in attitude and position due to the influence of environmental factors such as wind speed and wind direction during the flight of the drones. Some deviations are often normal dynamic adjustment behaviors. However, the existing technology can usually only monitor the drone swarm remotely, and it is difficult to accurately identify and judge whether the deviation is abnormal, which will affect the stability of the subsequent coordinated flight control and mission execution of the drone swarm.

[0063] To address the aforementioned issues, this invention discloses an intelligent multi-target tracking method and system. This method involves identifying multiple tracking targets and performing remote unified monitoring to obtain unified monitoring data; conducting collaborative stability analysis on the multiple tracking targets to match multiple corresponding collaborative targets; performing mutual collaborative monitoring on the corresponding tracking targets to obtain multiple collaborative monitoring data; performing deviation analysis on the unified monitoring data and multiple collaborative monitoring data to select suspicious targets and calculate a dynamic suspicion index; selecting multiple intervention targets when the dynamic suspicion index exceeds a preset threshold; and using these intervention targets to perform multi-position intervention monitoring on the suspicious targets, obtaining multiple intervention monitoring data, and identifying and determining whether abnormal deviations exist. This method combines remote unified monitoring with mutual collaborative monitoring to select suspicious targets with deviations, calculate a dynamic suspicion index, and select multiple intervention targets for multi-position intervention monitoring when the index exceeds a threshold, thereby identifying and determining whether abnormal deviations exist. It can accurately identify and determine whether deviations of UAV targets in a UAV swarm are abnormal, providing a foundation for the stability of subsequent collaborative flight control and mission execution of UAV swarms.

[0064] Specifically, Figure 1 shows a flowchart of the intelligent multi-target tracking method provided in an embodiment of the present invention.

[0065] In a preferred embodiment of the present invention, an intelligent multi-target tracking method specifically includes the following steps:

[0066] Step S101: Identify multiple tracking targets and perform remote unified monitoring on the multiple tracking targets to obtain unified monitoring data.

[0067] In this embodiment of the invention, a multi-target tracking task for a cluster of unmanned aerial vehicles (UAVs) is received, and then the UAVs are identified to determine multiple tracking targets. Based on the multi-target tracking task, remote monitoring parameters such as dynamic monitoring position, dynamic monitoring height, and dynamic monitoring angle are planned. Then, the multi-target tracking task is executed according to the remote monitoring parameters to perform remote unified monitoring of multiple tracking targets and obtain unified monitoring data.

[0068] It is understandable that multiple tracking targets refer to multiple drones in a drone swarm.

[0069] It is understandable that remote unified monitoring is the process of remotely filming multiple tracking targets (drones) in a drone swarm; unified monitoring data is video data.

[0070] Specifically, in another preferred embodiment provided by the present invention, the step of determining multiple tracking targets and performing remote unified monitoring on the multiple tracking targets to obtain unified monitoring data specifically includes the following steps:

[0071] Receive multi-target tracking tasks;

[0072] The multi-target tracking task is performed to identify multiple tracking targets;

[0073] Based on the multi-target tracking task, plan the remote monitoring parameters;

[0074] According to the remote monitoring parameters, multiple tracking targets are remotely monitored in a unified manner to obtain unified monitoring data.

[0075] Furthermore, the intelligent multi-target tracking method also includes the following steps:

[0076] Step S102: Perform collaborative stability analysis on multiple tracking targets, match multiple corresponding collaborative targets, perform mutual collaborative monitoring on multiple corresponding tracking targets, and obtain multiple collaborative monitoring data.

[0077] In this embodiment of the invention, based on the multi-target tracking task, dynamic routes corresponding to multiple tracking targets are planned. Based on the multiple dynamic routes, the temporal positions of multiple tracking targets are determined. Then, a collaborative stability analysis is performed on the multiple tracking targets. From the multiple tracking targets, multiple corresponding collaborative targets are matched. Then, collaborative monitoring sub-tasks corresponding to multiple collaborative targets are created. Then, based on the multiple collaborative monitoring sub-tasks, collaborative monitoring and control are performed on the multiple collaborative targets to realize mutual collaborative monitoring of multiple corresponding tracking targets and obtain multiple collaborative monitoring data.

