Object motion trajectory extraction method and device, computer equipment and medium

Abstract

The application relates to the technical field of target tracking detection, and discloses an object motion trajectory extraction method and device, computer equipment and a medium. The method comprises the following steps: acquiring a plurality of trajectory points, the sampling time of the plurality of trajectory points is different, and the plurality of trajectory points correspond to a plurality of target objects; the sampling time of the plurality of trajectory points is utilized to determine a trajectory endpoint corresponding to a first target object from the plurality of trajectory points; the trajectory endpoint is taken as a current query point, and a target screening area is determined with the current query point as the center; based on the size relationship between the sampling time of the trajectory point in the target screening area and the sampling time of the current query point, a next trajectory point belonging to the first target object is determined; the next trajectory point is updated to the current query point, the trajectory point belonging to the first target object is determined, and the motion trajectory of the first target object is obtained, thereby solving the problem that the trajectories of the target objects cannot be accurately and effectively extracted.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of target tracking detection, and particularly relates to an object motion trajectory extraction method and device, computer equipment and a medium. BACKGROUND

[0002] In the field of target tracking detection, for single target tracking detection, a multi-stage hypothesis test can be used to strictly recursively iterate according to a time gradient in combination with the motion characteristics of the observed target object, the track points of the same target object at different times are associated, and the entire motion trajectory is obtained. For multi-target tracking detection, multiple trajectories are obtained, which are mixed together. In theory, each trajectory can still be extracted by the mutual difference of the motion characteristics of each target object.

[0003] However, sometimes, due to limited device capabilities or special scenarios, only the position information of a target object can be randomly obtained at the same sampling time, that is, a track point is obtained, which leads to the fact that the track points of each target object appear intermittently and staggered in the time sequence. Due to the existence of observation errors, the traditional method cannot accurately and effectively extract the trajectories of each target object. SUMMARY

[0004] Therefore, the present application provides an object motion trajectory extraction method and device, computer equipment and a medium to solve the problem that the trajectories of each target object cannot be accurately and effectively extracted.

[0005] In a first aspect, the present application provides an object motion trajectory extraction method, comprising: obtaining a plurality of track points, the sampling times of the plurality of track points being different and corresponding to a plurality of target objects; determining a track endpoint corresponding to a first target object from the plurality of track points by using the sampling times of the plurality of track points; taking the track endpoint as a current query point and determining a target screening region centered on the current query point; determining a next track point belonging to the first target object based on the size relationship between the sampling time of the track point in the target screening region and the sampling time of the current query point; updating the next track point to the current query point, determining the track point belonging to the first target object, and obtaining the motion trajectory of the first target object.

[0006] In the object trajectory extraction method of the present application, the trajectory endpoint corresponding to the first target object is determined from the plurality of trajectory points according to the sampling time of the plurality of trajectory points; the trajectory endpoint is taken as the current query point, and a target screening region is determined with the current query point as the center; for the trajectory points in the target screening region, the next trajectory point of the first target object is accurately screened out by using the size relationship between the sampling time of the next trajectory point and the current query point; and then the next trajectory point is updated as the current query point to continue determining the target screening region, i.e. the target screening region is continued to be determined with the next trajectory point as the center, and the trajectory points belonging to the first target object are determined from the plurality of trajectory points through iterative circulation, so as to obtain the motion trajectory of the first target object. The trajectory points belonging to the first target object are screened by the target screening region, the trajectory of the target object is effectively and accurately extracted, the calculation amount of target trajectory tracking is greatly reduced, the real-time performance is improved, and the problem that the trajectories of the target objects cannot be accurately and effectively extracted is solved.

[0007] In an optional embodiment, the trajectory endpoint is a trajectory starting point or a trajectory ending point; and the trajectory endpoint corresponding to the first target object is determined from the plurality of trajectory points, including: if the trajectory endpoint is the trajectory starting point, the trajectory point with the earliest sampling time is determined as the trajectory endpoint; or if the trajectory endpoint is the trajectory ending point, the trajectory point with the latest sampling time is determined as the trajectory endpoint.

[0008] The trajectory point with the earliest sampling time is determined as the trajectory endpoint, or the trajectory point with the latest sampling time is determined as the trajectory endpoint, so that the trajectory endpoint can be efficiently and quickly determined, and the calculation amount can be further reduced and the real-time performance can be improved.

[0009] In an optional embodiment, the target screening region is determined with the current query point as the center, including: a preset motion speed of the first target object is obtained; the target step is determined by using the attribute feature of the preset motion speed; and the target screening region is determined by diffusing the target step outward from the current query point.

[0010] The target step is determined in combination with the attribute feature of the preset motion speed of the first target object, so that the determined target step is reasonable; and the target screening region is determined by diffusing the target step outward from the current query point in combination with the target step, so that the determined target screening region is reasonable, i.e. the next trajectory point of the first target object is not in the target screening region due to the target screening region being too small, or too many trajectory points fall into the target screening region due to the target screening region being too large, and the next trajectory point of the first target object cannot be accurately determined.

[0011] In an optional implementation, the next trajectory point belonging to the first target object is determined based on a size relationship between a sampling time of the trajectory point in the target screening region and a sampling time of the current query point, and includes: determining time difference values between the sampling time of the trajectory point in the target screening region and the sampling time of the current query point, to obtain a plurality of time difference values; determining the trajectory point in the target screening region corresponding to the smallest time difference value as a candidate trajectory point; and determining the next trajectory point belonging to the first target object based on a time gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point.

[0012] By screening the trajectory point with the smallest time difference value as the candidate trajectory point, the current query point and the next trajectory point can be closely connected in the time sequence, and situations such as trajectory breakage or mismatch caused by a too large time interval can be avoided; the time gradient direction (i.e., the order of the sampling times) is used to further determine whether the candidate trajectory point is the next trajectory point of the first target object, so that the trajectory point can be extended in the time increasing direction or the time decreasing direction, and the next trajectory point can be further accurately and reasonably determined.

[0013] In an optional implementation, the trajectory endpoint is a trajectory starting point or a trajectory ending point; the next trajectory point belonging to the first target object is determined based on a time gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point, and includes: if the trajectory endpoint is the trajectory starting point and the sampling time of the candidate trajectory point is later than the sampling time of the current query point, the candidate trajectory point is determined as the next trajectory point of the first target object; and if the trajectory endpoint is the trajectory ending point and the sampling time of the candidate trajectory point is earlier than the sampling time of the current query point, the candidate trajectory point is determined as the next trajectory point of the first target object.

