Trajectory point state recognition method and device, electronic equipment, storage medium and product

Abstract

The application relates to the technical field of trajectory data mining, and particularly discloses a trajectory point state recognition method and device, electronic equipment, a storage medium and a product. The method comprises the following steps: performing compression processing on acquired original trajectory data to obtain target trajectory data of different target objects; performing trajectory analysis on the target trajectory data of each target object according to a preset space-time density constraint condition to generate trajectory point state recognition results of each target object; and the preset space-time density constraint condition comprises a space constraint based on a preset stay radius and a time constraint based on a preset stay time length. According to the scheme, the original trajectory data is compressed, high-quality data input can be provided for trajectory analysis; the space-time density constraint condition combining the space constraint based on the preset stay radius and the time constraint based on the preset stay time length can accurately recognize stay points and motion points in the trajectory data, and the precision and efficiency of trajectory point state recognition are improved.

Description

Technical Field

[0001] This invention relates to the field of trajectory data mining technology, and in particular to a method, apparatus, electronic device, storage medium and product for trajectory point state recognition. Background Technology

[0002] Trajectory data processing refers to the process of collecting and analyzing location information recorded by positioning devices to achieve functions such as path tracking, behavior analysis, and speed calculation for moving objects or people. It is widely used in fields such as traffic management, logistics tracking, and personal health monitoring.

[0003] Most existing trajectory point recognition methods employ density clustering algorithms to group temporally continuous trajectory points that meet preset conditions into a single class, and then define the resulting clusters as the dwell areas of the trajectory data to be processed. However, these methods do not effectively utilize the spatial information of the trajectory data and cannot accurately identify dwell points within the trajectory, resulting in poor trajectory point status recognition performance. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, storage medium, and product for trajectory point state recognition, in order to solve the problem of poor trajectory point state recognition performance in the prior art.

[0005] According to one aspect of the present invention, a trajectory point state recognition method is provided, the method comprising:

[0006] The acquired raw trajectory data is compressed to obtain target trajectory data for different target objects;

[0007] Based on preset spatiotemporal density constraints, trajectory analysis is performed on the target trajectory data of each target object to generate trajectory point status recognition results for each target object; among which, preset spatiotemporal density constraints include spatial constraints based on preset dwell radius and temporal constraints based on preset dwell duration.

[0008] According to another aspect of the present invention, a trajectory point state recognition device is provided, the device comprising:

[0009] The compression processing module is used to compress the acquired raw trajectory data to obtain target trajectory data for different target objects;

[0010] The trajectory analysis module is used to perform trajectory analysis on the target trajectory data of each target object according to preset spatiotemporal density constraints, and generate trajectory point status recognition results for each target object; wherein, the preset spatiotemporal density constraints include spatial constraints based on preset dwell radius and temporal constraints based on preset dwell duration.

[0011] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0012] At least one processor; and

[0013] A memory communicatively connected to the at least one processor; wherein,

[0014] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the trajectory point state recognition method according to any embodiment of the present invention.

[0015] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the trajectory point state recognition method according to any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the trajectory point state recognition method according to any embodiment of the present invention.

[0017] The technical solution of this invention compresses the acquired raw trajectory data to obtain target trajectory data for different target objects; it then performs trajectory analysis on the target trajectory data of each target object according to preset spatiotemporal density constraints to generate trajectory point state recognition results for each target object; wherein, the preset spatiotemporal density constraints include spatial constraints based on a preset dwell radius and temporal constraints based on a preset dwell duration. This technical solution, by compressing the raw trajectory data, effectively reduces redundant information and provides high-quality data input for trajectory analysis; by employing spatiotemporal density constraints combining spatial constraints based on a preset dwell radius and temporal constraints based on a preset dwell duration, it can accurately identify dwell points and movement points in the trajectory data, improving the accuracy and efficiency of trajectory point state recognition.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a trajectory point state recognition method provided in Embodiment 1 of the present invention;

[0021] Figure 2 This is a flowchart of a trajectory point state recognition method provided in Embodiment 2 of the present invention;

[0022] Figure 3 This is a flowchart of the data compression process provided in Embodiment 2 of the present invention;

[0023] Figure 4 This is a flowchart of the trajectory analysis process provided in Embodiment 2 of the present invention;

[0024] Figure 5 This is a flowchart of a trajectory point state recognition method provided in Embodiment 3 of the present invention;

[0025] Figure 6 This is a flowchart of another trajectory point state recognition method provided in Embodiment 3 of the present invention;

[0026] Figure 7 This is a schematic diagram of the structure of a trajectory point state recognition device according to Embodiment 4 of the present invention;

[0027] Figure 8 This is a schematic diagram of the structure of an electronic device that implements the trajectory point state recognition method of the present invention. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Example 1

[0031] Figure 1 This is a flowchart of a trajectory point state recognition method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations requiring precise motion state recognition of trajectory data of a target object. The method can be executed by a trajectory point state recognition device, which can be implemented in hardware and / or software. This trajectory point state recognition device can be configured in an electronic device, which may include, but is not limited to, a computer, a server, etc. Figure 1 As shown in the figure, the trajectory point state recognition method provided in this embodiment includes the following steps:

[0032] S110. Compress the acquired raw trajectory data to obtain target trajectory data for different target objects.