[0078] Understandably, based on the temporal positions of multiple tracking targets, the temporal distance between each pair of tracking targets is calculated, and the average distance between multiple tracking targets in the entire multi-target tracking task is statistically analyzed; the matched cooperative target has the shortest average distance to the corresponding tracking target.

[0079] Specifically, in another preferred embodiment provided by the present invention, the step of performing cooperative stability analysis on multiple tracking targets, matching multiple corresponding cooperative targets, performing mutual cooperative monitoring on multiple corresponding tracking targets, and obtaining multiple cooperative monitoring data specifically includes the following steps:

[0080] Based on the multi-target tracking task, plan dynamic routes for multiple tracking targets;

[0081] Based on multiple dynamic routes, a cooperative stability analysis is performed on multiple tracking targets, and multiple corresponding cooperative targets are matched from the multiple tracking targets;

[0082] Create multiple collaborative monitoring subtasks corresponding to the aforementioned collaborative objectives;

[0083] Based on the multiple collaborative monitoring sub-tasks, multiple corresponding tracking targets are monitored collaboratively to obtain multiple collaborative monitoring data.

[0084] Furthermore, the intelligent multi-target tracking method also includes the following steps:

[0085] Step S103: Perform deviation analysis on the unified monitoring data and multiple collaborative monitoring data, select suspicious targets, and calculate the dynamic suspiciousness index.

[0086] In this embodiment of the invention, based on unified monitoring data, attitude deviation is identified for multiple tracking targets, and multiple attitude deviation data are recorded. Furthermore, based on multiple collaborative monitoring data, distance deviation is identified for multiple corresponding tracking targets, and multiple distance deviation data are recorded. According to the multiple attitude deviation data and multiple distance deviation data, suspicious targets with deviation values ​​not exceeding a preset abnormal value are selected from the multiple tracking targets. Then, from the multiple attitude deviation data and multiple distance deviation data, attitude suspicious deviation data and distance suspicious deviation data corresponding to the suspicious targets are filtered. Based on the attitude suspicious deviation data and distance suspicious deviation data, a dynamic suspiciousness index for the suspicious targets is calculated. Specifically, the formula for calculating the dynamic suspiciousness index is:

[0087] ;

[0088] ;

[0089] in, This is a dynamic suspicion index; For the time interval of deviation; for Attitude deviation value at any given time; for Distance deviation at any given time; This is the preset correction factor.

[0090] Specifically, in another preferred embodiment provided by the present invention, the deviation analysis of the unified monitoring data and multiple collaborative monitoring data, the selection of suspicious targets, and the calculation of the dynamic suspiciousness index specifically include the following steps:

[0091] Based on the unified monitoring data, attitude deviations of multiple tracking targets are identified, and multiple attitude deviation data are recorded.

[0092] Based on multiple collaborative monitoring data, distance deviation is identified for multiple corresponding tracking targets, and multiple distance deviation data are recorded.

[0093] Based on the multiple attitude deviation data and the multiple distance deviation data, a suspicious target is selected from the multiple tracking targets;

[0094] Based on multiple attitude deviation data and multiple distance deviation data, the dynamic suspicion index of the suspicious target is calculated.

[0095] Furthermore, the intelligent multi-target tracking method also includes the following steps:

[0096] Step S104: When the dynamic suspicion index exceeds the preset index threshold, select multiple intervention targets.

[0097] In this embodiment of the invention, by comparing the dynamic suspicious index with a preset index threshold, if the dynamic suspicious index is greater than the index threshold, it is determined that the deviation of the suspicious target has an abnormal trend. At this time, the index difference between the dynamic suspicious index and the index threshold is calculated, and then the number of interventions is matched according to the index difference. Then, the number of intervention targets is selected from multiple tracking targets.