[0014] The time gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point can be used to further accurately and reasonably determine the next trajectory point, so that the trajectory point of other target objects can be avoided to be associated to the motion trajectory of the first target object.

[0015] In an optional embodiment, the trajectory end point is a trajectory start point or a trajectory end point; the next trajectory point is updated as the current query point, the trajectory points belonging to the first target object are determined, and the motion trajectory of the first target object is obtained, including: the next trajectory point is updated as the current query point, and the target screening area is determined again with the current query point as the center; if the trajectory end point is the trajectory start point, and there is a sampling time of the trajectory point in the target screening area that is earlier than the sampling time of the current query point, the current query point is determined as the trajectory end point, and the motion trajectory of the first target object is obtained; if the trajectory end point is the trajectory end point, and there is a sampling time of the trajectory point in the target screening area that is later than the sampling time of the current query point, the current query point is determined as the trajectory start point, and the motion trajectory of the first target object is obtained.

[0016] In the case where the trajectory end point is the trajectory start point, if the sampling time of the trajectory point in the target screening area is earlier than the sampling time of the current query point, it indicates that there is no trajectory point later than the sampling time of the current query point in the current plurality of trajectory points, so the current query point can be efficiently and quickly determined as the trajectory end point of the first target object. In the case where the trajectory end point is the trajectory end point, if the sampling time of the trajectory point in the target screening area is later than the sampling time of the current query point, it indicates that there is no trajectory point earlier than the sampling time of the current query point in the current plurality of trajectory points, so the current query point can be efficiently and quickly determined as the trajectory start point of the first target object.

[0017] In an optional embodiment, after the next trajectory point is updated as the current query point, the trajectory points belonging to the first target object are determined, and the motion trajectory of the first target object is obtained, the method further includes: determining whether there are a plurality of remaining trajectory points in the plurality of trajectory points, the remaining trajectory points being trajectory points other than the trajectory points in the motion trajectory of the first target object; if there are a plurality of remaining trajectory points, the sampling time of the plurality of remaining trajectory points is used to determine the trajectory end point of the second target object from the plurality of remaining trajectory points; the second target object is updated as the first target object, and the step of taking the trajectory end point as the current query point and determining the target screening area with the current query point as the center is entered.

[0018] After the motion trajectory of the first target object is obtained, if there are remaining trajectory points in the plurality of trajectory points, the foregoing steps are iterated, i.e., the trajectory points belonging to the second target object are continuously screened by the target screening area, and the motion trajectory of the second target object is obtained. That is, through loop iteration, all trajectory points are extracted, and thus the motion trajectories of a plurality of target objects are obtained, thereby solving the problem that the trajectories of the target objects cannot be accurately and effectively extracted.

[0019] In a second aspect, the present application provides an object motion trajectory extraction device, comprising: an acquisition module configured to acquire a plurality of trajectory points, the plurality of trajectory points being different in sampling time and corresponding to a plurality of target objects; a first determination module configured to determine a trajectory endpoint corresponding to a first target object from the plurality of trajectory points according to the sampling times of the plurality of trajectory points; a second determination module configured to take the trajectory endpoint as a current query point and determine a target screening region centered on the current query point; a third determination module configured to determine a next trajectory point belonging to the first target object based on the size relationship between the sampling time of the trajectory point in the target screening region and the sampling time of the current query point; and a fourth determination module configured to update the next trajectory point to the current query point, determine the trajectory points belonging to the first target object, and obtain the motion trajectory of the first target object.

[0020] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, which are communicatively connected with each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the object motion trajectory extraction method of the first aspect or any of the corresponding embodiments thereof.

[0021] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer instructions, and the computer instructions are used to make a computer execute the object motion trajectory extraction method of the first aspect or any of the corresponding embodiments thereof.

[0022] In a fifth aspect, the present application provides a computer program product, which comprises computer instructions, and the computer instructions are used to make a computer execute the object motion trajectory extraction method of the first aspect or any of the corresponding embodiments thereof. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the description of the specific embodiments or the prior art. Obviously, the drawings described below are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0024] Figure 1 is a flowchart of an object motion trajectory extraction method according to an embodiment of the present application;

[0025] Figure 2 is a schematic diagram of determining a candidate trajectory point according to an embodiment of the present application;

[0026] Figure 3 is another schematic diagram of determining a candidate trajectory point according to an embodiment of the present application;

[0027] Figure 4 is a schematic diagram of determining a motion trajectory of a first target object according to an embodiment of the present application;

[0028] Figure 5 is a flowchart of another method for extracting a motion trajectory of an object according to an embodiment of the present application;

[0029] Figure 6 is a flowchart of still another method for extracting a motion trajectory of an object according to an embodiment of the present application;

[0030] Figure 7 is a schematic diagram of motion trajectories of four aircrafts according to an embodiment of the present application;

[0031] Figure 8 is a schematic diagram of a plurality of trajectory points obtained by sampling the motion trajectories of the four aircrafts according to an embodiment of the present application;

[0032] Figure 9 is a schematic diagram of the plurality of trajectory points after adding perturbations according to an embodiment of the present application;

[0033] Figure 10 is a schematic diagram of determining trajectory endpoints of the four aircrafts according to an embodiment of the present application;

[0034] Figure 11 is a schematic diagram of the motion trajectories of the four aircrafts obtained by using the method for extracting a motion trajectory of an object according to an embodiment of the present application;

[0035] Figure 12 is a schematic diagram of the motion trajectories of the four aircrafts obtained by using a traditional method;

[0036] Figure 13 is a structural block diagram of a device for extracting a motion trajectory of an object according to an embodiment of the present application;

[0037] Figure 14 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0038] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0039] In the field of target tracking detection, the target track of an aircraft or a ship should be a straight line or a smooth curve in an ideal case. Due to observation errors, the actual position of a target object at each sampling time deviates from the theoretical position. Thus, for tracking detection of a single target, the motion characteristics of the observed target object can be combined, a multi-stage hypothesis testing method can be used, and the track points of the same target object at different times can be associated by strict time gradient recursion iteration, so as to obtain the entire motion track of the target object.

[0040] Currently, multi-target tracking detection methods exist in the simultaneous observation process of multiple target objects, that is, the position information of all target objects in the field of view is obtained at the same sampling time, and the track is tracked simultaneously. The tracks are continuous in time, so theoretically, the tracks of each target object can still be extracted by the mutual difference of the motion characteristics of each target object.