[0033] The raw trajectory data can refer to a dataset containing numerous movement trajectories of target objects, collected using a Global Positioning System (GPS) device. This dataset contains a large number of trajectory points, each of which may include a timestamp 't' and longitude. ,latitude The target object includes attribute information such as the target object identifier (ID). The target object can refer to the subject that generates the trajectory, such as a vehicle, person, equipment, or animal. Each target object has a unique target object identifier (such as device ID, user ID, etc.) to distinguish the trajectories of different subjects.

[0034] Compression processing can refer to the process of simplifying and purifying the original trajectory data. The purpose is to reduce the amount of data, remove noise, and output high-quality trajectory data suitable for subsequent trajectory analysis. For example, compression processing can include, but is not limited to, operations such as object grouping, time window extraction, latitude and longitude correction, merging of duplicate points, and removal of drift points.

[0035] In this embodiment of the invention, the large-scale raw trajectory data can be first grouped according to the target object identifier of the trajectory points to obtain the initial trajectory data corresponding to each target object. Then, data compression operations such as spatiotemporal range filtering, duplicate point merging, and drift point removal are performed on the initial trajectory data of each target object. This reduces the amount of initial trajectory data while retaining key motion features, and outputs high-quality target trajectory data corresponding to each target object.

[0036] S120. Perform trajectory analysis on the target trajectory data of each target object according to the preset spatiotemporal density constraints, and generate the trajectory point status recognition results of each target object; wherein, the preset spatiotemporal density constraints include spatial constraints based on preset dwell radius and temporal constraints based on preset dwell time.

[0037] Among them, the preset spatiotemporal density constraint condition can refer to the dual constraint rule used to determine the state of trajectory points (movement). It consists of two parts: a spatial constraint based on the preset dwell radius and a temporal constraint based on the preset dwell time. It is the core judgment basis for trajectory analysis.

[0038] The preset dwell radius (spatial constraint) can refer to a pre-configured distance threshold (such as 250 meters) used to define spatial aggregation. It represents the activity of the target object within the dwell radius as a spatial aggregation state and is a spatial standard for judging the state of trajectory points.

[0039] The preset dwell time (time constraint) can refer to a pre-configured time threshold (such as 20 minutes) used to define the duration of time. It represents the time that the target object must spend in a spatially clustered state to reach the time threshold in order to constitute a "dwelling". It is the time standard for judging the state of trajectory points.

[0040] Trajectory analysis refers to the process of identifying the status of trajectory points based on target trajectory data. Specifically, it may include point cluster aggregation, spatiotemporal constraint judgment, and status identification. The trajectory point status identification result can be the final output of trajectory analysis on the target trajectory results. It can be a sequence of point clusters arranged in chronological order, with each cluster labeled as a stationary cluster or a moving cluster, and containing information such as the cluster's start time, end time, center point latitude and longitude, and the number of merged trajectory points. This information visually reflects the movement and stationary patterns of the target object.

[0041] In this embodiment of the invention, to improve the overall processing efficiency of trajectory data, parallel computing can be used to simultaneously execute corresponding trajectory analysis operations for target trajectory data of different target objects, such as through multi-threading or distributed computing. The trajectory analysis process for a single target object is described below: The target trajectory data is traversed, and the trajectory points in the target trajectory data are clustered in chronological order. The state (stationary or moving) of the point clusters is determined by preset spatiotemporal density constraints. Spatially, it is checked whether all trajectory points within a point cluster are located within a circular area centered on the cluster's center point and with a preset stationary radius (spatial constraint). Temporally, the duration of the point cluster from the first to the last trajectory point is calculated to determine if it reaches a preset stationary duration (temporal constraint). If both spatial and temporal constraints are met, the point cluster is a stationary point cluster; otherwise, it is a moving point cluster. After traversing all trajectory points and completing point cluster identification, adjacent stationary point clusters with close spatial distances can be further merged to ultimately form a point cluster sequence reflecting the stationary and moving states of the target object. In simple terms, the goal of point clustering is to identify the largest point cluster within a preset dwell radius through clustering, and the largest point cluster must also meet the time constraint based on a preset dwell time.