[0098] Understandably, the larger the index difference, the more interventions are made.

[0099] Understandably, the selection of intervention targets is based on the average distance between the suspected target and multiple other tracked targets. By arranging multiple average distances from smallest to largest, the tracked target corresponding to the average distance of the number of interventions ranked first is selected as the intervention target.

[0100] Specifically, in another preferred embodiment provided by the present invention, the step of selecting multiple intervention targets when the dynamic suspicion index exceeds a preset index threshold specifically includes the following steps:

[0101] The dynamic suspicion index is compared with a preset index threshold;

[0102] When the dynamic suspicion index is greater than the index threshold, the index difference between the dynamic suspicion index and the index threshold is calculated;

[0103] The number of interventions is matched based on the index difference;

[0104] From the multiple tracking targets, select the number of targets to be intervened.

[0105] Furthermore, the intelligent multi-target tracking method also includes the following steps:

[0106] Step S105: Through multiple intervention targets, perform multi-position intervention monitoring on the suspicious target, obtain multiple intervention monitoring data, and identify and determine whether there is any abnormal deviation.

[0107] In this embodiment of the invention, multiple intervention monitoring subtasks for multiple intervention targets are created. Based on the multiple intervention monitoring subtasks, multi-position intervention monitoring control is performed on multiple corresponding intervention targets for suspected targets. Then, intervention monitoring data fed back by multiple intervention targets is obtained. By performing comprehensive distance and attitude recognition on multiple intervention monitoring data, it is determined whether there is any abnormal deviation.

[0108] Understandably, the process of multi-target intervention monitoring involves monitoring suspicious targets at close range through multiple intervention targets, thereby enabling more accurate distance and attitude identification of suspicious targets, and providing accurate and comprehensive data for judging whether there are abnormal deviations.

[0109] Specifically, in another preferred embodiment provided by the present invention, the step of performing multi-point intervention monitoring on the suspected target through multiple intervention targets, acquiring multiple intervention monitoring data, and identifying and determining whether there is any abnormal deviation specifically includes the following steps:

[0110] Create multiple intervention monitoring subtasks for the aforementioned intervention targets;

[0111] Based on the multiple intervention monitoring sub-tasks, multi-point intervention monitoring and control are performed on the multiple intervention targets targeting the suspected targets.

[0112] Acquire multiple intervention monitoring data;

[0113] The multiple intervention monitoring data are comprehensively identified to determine whether there are any abnormal deviations.

[0114] Furthermore, Figure 2 shows the application architecture diagram of the intelligent multi-target tracking system provided in the embodiment of the present invention.

[0115] Specifically, in another preferred embodiment provided by the present invention, an intelligent multi-target tracking system includes:

[0116] The remote unified monitoring module 101 is used to identify multiple tracking targets and perform remote unified monitoring on the multiple tracking targets to obtain unified monitoring data.

[0117] In this embodiment of the invention, the remote unified monitoring module 101 receives a multi-target tracking task for a cluster of drones, then performs target identification on the multi-target tracking task to determine multiple tracking targets, and then plans remote monitoring parameters such as dynamic monitoring position, dynamic monitoring height, and dynamic monitoring angle based on the multi-target tracking task. Finally, according to the remote monitoring parameters, the multi-target tracking task is executed to perform remote unified monitoring on multiple tracking targets and obtain unified monitoring data.

[0118] The mutual collaborative monitoring module 102 is used to perform collaborative stability analysis on multiple tracking targets, match multiple corresponding collaborative targets, perform mutual collaborative monitoring on multiple corresponding tracking targets, and obtain multiple collaborative monitoring data.