[0041] However, sometimes, due to the limited ability of observation equipment or special scenes, only the position information of a target object can be randomly obtained at the same sampling time, that is, only the track point of a target object can be obtained, which leads to the fact that the track points of each target object appear discontinuously and staggered in time sequence. The traditional track tracking association method based on consistent speed is only suitable for continuous tracks in time sequence, and is not suitable for time staggered tracks. On the other hand, when the tracks of each target object are staggered in geographical position, due to the existence of observation errors, the traditional method cannot accurately and effectively extract the tracks of each target object.

[0042] Therefore, the present application provides a method and device for extracting the motion track of an object, a computer device and a medium. The method comprises the following steps: obtaining a plurality of track points, the sampling times of the plurality of track points are different and correspond to a plurality of target objects; determining a track endpoint corresponding to a first target object from the plurality of track points by using the sampling times of the plurality of track points; taking the track endpoint as a current query point and determining a target screening region centered on the current query point; determining a next track point belonging to the first target object based on the size relationship between the sampling times of the track points in the target screening region and the sampling time of the current query point; updating the next track point to the current query point, determining the track points belonging to the first target object, and obtaining the motion track of the first target object.

[0043] In the object motion trajectory extraction method of the present application, according to the sampling time of the plurality of trajectory points, the trajectory endpoint corresponding to the first target object is determined from the plurality of trajectory points; the trajectory endpoint is taken as the current query point, and the target screening region is determined with the current query point as the center; for the trajectory points in the target screening region, the next trajectory point of the first target object is accurately screened out by using the size relationship of the sampling time of the trajectory points and the current query point; and then the next trajectory point is updated as the current query point to continue determining the target screening region, i.e., the target screening region is continued to be determined with the next trajectory point as the center, and the trajectory points belonging to the first target object are determined from the plurality of trajectory points through cyclic iteration, so as to obtain the motion trajectory of the first target object. The trajectory points belonging to the first target object are screened through the target screening region, the trajectory of the target object is effectively and accurately extracted, the calculation amount of target trajectory tracking is reduced, the real-time performance is improved, and the problem that the trajectories of the target objects cannot be accurately and effectively extracted is solved.

[0044] It should be noted that the application scenarios described in the above embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems. The object motion trajectory extraction method provided by the embodiments of the present application can be applied to various application scenarios in which the trajectories of target objects need to be separated and extracted.

[0045] According to the embodiments of the present application, an object motion trajectory extraction method is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0046] In the present embodiment, an object motion trajectory extraction method is provided, which can be used in a computer device, for example, a server, a trajectory point sampling device, various edge devices corresponding to the trajectory point sampling device, and other devices, Figure 1 is a flowchart of the object motion trajectory extraction method according to the embodiments of the present application, as shown in Figure 1 , the flowchart includes the following steps:

[0047] In step S102, a plurality of trajectory points are obtained, and the sampling time of the plurality of trajectory points is different and corresponds to a plurality of target objects.

[0048] A trajectory point can be a spatial position of an object at a specific moment during movement. From a mathematical perspective, it can be a coordinate point on a trajectory (i.e., a curve or a straight line formed by all trajectory points) of the object, which is usually represented by a three-dimensional coordinate system (x, y, z) or a two-dimensional coordinate system (x, y). From a physical perspective, each trajectory point corresponds to a position of the object at a specific moment, reflecting the instantaneous state of the object movement.

[0049] For obtaining a plurality of trajectory points, in actual applications, a plurality of target objects can be sampled by a trajectory point sampling device or various edge devices corresponding to the trajectory point sampling device to obtain a plurality of trajectory points. In simulation experiments, a plurality of trajectory points can also be randomly generated by a normal distribution.

[0050] As shown in the foregoing, due to limited device capability or special scenarios, only one position information of a target object can be randomly obtained at the same sampling moment, i.e., one trajectory point is obtained. For example, a first trajectory point of a first target object is obtained at a first sampling moment, a second trajectory point of the first target object is obtained at a second sampling moment, a third trajectory point of a second target object is obtained at a third sampling moment, and a fourth trajectory point of a third target object is obtained at a fourth sampling moment, so that a plurality of trajectory points discontinuous and staggered in a time sequence can be obtained.

[0051] In step S104, a trajectory endpoint corresponding to the first target object is determined from the plurality of trajectory points by using sampling moments of the plurality of trajectory points.

[0052] The sampling moment of a trajectory point is used to record a time point at which the trajectory point is sampled. The sampling moment of each trajectory point can be artificially set or randomly generated by a system, and the sampling time interval of any two adjacent trajectory points can be the same or different.

[0053] It should be understood that the first target object can be a first target object whose motion trajectory is extracted. The first target object can be a target object corresponding to a first trajectory point sampled. For example, in the foregoing example, the first target object can be a first target object sampled at a first sampling moment.

[0054] The trajectory endpoint can be a boundary point of the motion trajectory of the first target object in the time dimension. The trajectory endpoint can be a trajectory starting point or a trajectory ending point. The trajectory starting point can be a first trajectory point of the first target object in an observation or analysis time period, which corresponds to the earliest sampling moment. The trajectory ending point can be a last trajectory point of the first target object in the observation or analysis time period, which corresponds to the latest sampling moment.

[0055] Step S106, taking the trajectory endpoint as the current query point, and determining a target screening region centered on the current query point.

[0056] There are various implementation manners for determining the target screening region centered on the current query point. For example, a target screening region of fixed shape can be defined by mathematical formula or geometric model, which can be circular, rectangular, polygonal, etc. For another example, the space can be divided into regular grids (such as grid, hexagonal grid), and the target screening region can be determined by the grid where the current query point is located and the adjacent grids.

[0057] It should be understood that the target screening region can be a planar region in two-dimensional space, and the target screening region can be a spatial region in three-dimensional space.

[0058] Step S108, determining the next trajectory point belonging to the first target object based on the size relationship between the sampling time of the trajectory point in the target screening region and the sampling time of the current query point.

[0059] The next trajectory point of the first target object can be the next trajectory point adjacent to the current query point in position. The next trajectory point can be used to continue to determine the next trajectory point adjacent to the next trajectory point. In this way, through multiple loop iterations, multiple next trajectory points of the first target object can be obtained.