[0042] The technical solution of this invention compresses the acquired raw trajectory data to obtain target trajectory data for different target objects; it then performs trajectory analysis on the target trajectory data of each target object according to preset spatiotemporal density constraints to generate trajectory point state recognition results for each target object; wherein, the preset spatiotemporal density constraints include spatial constraints based on a preset dwell radius and temporal constraints based on a preset dwell duration. This technical solution, by compressing the raw trajectory data, effectively reduces redundant information and provides high-quality data input for trajectory analysis; by employing spatiotemporal density constraints combining spatial constraints based on a preset dwell radius and temporal constraints based on a preset dwell duration, it can accurately identify dwell points and movement points in the trajectory data, improving the accuracy and efficiency of trajectory point state recognition.

[0043] Example 2

[0044] Figure 2 This is a flowchart of a trajectory point state recognition method provided in Embodiment 2 of the present invention. It is further optimized and extended based on the above embodiments and can be combined with various optional technical solutions in the above embodiments. For example... Figure 2 As shown in the figure, the trajectory point state recognition method provided in this embodiment includes the following steps:

[0045] S210. Extract the target object identifier of each trajectory point in the original trajectory data.

[0046] In this embodiment of the invention, after obtaining a large amount of raw trajectory data, the target object identifier contained in each trajectory point can be extracted, such as device ID, user ID, vehicle ID, etc.

[0047] S220. Divide the original trajectory data into initial independent trajectory data corresponding to different target objects according to the target object identifier.

[0048] The initial independent trajectory data can refer to the data set consisting of all the original trajectory points of a single target object after being divided according to the target object, which has not yet undergone data compression processing.

[0049] In this embodiment of the invention, the original trajectory data can be divided into several groups of initial independent trajectory data according to the target object identifier of each trajectory point in the original trajectory data. Each group of initial independent trajectory data contains only all the original trajectory points of a single target object. For example, all the original trajectory points of vehicle A are classified as initial independent trajectory data 1, all the original trajectory points of user B are classified as initial independent trajectory data 2, etc., to ensure that the trajectory data of different target objects are not confused.

[0050] S230. Perform data compression operations in parallel on each initial independent trajectory data to obtain the corresponding target trajectory data.

[0051] In this embodiment of the invention, data compression operations can be performed in parallel on each of the initial independent trajectory data after division. For example, the data compression tasks of different target objects can be assigned to independent computing nodes, processor cores or threads for simultaneous processing, and finally, simplified target trajectory data corresponding to each target can be generated.

[0052] Furthermore, based on the above embodiments of the invention, such as Figure 3 As shown, in step S230, a data compression operation is performed on the initial independent trajectory data to obtain the corresponding target trajectory data. This specifically includes the following steps:

[0053] S11. According to the preset time window and preset latitude and longitude coordinate range, the initial independent trajectory data is filtered for trajectory points to obtain the first trajectory data;

[0054] S12. Arrange the trajectory points in the first trajectory data in chronological order, merge trajectory points with consecutive latitude and longitude coordinates and retain the earliest timestamp to obtain the second trajectory data.

[0055] S13. Traverse the second trajectory data, determine the first distance between the current trajectory point and the previous trajectory point, and the second distance between the previous trajectory point and the next trajectory point of the current trajectory point. If the first distance and the second distance meet the preset drift point determination conditions, then determine the current trajectory point as a drift point and remove it.

[0056] S14. The second trajectory data after the drift points are removed is used as the third trajectory data, and the trajectory points with the same longitude and latitude coordinates in the third trajectory data are merged to obtain the target trajectory data.

[0057] The preset time window refers to an effective time range set according to business needs or the activity patterns of the target object. It is used to filter trajectory points in the trajectory data that belong to the effective time period and eliminate redundant points in irrelevant time periods. In one embodiment, the preset time window can be set to a length of 24 hours, covering the activity troughs of most target objects. For example, the preset time window can be selected from 04:00 to 04:00 the next day. It should be understood that in practical applications, the length and duration of the preset time window can be set according to actual business needs or the activity patterns of the target object; this embodiment does not impose specific limitations on this.

[0058] The preset latitude and longitude coordinate range can refer to the effective geographical boundary set based on the actual trajectory analysis scenario, used to filter out abnormal positioning points in the trajectory data that exceed this range. For example, the preset latitude and longitude coordinate range can be the latitude and longitude coordinate range of a country, a city, a region, etc.

[0059] Drift points can refer to abnormal trajectory points caused by signal noise interference, equipment failure, etc. They are characterized by a sudden and significant change in distance from the preceding and following trajectory points (such as the distance between two adjacent points suddenly changing from 10 meters to 1000 meters), which will interfere with the continuity of the trajectory and therefore need to be removed.

[0060] The preset drift point determination condition can refer to the determination condition used to identify drift points. It is based on the logic setting that the trajectory point will not suddenly change drastically during normal movement, and is used to distinguish abnormal drift points from normal movement points. For example, the preset drift point determination condition can include at least: whether the first distance is greater than a preset multiple (such as 3 times) of the second distance, wherein the first distance can refer to the distance between the current trajectory point and the previous trajectory point, and the second distance can refer to the distance between two adjacent trajectory points corresponding to the current trajectory point.