[0119] In this embodiment of the invention, the mutual collaborative monitoring module 102 plans dynamic routes corresponding to multiple tracking targets according to the multi-target tracking task, determines the temporal positions of multiple tracking targets based on the multiple dynamic routes, performs collaborative stability analysis on the multiple tracking targets, matches multiple corresponding collaborative targets from the multiple tracking targets, creates collaborative monitoring sub-tasks corresponding to the multiple collaborative targets, and then performs collaborative monitoring and control on the multiple collaborative targets according to the multiple collaborative monitoring sub-tasks, so as to realize mutual collaborative monitoring of multiple corresponding tracking targets and obtain multiple collaborative monitoring data.

[0120] Specifically, in another preferred embodiment provided by the present invention, the mutually cooperating monitoring module 102 specifically includes:

[0121] A dynamic route planning unit is used to plan dynamic routes for multiple tracking targets based on the multi-target tracking task.

[0122] The cooperative stability analysis unit is used to perform cooperative stability analysis on multiple tracking targets based on multiple dynamic routes, and to match multiple corresponding cooperative targets from the multiple tracking targets;

[0123] The subtask creation unit is used to create multiple collaborative monitoring subtasks corresponding to the collaborative objectives.

[0124] The mutual collaborative monitoring unit is used to perform mutual collaborative monitoring on multiple corresponding tracking targets according to multiple collaborative monitoring sub-tasks, and to obtain multiple collaborative monitoring data.

[0125] Furthermore, the intelligent multi-target tracking system also includes:

[0126] The suspicion index calculation module 103 is used to perform deviation analysis on the unified monitoring data and multiple collaborative monitoring data, select suspicious targets, and calculate the dynamic suspicion index.

[0127] In this embodiment of the invention, the suspicion index calculation module 103 identifies attitude deviations of multiple tracking targets based on unified monitoring data, records multiple attitude deviation data, and identifies distance deviations of multiple corresponding tracking targets based on multiple collaborative monitoring data, records multiple distance deviation data. Based on the multiple attitude deviation data and multiple distance deviation data, it selects suspicious targets from the multiple tracking targets whose deviation values ​​are no greater than a preset abnormal value. Then, it filters the attitude suspicion deviation data and distance suspicion deviation data corresponding to the suspicious targets from the multiple attitude deviation data and multiple distance deviation data. Based on the attitude suspicion deviation data and distance suspicion deviation data, it calculates the dynamic suspicion index of the suspicious targets. Specifically, the calculation formula for the dynamic suspicion index is:

[0128] ;

[0129] ;

[0130] in, This is a dynamic suspicion index; For the time interval of deviation; for Attitude deviation value at any given time; for Distance deviation at any given time; This is the preset correction factor.

[0131] Specifically, in another preferred embodiment provided by the present invention, the suspicion index calculation module 103 specifically includes:

[0132] The attitude deviation recognition unit is used to identify attitude deviations of multiple tracking targets based on the unified monitoring data and record multiple attitude deviation data.

[0133] The distance deviation identification unit is used to identify distance deviations from multiple corresponding tracking targets based on multiple collaborative monitoring data, and to record multiple distance deviation data.

[0134] A suspicious target selection unit is configured to select a suspicious target from a plurality of tracked targets based on a plurality of attitude deviation data and a plurality of distance deviation data;

[0135] The dynamic suspicion index calculation unit is used to calculate the dynamic suspicion index of the suspicious target based on multiple attitude deviation data and multiple distance deviation data.

[0136] Furthermore, the intelligent multi-target tracking system also includes:

[0137] The intervention target selection module 104 is used to select multiple intervention targets when the dynamic suspicion index exceeds a preset index threshold.

[0138] In this embodiment of the invention, the intervention target selection module 104 compares the dynamic suspicious index with a preset index threshold. If the dynamic suspicious index is greater than the index threshold, it determines that the deviation of the suspicious target has an abnormal trend. At this time, the index difference between the dynamic suspicious index and the index threshold is calculated, and the number of interventions is matched according to the index difference. Then, the number of intervention targets is selected from multiple tracking targets.