[0060] Due to the reasons such as multi-trajectory intersection and large positioning error, there are often multiple optional trajectory points in the target screening region centered on the current query point, which can come from other target objects or trajectory points spaced by multiple sampling times. In this way, the next trajectory point can be screened according to the principle that the time gradient of adjacent trajectory points on the same motion trajectory is the smallest. For example, the sampling time of the current query point is compared with the sampling time of the trajectory points in the target screening region one by one. If the trajectory endpoint is the trajectory starting point, the trajectory point corresponding to the sampling time later than and closest to the sampling time of the current query point is determined as the next trajectory point. If the trajectory endpoint is the trajectory termination point, the trajectory point corresponding to the sampling time earlier than and closest to the sampling time of the current query point is determined as the next trajectory point.

[0061] Step S110, updating the next trajectory point as the current query point, determining the trajectory point belonging to the first target object, and obtaining the motion trajectory of the first target object.

[0062] The motion trajectory of the first target object can be the position change sequence of the first target object in a period of time, which is usually composed of a series of trajectory points arranged in time sequence. These trajectory points can reflect the moving path and motion characteristics of the first target object in space.

[0063] After the next trajectory point is updated as the current query point, a step of determining a target screening region centered on the current query point can be performed, so that a plurality of next trajectory points of the first target object can be determined through iterative loops until another trajectory endpoint of the first target object is determined again. In this way, the motion trajectory of the first target object can be formed by the two trajectory endpoints and the plurality of next trajectory points between the two trajectory endpoints.

[0064] The method for extracting the motion trajectory of the object provided in the application determines the trajectory endpoint corresponding to the first target object from the plurality of trajectory points according to the sampling time of the plurality of trajectory points, determines the target screening region centered on the current query point by taking the trajectory endpoint as the current query point, accurately screens the next trajectory point of the first target object from the trajectory points in the target screening region according to the size relationship between the sampling time of the current query point and the sampling time of the next trajectory point, and then updates the next trajectory point as the current query point to continue determining the target screening region, i.e., continues to determine the target screening region centered on the next trajectory point, iteratively determines the trajectory points belonging to the first target object from the plurality of trajectory points, and thus obtains the motion trajectory of the first target object. The scheme screens the trajectory points belonging to the first target object by using the target screening region, effectively and accurately extracts the trajectory of the target object, reduces the calculation amount of the target trajectory tracking and improves the real-time performance, and thus solves the problem that the trajectories of the target objects cannot be accurately and effectively extracted.

[0065] When the first target object is tracked and detected, the trajectory endpoint of the first target object needs to be determined first, and then the next trajectory point adjacent to the trajectory endpoint is determined, so that the motion trajectory of the first target object is determined through iterative loops. The trajectory endpoint of the first target object can be the trajectory starting point of the first target object or the trajectory ending point of the first target object. Based on this, in some optional embodiments, the trajectory endpoint corresponding to the first target object is determined from the plurality of trajectory points, including: if the trajectory endpoint is the trajectory starting point, the trajectory point with the earliest sampling time is determined as the trajectory endpoint; and if the trajectory endpoint is the trajectory ending point, the trajectory point with the latest sampling time is determined as the trajectory endpoint.

[0066] In the above implementation, the trajectory point with the earliest sampling time is determined as the trajectory endpoint or the trajectory point with the latest sampling time is determined as the trajectory endpoint, which can efficiently and quickly determine the trajectory endpoint, and can further reduce the calculation amount and improve the real-time performance.

[0067] As shown in the foregoing, the first target object can be a first target object sampled at a first sampling time, and the trajectory point with the earliest sampling time or the trajectory point with the latest sampling time is determined as the trajectory endpoint of the first target object, so that the trajectory endpoint of the first target object can be accurately determined. After the trajectory of the first target object is extracted, the trajectory point with the earliest sampling time or the trajectory point with the latest sampling time is determined as the trajectory endpoint of the second target object from the remaining trajectory points, so that the trajectory point of the second target object can also be accurately and quickly determined. Thus, according to the principle of extending in the direction of the time gradient, when the trajectory of each target object is tracked, the trajectory point with the highest or lowest time dimension in the current trajectory point can be selected as the trajectory endpoint of the target object.

[0068] In some optional embodiments, the target screening area is determined with the current query point as the center, including: obtaining a preset motion speed of the first target object; determining a target step length by using an attribute feature of the preset motion speed; and spreading the target step length outward from the current query point to determine the target screening area.

[0069] The target step length can be determined in various ways by using the attribute feature of the preset motion speed. For example, in order to reduce the calculation amount, the moving distance of the preset motion speed of the first target object in a unit time can be directly determined as the target step length. For another example, considering the existence of positioning errors, it is necessary to ensure that the next trajectory point on the motion trajectory is within the target screening range. Due to staggered observation, the time interval between adjacent trajectory points on the same motion trajectory cannot be guaranteed, and therefore it is necessary to appropriately expand the range of the target screening area. Since the next trajectory point is determined based on the time nearest neighbor principle, even if the target screening area is relatively large, the probability of false association will not increase, and therefore the target step length can be determined in combination with the preset motion speed and the observation error.

[0070] Optionally, as shown in Figure 2 and Figure 3 , the current query point can be taken as the center of a circle, and the target step length can be taken as the radius to spread the target step length outward to obtain a circular target screening area. Since the circle itself has geometric symmetry, the determination of the circular target screening area can improve the accuracy and computational efficiency of the subsequent determination of the next trajectory point of the first target object.

[0071] The attribute feature of the preset motion speed of the first target object is combined to determine the target step length, so that the determined target step length is reasonable. Then, the target step length is combined to spread the target step length outward from the current query point as the center to determine the target screening region, so that the determined target screening region is reasonable. The next track point of the first target object will not be located outside the target screening region due to the target screening region being too small. Too many track points will not fall into the target screening region due to the target screening region being too large, and the next track point of the first target object cannot be accurately determined.

[0072] In some optional embodiments, the next track point belonging to the first target object is determined based on the size relationship between the sampling time of the track point in the target screening region and the sampling time of the current query point, and includes: determining the time difference between the sampling time of the track point in the target screening region and the sampling time of the current query point to obtain a plurality of time difference values; determining the track point in the target screening region corresponding to the smallest time difference value as the candidate track point; and determining the next track point belonging to the first target object by using the time gradient direction between the sampling time of the candidate track point and the sampling time of the current query point.

[0073] Optionally, the time gradient direction can be a time increasing direction or a time decreasing direction.

[0074] By screening the track point with the smallest time difference value as the candidate track point, it can be ensured that the current query point and the next track point closely connect in the time sequence, and the situations such as track breakage or mismatch caused by too large time interval can be avoided. By using the time gradient direction (i.e. the order of sampling time) to further determine whether the candidate track point is the next track point of the first target object, it can be ensured that the track point extends in the time increasing direction or the time decreasing direction, and the next track point can be further accurately and reasonably determined.

[0075] In some optional embodiments, the track endpoint is a track starting point or a track ending point. The next track point belonging to the first target object is determined by using the time gradient direction between the sampling time of the candidate track point and the sampling time of the current query point, and includes: if the track endpoint is the track starting point and the sampling time of the candidate track point is later than the sampling time of the current query point, the candidate track point is determined as the next track point of the first target object; and if the track endpoint is the track ending point and the sampling time of the candidate track point is earlier than the sampling time of the current query point, the candidate track point is determined as the next track point of the first target object.

[0076] As Figure 2As shown, when the endpoint of the trajectory is the starting point of the trajectory, the sampling time of the next trajectory point must be later than the sampling time of the current query point. Based on this, the sampling times of the candidate trajectory points and the sampling time of the current query point are in an increasing direction in time.

[0077] like Figure 3 As shown, when the trajectory endpoint is the trajectory termination point, the sampling time of the next trajectory point must be earlier than the sampling time of the current query point. Based on this, the sampling time of the candidate trajectory point and the sampling time of the current query point are in the direction of decreasing time.

[0078] In the above implementation, by utilizing the temporal gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point, the next trajectory point can be determined more accurately and reasonably, avoiding associating the trajectory points of other target objects with the motion trajectory of the first target object.

[0079] In some optional implementations, the trajectory endpoint is either the trajectory start point or the trajectory end point; updating the next trajectory point to the current query point, determining the trajectory point belonging to the first target object, and obtaining the motion trajectory of the first target object includes: updating the next trajectory point to the current query point, continuing to determine the target filtering area with the current query point as the center; if the trajectory endpoint is the trajectory start point, and the sampling time of the trajectory points within the target filtering area is earlier than the sampling time of the current query point, then the current query point is determined as the trajectory end point, and the motion trajectory of the first target object is obtained; if the trajectory endpoint is the trajectory end point, and the sampling time of the trajectory points within the target filtering area is later than the sampling time of the current query point, then the current query point is determined as the trajectory start point, and the motion trajectory of the first target object is obtained.

[0080] When the trajectory endpoint is the trajectory start point, if the sampling time of all trajectory points within the target filtering area is earlier than the sampling time of the current query point, it indicates that there are no trajectory points among the current multiple trajectory points with a sampling time later than the current query point. In this way, the trajectory termination point of the current query point as the first target object can be determined efficiently and quickly.

[0081] When the trajectory endpoint is the trajectory termination point, if the sampling time of all trajectory points within the target filtering area is later than the sampling time of the current query point, it indicates that there is no trajectory point among the current multiple trajectory points with an earlier sampling time than the current query point. In this way, the current query point can be efficiently and quickly determined as the trajectory starting point of the first target object.

[0082] like Figure 4As shown, after the trajectory starting point, the trajectory ending point and the plurality of next trajectory points belonging to the first target object are determined, the trajectory starting point, the plurality of next trajectory points and the trajectory ending point are connected in the time increasing direction to obtain the motion trajectory of the first target object.

[0083] In the embodiment, a method for extracting a motion trajectory of an object is provided, which can be used in a computer device, such as a server, a trajectory point sampling device, various edge devices corresponding to the trajectory point sampling device, and other devices. Figure 5 FIG. 1 is a flowchart of a method for extracting a motion trajectory of an object according to an embodiment of the present application. Figure 5 As shown, the flowchart includes the following steps:

[0084] In step S502, a plurality of trajectory points are obtained, the plurality of trajectory points are sampled at different time points and correspond to a plurality of target objects. For details, refer to step S102 of the embodiment shown in FIG. 1, which will not be repeated here. Figure 1

[0085] In step S504, the trajectory endpoints corresponding to the first target object are determined from the plurality of trajectory points by using the sampling time points of the plurality of trajectory points. For details, refer to step S104 of the embodiment shown in FIG. 1, which will not be repeated here. Figure 1

[0086] In step S506, the trajectory endpoints are taken as the current query point, and a target screening area is determined with the current query point as the center. For details, refer to step S106 of the embodiment shown in FIG. 1, which will not be repeated here. Figure 1

[0087] In step S508, the next trajectory point belonging to the first target object is determined based on the size relationship between the sampling time point of the trajectory point in the target screening area and the sampling time point of the current query point. For details, refer to step S108 of the embodiment shown in FIG. 1, which will not be repeated here. Figure 1

[0088] In step S510, the next trajectory point is updated as the current query point, the trajectory point belonging to the first target object is determined, and the motion trajectory of the first target object is obtained. For details, refer to step S110 of the embodiment shown in FIG. 1, which will not be repeated here. Figure 1

[0089] In step S512, it is determined whether there are a plurality of remaining trajectory points in the plurality of trajectory points, the remaining trajectory points being the trajectory points other than the trajectory points in the motion trajectory of the first target object.

[0090] ​​​​​There are many ways to determine whether there are remaining trajectory points in the plurality of trajectory points. For example, a pre-set screening condition is used to screen the identification information of each trajectory point to determine whether there are remaining trajectory points in the plurality of trajectory points. For another example, whether the trajectory points belong to the movement trajectory of the first target object is determined to determine whether there are remaining trajectory points in the plurality of trajectory points.

[0091] In step S514, if there are remaining trajectory points, the sampling time of the remaining trajectory points is used to determine the trajectory endpoint of the second target object from the remaining trajectory points.

[0092] If there are no remaining trajectory points, it indicates that all the trajectory points have been determined, i.e., the trajectories of all target objects have been separated.

[0093] In step S516, the second target object is updated to the first target object, and the step of taking the trajectory endpoint as the current query point and determining the target screening area centered on the current query point is entered.

[0094] After obtaining the movement trajectory of the first target object, if there are remaining trajectory points in the plurality of trajectory points, the foregoing steps are iterated, i.e., the trajectory points belonging to the second target object are continuously screened using the target screening area to obtain the movement trajectory of the second target object. That is, through cyclic iteration, all trajectory points are extracted to obtain the movement trajectories of the plurality of target objects, thereby solving the problem of being unable to accurately and effectively extract the trajectories of the target objects.