[0139] The multi-position intervention monitoring module 105 is used to perform multi-position intervention monitoring on the suspected target through multiple intervention targets, acquire multiple intervention monitoring data, and identify and determine whether there is any abnormal deviation.

[0140] In this embodiment of the invention, the multi-position intervention monitoring module 105 creates multiple intervention monitoring sub-tasks for multiple intervention targets. Based on the multiple intervention monitoring sub-tasks, it performs multi-position intervention monitoring control on multiple corresponding intervention targets for suspected targets, and then obtains intervention monitoring data fed back by multiple intervention targets. By performing comprehensive distance and attitude recognition on multiple intervention monitoring data, it determines whether there is any abnormal deviation.

[0141] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. An intelligent multi-target tracking method, characterized in that, The method specifically includes the following steps: identifying multiple tracking targets and performing remote unified monitoring on the multiple tracking targets to obtain unified monitoring data; performing collaborative stability analysis on the multiple tracking targets, matching multiple corresponding collaborative targets, and performing mutual collaborative monitoring on the multiple corresponding tracking targets to obtain multiple collaborative monitoring data; performing deviation analysis on the unified monitoring data and the multiple collaborative monitoring data, selecting suspicious targets, and calculating a dynamic suspiciousness index; when the dynamic suspiciousness index exceeds a preset index threshold, selecting multiple intervention targets; and performing multi-position intervention monitoring on the suspicious targets through the multiple intervention targets to obtain multiple intervention monitoring data, and identifying and determining whether there is any abnormal deviation. The deviation analysis of the unified monitoring data and multiple collaborative monitoring data, the selection of suspicious targets, and the calculation of a dynamic suspicion index specifically include the following steps: Based on the unified monitoring data, perform attitude deviation identification on multiple tracking targets and record multiple attitude deviation data; based on the multiple collaborative monitoring data, perform distance deviation identification on multiple corresponding tracking targets and record multiple distance deviation data; select suspicious targets from the multiple tracking targets according to the multiple attitude deviation data and the multiple distance deviation data; calculate the dynamic suspicion index of the suspicious target based on the multiple attitude deviation data and the multiple distance deviation data; the formula for calculating the dynamic suspicion index is: ； ;in, This is a dynamic suspicion index; For the time interval of deviation; for Attitude deviation value at any given time; for Distance deviation at any given time; This is the preset correction factor.

2. The intelligent multi-target tracking method according to claim 1, characterized in that, The process of identifying multiple tracking targets and performing remote unified monitoring on these targets to obtain unified monitoring data specifically includes the following steps: receiving a multi-target tracking task; identifying targets in the multi-target tracking task to determine multiple tracking targets; planning remote monitoring parameters based on the multi-target tracking task; and performing remote unified monitoring on the multiple tracking targets according to the remote monitoring parameters to obtain unified monitoring data.

3. The intelligent multi-target tracking method according to claim 1, characterized in that, The step of performing collaborative stability analysis on multiple tracking targets, matching multiple corresponding collaborative targets, performing mutual collaborative monitoring on multiple corresponding tracking targets, and obtaining multiple collaborative monitoring data specifically includes the following steps: planning dynamic routes for multiple tracking targets based on the multi-target tracking task; performing collaborative stability analysis on multiple tracking targets based on the multiple dynamic routes, and matching multiple corresponding collaborative targets from the multiple tracking targets; creating collaborative monitoring sub-tasks corresponding to multiple collaborative targets; and performing mutual collaborative monitoring on multiple corresponding tracking targets according to the multiple collaborative monitoring sub-tasks to obtain multiple collaborative monitoring data.

4. The intelligent multi-target tracking method according to claim 1, characterized in that, The step of selecting multiple intervention targets when the dynamic suspicion index exceeds a preset index threshold specifically includes the following steps: comparing the dynamic suspicion index with the preset index threshold; when the dynamic suspicion index is greater than the index threshold, calculating the index difference between the dynamic suspicion index and the index threshold; matching the number of interventions based on the index difference; and selecting the number of intervention targets from the multiple tracking targets.