[0095] It should be understood that in the case where the target objects corresponding to the plurality of trajectory points are more than two, there can be multiple second target objects.

[0096] The object motion trajectory extraction method provided in the embodiment determines a trajectory endpoint corresponding to the first target object from the plurality of trajectory points according to the sampling time of the plurality of trajectory points, takes the trajectory endpoint as a current query point, determines a target screening region with the current query point as the center, accurately screens the next trajectory point of the first target object by using the size relationship between the sampling time of the current query point and the sampling time of the trajectory point in the target screening region, updates the next trajectory point as the current query point to continue determining the target screening region, that is, continues to determine the target screening region with the next trajectory point as the center, and iteratively determines all the trajectory points belonging to the first target object from the plurality of trajectory points, thereby obtaining the motion trajectory of the first target object. After obtaining the motion trajectory of the first target object, it is determined whether there are remaining trajectory points. If there are remaining trajectory points, the above steps are continued, and the trajectories of a plurality of target objects are extracted through multiple iterative cycles. The scheme screens the trajectory points belonging to the first target object through the target screening region, realizes effective and accurate extraction of the trajectories of the target objects, can reduce the calculation amount of target trajectory tracking and improve the real-time performance, and further solves the problem that the trajectories of the target objects cannot be accurately and effectively extracted.

[0097] In order to facilitate understanding of the object motion trajectory extraction method of the present application, as shown in Figure 6 The flowchart of the object motion trajectory extraction method provided in the embodiment of the present application is shown in the figure. The flowchart includes steps S601 to S608.

[0098] In step S601, in the process of extracting the motion trajectory of the first target object, the trajectory endpoint is determined from the plurality of trajectory points by using the sampling time of the plurality of trajectory points. The trajectory endpoint can be the trajectory starting point of the first target object or the trajectory ending point of the first target object.

[0099] In step S602, the target step is determined in combination with the preset motion speed of the target object and the observation error.

[0100] In step S603, the trajectory endpoint is taken as the current query point, the target step is diffused outward from the current query point as the center, and the target screening region is determined.

[0101] In step S604, the next trajectory point is determined by using the target screening region. That is, the next trajectory point is determined by using the size relationship between the sampling time of the current query point and the sampling time of the trajectory point in the target screening region.

[0102] Step S605: Determine whether there is a trajectory point within the target filtering area that meets predetermined conditions. If a trajectory point within the target filtering area meets the predetermined conditions, then that trajectory point is determined as the next trajectory point, and step S606 is executed; if no trajectory point within the target filtering area meets the predetermined conditions, then step S607 is executed. The predetermined conditions are: if the trajectory endpoint is the trajectory start point, there exists a trajectory point within the target filtering area whose sampling time difference with the current query point is the smallest and conforms to the time-increasing direction; if the trajectory endpoint is the trajectory end point, there exists a trajectory point within the target filtering area whose sampling time difference with the current query point is the smallest and conforms to the time-decreasing direction.

[0103] Step S606: Update the next trajectory point to the current query point.

[0104] Step S607: Determine the current query point as the starting point or ending point of the trajectory.

[0105] Step S608: Determine whether there are any remaining trajectory points among the multiple trajectory points. If there are remaining trajectory points among the multiple trajectory points, proceed to step S601; otherwise, end the process.

[0106] To facilitate understanding of the object motion trajectory extraction method of this application, the following example of extracting the motion trajectories of multiple aircraft through simulation experiments will be used to further illustrate the object motion trajectory extraction method of this application.

[0107] like Figure 7 The image shows the motion trajectories of four aircraft obtained through simulation. Each trajectory consists of 125 trajectory points, and all four aircraft move at the same altitude. For these trajectories, the motion trajectories of the four aircraft are sampled at 1-second intervals to obtain the following results. Figure 8 The diagram shows multiple trajectory points observed in staggered patterns. Then, the normal distribution is used... For example Figure 8 After adding random perturbations to the trajectory points shown, the following results are obtained: Figure 9 The multiple trajectory points shown, among which, In this context, 0 represents the mean, and 0.002 represents the variance. Extracting the motion trajectory of trajectory points after adding random perturbations improves the robustness of the extraction method presented in this application. The specific steps include:

[0108] First step: Through, as... Figure 9 The trajectory points shown are analyzed, and the trajectory point at the last sampling time among multiple trajectory points is taken as the trajectory endpoint, specifically as follows: Figure 10As shown. After extracting the trajectory of the first aircraft, the trajectory point at the end of the sampling time from the remaining trajectory points is selected to start the next round of trajectory extraction.

[0109] The second step is to determine the target step size based on the preset speed of the observed aircraft and the observation error. For example, the target step size for the aircraft can be 0.02 degrees. It should be understood that when the target step size is in degrees, coordinate transformation can be performed to obtain the corresponding actual distance.

[0110] The third step: Determine the next trajectory point within the target selection area: Due to the interweaving of multiple trajectory points and large positioning errors, there are often multiple selectable trajectory points within the target selection area centered on the current query point. These trajectory points may come from other aircraft or trajectory points spaced apart by multiple sampling times. However, based on the principle of minimizing the time gradient between adjacent trajectory points on the same motion trajectory, the next trajectory point can be selected, thus obtaining the entire motion trajectory.

[0111] According to the object motion trajectory extraction method of this application, for example... Figure 9 Extracting multiple trajectory points as shown, we obtain the following: Figure 11 The motion trajectory extraction results are shown. Results obtained using traditional methods are as follows... Figure 12 As shown in Table 1, the false association rates of the two methods are compared. Table 1 shows that this application can accurately separate the motion trajectories of each target object from multiple trajectory points with overlapping time sequences, and the false association rate is significantly lower than that of traditional methods. The false association rate is the ratio of falsely associated trajectory points to the total number of trajectory points in the trajectory tracking result.

[0112] Table 1 Comparison of false association rates between the two methods

[0113]

[0114] This embodiment also provides an object motion trajectory extraction device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as already described. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0115] This embodiment provides a device for extracting the trajectory of an object's motion, such as... Figure 13 As shown, the device includes an acquisition module 1310, a first determination module 1320, a second determination module 1330, a third determination module 1340, and a fourth determination module 1350.

[0116] The acquisition module 1310 is configured to acquire a plurality of trajectory points, the plurality of trajectory points being different in sampling time and corresponding to a plurality of target objects.

[0117] The first determination module 1320 is configured to determine, from the plurality of trajectory points, a trajectory endpoint corresponding to the first target object by using the sampling time of the plurality of trajectory points.

[0118] The second determination module 1330 is configured to take the trajectory endpoint as a current query point and determine a target screening region centered on the current query point.

[0119] The third determination module 1340 is configured to determine, based on a size relationship between the sampling time of the trajectory point in the target screening region and the sampling time of the current query point, a next trajectory point belonging to the first target object.

[0120] The fourth determination module 1350 is configured to update the next trajectory point as the current query point, determine the trajectory point belonging to the first target object, and obtain a motion trajectory of the first target object.

[0121] In some optional embodiments, the trajectory endpoint is a trajectory starting point or a trajectory ending point; the first determination module is further configured to, if the trajectory endpoint is the trajectory starting point, determine, as the trajectory endpoint, a trajectory point with the earliest sampling time; or if the trajectory endpoint is the trajectory ending point, determine, as the trajectory endpoint, a trajectory point with the latest sampling time.

[0122] In some optional embodiments, the second determination module is further configured to acquire a preset motion speed of the first target object; determine a target step length by using an attribute feature of the preset motion speed; and diffuse the target step length outward from the current query point to determine the target screening region.

[0123] In some optional embodiments, the third determination module is further configured to determine a plurality of time difference values between the sampling time of the trajectory point in the target screening region and the sampling time of the current query point, obtain the plurality of time difference values; determine, as a candidate trajectory point, a trajectory point in the target screening region corresponding to a smallest time difference value; and determine, by using a time gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point, the next trajectory point belonging to the first target object.

[0124] In some optional embodiments, the trajectory endpoint is a trajectory starting point or a trajectory ending point; the third determination module is further configured to, if the trajectory endpoint is the trajectory starting point and the sampling time of the candidate trajectory point is later than the sampling time of the current query point, determine, as the next trajectory point belonging to the first target object, the candidate trajectory point; or if the trajectory endpoint is the trajectory ending point and the sampling time of the candidate trajectory point is earlier than the sampling time of the current query point, determine, as the next trajectory point belonging to the first target object, the candidate trajectory point.

[0125] In some optional embodiments, the trajectory end point is a trajectory start point or a trajectory end point; the fourth determination module is further configured to update the next trajectory point as the current query point, and continue to determine the target screening region with the current query point as the center; if the trajectory end point is the trajectory start point, and there is a sampling time point of the trajectory point in the target screening region that is earlier than the sampling time point of the current query point, the current query point is determined as the trajectory end point, and the motion trajectory of the first target object is obtained; if the trajectory end point is the trajectory end point, and there is a sampling time point of the trajectory point in the target screening region that is later than the sampling time point of the current query point, the current query point is determined as the trajectory start point, and the motion trajectory of the first target object is obtained.

[0126] In some optional embodiments, the device further comprises a fifth determination module, a sixth determination module and a seventh determination module, wherein the fifth determination module is configured to determine whether there are a plurality of remaining trajectory points in the plurality of trajectory points, the remaining trajectory points being trajectory points other than the trajectory points in the motion trajectory of the first target object; the sixth determination module is configured to, if there are a plurality of remaining trajectory points, determine the trajectory end point of the second target object from the plurality of remaining trajectory points by using the sampling time points of the plurality of remaining trajectory points; and the seventh determination module is configured to update the second target object as the first target object, and enter the step of taking the trajectory end point as the current query point and determining the target screening region with the current query point as the center.

[0127] The device for extracting a motion trajectory of an object provided in the application determines a trajectory end point corresponding to a first target object from a plurality of trajectory points according to sampling time points of the plurality of trajectory points; takes the trajectory end point as a current query point, and determines a target screening region with the current query point as the center; for a trajectory point in the target screening region, accurately screens a next trajectory point of the first target object by using a size relationship between the sampling time point of the trajectory point and the sampling time point of the current query point; and then updates the next trajectory point as the current query point to continue to determine the target screening region, i.e., continues to determine the target screening region with the next trajectory point as the center, iteratively determines all trajectory points belonging to the first target object from the plurality of trajectory points, and thus obtains a motion trajectory of the first target object. The scheme screens the trajectory points belonging to the first target object by using the target screening region, effectively and accurately extracts the trajectory of the target object, can reduce the calculation amount of target trajectory tracking and improve real-time performance, and thus solves the problem that the trajectories of the target objects cannot be accurately and effectively extracted.

[0128] Further function descriptions of the above-mentioned modules and units are the same as those of the above-mentioned corresponding embodiments, which will not be described herein again.

[0129] The object trajectory extraction device in the embodiment is in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory executing one or more software or fixed programs, and / or other devices that can provide the above functions.

[0130] The embodiment of the present application also provides a computer device, which refers to Figure 14 , Figure 14 is a structural schematic diagram of a computer device provided by an optional embodiment of the present application, as shown in Figure 14 The computer device comprises one or more processors 1410, a memory 1420, and an interface for connecting components, including a high-speed interface and a low-speed interface. The components are communicatively connected by using different buses, and can be mounted on a common mainboard or mounted in other manners as needed. The processor can process instructions executed in the computer device, including instructions stored in the memory or the memory to display graphical information of a GUI on an external input / output device, such as a display device coupled to the interface. In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 14 The processor 1410 is taken as an example in the embodiment.

[0131] The processor 1410 can be a central processor, a network processor, or a combination thereof. The processor 1410 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a general array logic, or any combination thereof.

[0132] The memory 1420 stores instructions executable by the at least one processor 1410, so that the at least one processor 1410 executes the method shown in the above embodiment.

[0133] The memory 1420 can include a program storage area and a data storage area. The program storage area can store an operating system, application programs, and / or data required by at least one function. The data storage area can store data created by the computer device, etc. Additionally, the memory 1420 can include a volatile memory, such as a random access memory, and can also include a non-volatile memory, such as at least one disk memory device, a flash memory device, or other non-volatile solid state memory device. In some alternative embodiments, the memory 1420 can optionally include memory that is remote from the processor 1410, which can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a wide area network, a mobile communication network, and combinations thereof.

[0134] The memory 1420 can include a volatile memory, such as a random access memory, and can also include a non-volatile memory, such as at least one disk memory device, a flash memory device, or other non-volatile solid state memory device. In some alternative embodiments, the memory 1420 can optionally include memory that is remote from the processor 1410, which can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a wide area network, a mobile communication network, and combinations thereof.