5. The intelligent multi-target tracking method according to claim 1, characterized in that, The process of performing multi-point intervention monitoring on the suspected target through multiple intervention targets, acquiring multiple intervention monitoring data, and identifying and determining whether there is any abnormal deviation specifically includes the following steps: creating intervention monitoring sub-tasks for multiple intervention targets; performing multi-point intervention monitoring control on multiple intervention targets for the suspected target according to the multiple intervention monitoring sub-tasks; acquiring multiple intervention monitoring data; and comprehensively identifying the multiple intervention monitoring data to determine whether there is any abnormal deviation.

6. An intelligent multi-target tracking system, characterized in that, The system includes a remote unified monitoring module, a mutual collaborative monitoring module, a suspicion index calculation module, an intervention target selection module, and a multi-position intervention monitoring module. Specifically: the remote unified monitoring module identifies multiple tracking targets and performs remote unified monitoring on these targets to obtain unified monitoring data; the mutual collaborative monitoring module performs collaborative stability analysis on the multiple tracking targets, matches multiple corresponding collaborative targets, performs mutual collaborative monitoring on the corresponding tracking targets, and obtains multiple collaborative monitoring data; the suspicion index calculation module performs deviation analysis on the unified monitoring data and the multiple collaborative monitoring data, selects suspicious targets, and calculates a dynamic suspicion index; the intervention target selection module selects multiple intervention targets when the dynamic suspicion index exceeds a preset index threshold; and the multi-position intervention monitoring module is used to monitor multiple... The intervention target involves multi-position intervention monitoring of the suspected target, acquiring multiple intervention monitoring data, identifying and determining whether there are abnormal deviations; the deviation analysis of the unified monitoring data and multiple collaborative monitoring data, selecting suspected targets, and calculating a dynamic suspicion index specifically includes the following steps: based on the unified monitoring data, identifying attitude deviations of multiple tracking targets and recording multiple attitude deviation data; based on the multiple collaborative monitoring data, identifying distance deviations of multiple corresponding tracking targets and recording multiple distance deviation data; selecting suspected targets from the multiple tracking targets according to the multiple attitude deviation data and multiple distance deviation data; and calculating the dynamic suspicion index of the suspected target based on the multiple attitude deviation data and multiple distance deviation data; the calculation formula for the dynamic suspicion index is: ； ;in, This is a dynamic suspicion index; For the time interval of deviation; for Attitude deviation value at any given time; for Distance deviation at any given time; This is the preset correction factor.

7. The intelligent multi-target tracking system according to claim 6, characterized in that, The mutual collaborative monitoring module specifically includes: a dynamic route planning unit, used to plan dynamic routes for multiple tracking targets based on the multi-target tracking task; a collaborative stability analysis unit, used to perform collaborative stability analysis on multiple tracking targets based on the multiple dynamic routes, and match multiple corresponding collaborative targets from the multiple tracking targets; a sub-task creation unit, used to create collaborative monitoring sub-tasks corresponding to the multiple collaborative targets; and a mutual collaborative monitoring unit, used to perform mutual collaborative monitoring on multiple corresponding tracking targets based on the multiple collaborative monitoring sub-tasks, and obtain multiple collaborative monitoring data.

8. The intelligent multi-target tracking system according to claim 6, characterized in that, The suspiciousness index calculation module specifically includes: a posture deviation identification unit, used to identify posture deviations of multiple tracking targets based on the unified monitoring data and record multiple posture deviation data; a distance deviation identification unit, used to identify distance deviations of multiple corresponding tracking targets based on multiple collaborative monitoring data and record multiple distance deviation data; a suspicious target selection unit, used to select a suspicious target from multiple tracking targets according to the multiple posture deviation data and the multiple distance deviation data; and a dynamic suspiciousness index calculation unit, used to calculate the dynamic suspiciousness index of the suspicious target based on the multiple posture deviation data and the multiple distance deviation data.

Images