[0135] The computer device also includes an input device 1430 and an output device 1440. The processor 1410, the memory 1420, the input device 1430, and the output device 1440 can be connected through a bus or other means, Figure 14 The connection through the bus is taken as an example.

[0136] The input device 1430 can receive input digital or character information, and generate key signal input related to user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 1440 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), etc. The display device includes, but is not limited to, a liquid crystal display, a light emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0137] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded through a network and stored in a local storage medium, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special hardware. The storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0138] Part of the present application can be applied as a computer program product, for example, computer program instructions, when executed by a computer, through the operation of the computer, the method and / or technical solutions according to the present application can be invoked or provided. Those skilled in the art should understand that the form of computer program instructions in a computer readable medium includes but is not limited to source files, executable files, installation package files, etc. Correspondingly, the way of executing computer program instructions by a computer includes but is not limited to: the computer directly executes the instructions, or the computer executes the corresponding compiled program after compiling the instructions, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer readable medium can be any available computer readable storage medium or communication medium accessible to the computer.

[0139] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A method for extracting the trajectory of an object, characterized in that, include: Multiple trajectory points are acquired, and the sampling times of the multiple trajectory points are different and they correspond to multiple target objects; By utilizing the sampling times of multiple trajectory points, the trajectory endpoint corresponding to the first target object is determined from the multiple trajectory points; The trajectory endpoint is taken as the current query point, and the target filtering area is determined with the current query point as the center. Based on the relationship between the sampling time of the trajectory points within the target filtering area and the sampling time of the current query point, the next trajectory point belonging to the first target object is determined; Update the next trajectory point to the current query point, determine the trajectory point belonging to the first target object, and obtain the motion trajectory of the first target object; Based on the relationship between the sampling time of trajectory points within the target filtering area and the sampling time of the current query point, the next trajectory point belonging to the first target object is determined, including: Determine the time difference between the sampling time of the trajectory points within the target filtering area and the sampling time of the current query point to obtain multiple time differences; The trajectory points within the target screening area corresponding to the smallest time difference are determined as candidate trajectory points; By utilizing the temporal gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point, the next trajectory point belonging to the first target object is determined.

2. The method according to claim 1, characterized in that, The trajectory endpoints are either the trajectory start point or the trajectory end point; Determining the trajectory endpoint corresponding to the first target object from the plurality of trajectory points includes: If the trajectory endpoint is the trajectory starting point, then the trajectory point with the earliest sampling time is determined as the trajectory endpoint; If the trajectory endpoint is the trajectory termination point, then the trajectory point with the latest sampling time is determined as the trajectory endpoint.

3. The method according to claim 1, characterized in that, Determine the target filtering area centered on the current query point, including: Obtain the preset motion speed of the first target object; The target step length is determined by utilizing the attribute characteristics of the preset motion speed; The target filtering area is determined by expanding the target step size outward from the current query point.

4. The method according to claim 1, characterized in that, The trajectory endpoints are either the trajectory start point or the trajectory end point; Determining the next trajectory point belonging to the first target object by utilizing the temporal gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point includes: If the trajectory endpoint is the trajectory starting point, and the sampling time of the candidate trajectory point is later than the sampling time of the current query point, then the candidate trajectory point is determined as the next trajectory point of the first target object; If the trajectory endpoint is the trajectory termination point, and the sampling time of the candidate trajectory point is earlier than the sampling time of the current query point, then the candidate trajectory point is determined as the next trajectory point of the first target object.

5. The method according to claim 1, characterized in that, The trajectory endpoints are either the trajectory start point or the trajectory end point; Update the next trajectory point to the current query point, determine the trajectory points belonging to the first target object, and obtain the motion trajectory of the first target object, including: Update the next trajectory point to the current query point, and continue to determine the target filtering area with the current query point as the center; If the trajectory endpoint is the trajectory start point, and the sampling time of all trajectory points within the target filtering area is earlier than the sampling time of the current query point, then the current query point is determined as the trajectory end point, and the motion trajectory of the first target object is obtained. If the trajectory endpoint is the trajectory termination point, and the sampling time of all trajectory points within the target filtering area is later than the sampling time of the current query point, then the current query point is determined as the trajectory starting point, and the motion trajectory of the first target object is obtained.

6. The method according to claim 1, characterized in that, After updating the next trajectory point to the current query point, determining the trajectory point belonging to the first target object, and obtaining the motion trajectory of the first target object, the method further includes: Determine whether there are multiple remaining trajectory points among the multiple trajectory points, wherein the remaining trajectory points are the trajectory points other than those on the motion trajectory of the first target object; If there are multiple remaining trajectory points, the trajectory endpoint of the second target object is determined from the multiple remaining trajectory points by using the sampling time of the multiple remaining trajectory points; The second target object is updated to the first target object, and the process proceeds to the step of using the trajectory endpoint as the current query point and determining the target filtering area centered on the current query point.

7. A device for extracting the trajectory of an object, characterized in that, include: The acquisition module is used to acquire multiple trajectory points, wherein the sampling time of the multiple trajectory points is different and they correspond to multiple target objects; The first determining module is used to determine the trajectory endpoint corresponding to the first target object from the multiple trajectory points by utilizing the sampling time of the multiple trajectory points; The second determining module is used to take the trajectory endpoint as the current query point and determine the target filtering area with the current query point as the center. The third determining module is used to determine the next trajectory point belonging to the first target object based on the relationship between the sampling time of the trajectory points in the target filtering area and the sampling time of the current query point. The fourth determining module is used to update the next trajectory point to the current query point, determine the trajectory point belonging to the first target object, and obtain the motion trajectory of the first target object; The third determining module is further configured to determine the time difference between the sampling time of the trajectory point in the target filtering area and the sampling time of the current query point, and obtain multiple time differences; determine the trajectory point in the target filtering area corresponding to the smallest time difference as a candidate trajectory point; and determine the next trajectory point belonging to the first target object by using the time gradient direction between the sampling time of the candidate trajectory point and the sampling time of the current query point.

8. A computer device, characterized in that, include: The device includes a memory and a processor, which are interconnected and the memory stores computer instructions. The processor executes the computer instructions to perform the method for extracting the motion trajectory of an object as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to execute the method for extracting the motion trajectory of an object as described in any one of claims 1 to 6.

Images