[0061] The first trajectory data can refer to the trajectory data obtained after filtering the initial independent trajectory data in terms of spatiotemporal range. This spatiotemporal range filtering includes time window extraction and latitude / longitude correction. The second trajectory data can refer to the trajectory data obtained after merging duplicate points in the first trajectory data. The third trajectory data can refer to the trajectory data obtained after removing drift points from the second trajectory data.

[0062] In this embodiment of the invention, the specific process of performing data compression on initial independent trajectory data to obtain target trajectory data includes:

[0063] ① Spatiotemporal range filtering: Obtain a pre-configured preset time window and preset latitude and longitude coordinate range. Based on the preset time window and preset latitude and longitude coordinate range, filter the initial independent trajectory data, retain only trajectory points whose timestamps are within the preset time window and whose latitude and longitude coordinates are within the preset latitude and longitude coordinate range, and remove invalid trajectory points outside the time window or whose latitude and longitude coordinates are out of range to obtain the first trajectory data.

[0064] ② First-time deduplication: Sort the trajectory points in the first trajectory data in chronological order. Traverse the sorted trajectory points. If there are multiple consecutive trajectory points with the same latitude and longitude coordinates, for example, trajectory points... and If the latitude and longitude coordinates are the same, only the trajectory point with the earliest timestamp will be retained. This yields the second trajectory data after deduplication.

[0065] ③ Drift point removal: Traverse the second trajectory data, for every three consecutive trajectory points: the previous trajectory point Current trajectory point and the next trajectory point Perform distance analysis, that is, calculate the current trajectory points respectively. Compared with the previous trajectory point The first distance between and the previous trajectory point Between the next trajectory point The second distance Determine the first distance Second distance Does the preset drift point determination condition meet? If it does, then determine the current trajectory point. The drift points are identified and removed. After removing all drift points from the second trajectory data, the third trajectory data is obtained. The preset drift point determination condition can be: whether the first distance is greater than three times the second distance. .

[0066] In one embodiment, the first distance Second distance We can use the squared Euclidean distance, that is:

[0067]

[0068]

[0069] ④ Secondary deduplication: Since new adjacent duplicate points may appear after the drift points are removed, it is necessary to perform deduplication again on the third trajectory data to obtain the final target trajectory data; the deduplication process is the same as step ② above, and will not be repeated here.

[0070] This embodiment effectively reduces redundant or noisy data (such as repeatedly reported points at the same location, drift points with incorrect positioning, etc.) in the initial independent trajectory data of the target object by performing spatiotemporal range filtering, drift point removal, and two deduplication data compression operations. This significantly reduces the computational load of subsequent trajectory analysis (such as point cluster aggregation and state recognition), thereby improving the overall processing efficiency.

[0071] S240. Perform trajectory point state recognition operations in parallel on each target trajectory data to generate corresponding trajectory point state recognition results; wherein, the trajectory point state recognition operation is used to identify trajectory points in the target trajectory data as stationary points or moving points based on preset spatiotemporal density constraints.

[0072] In this embodiment of the invention, for the target trajectory data of each target object, the corresponding trajectory point state recognition operation can be executed simultaneously by allocating independent processing threads or computing nodes. The processing of different target objects is independent of each other, does not occupy computing resources, and completes the trajectory point (motion) state recognition task only within its own computing unit, and finally outputs the trajectory point state recognition result corresponding to each target object.

[0073] Furthermore, based on the above embodiments of the invention, such as Figure 4 As shown, S240 performs trajectory point state recognition operation on the target trajectory data to generate the corresponding trajectory point state recognition result, which specifically includes the following steps:

[0074] S21. Initialize the first trajectory point in the target trajectory data as the center point of the current point cluster;

[0075] S22. Traverse subsequent trajectory points in chronological order and determine whether all trajectory points in the current point cluster are within the target circular area with the center point as the center and the preset dwell radius as the radius.

[0076] S23. If all trajectory points are within the target circular area and there are undetected trajectory points in the target trajectory data, then merge the current trajectory points into the current point cluster, update the center point of the current point cluster, and then return to the step of traversing the subsequent trajectory points in chronological order to determine whether all trajectory points in the current point cluster are within the target circular area with the center point as the center and the preset dwell radius as the radius.

[0077] S24. If all trajectory points are within the target circular area and there are no undetected trajectory points in the target trajectory data, then mark the current point cluster as a stationary point cluster and record the point cluster information;

[0078] S25. If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is longer than the preset dwell time, then the current point cluster after removing the last trajectory point is marked as a dwell point cluster, and the point cluster information is recorded. Then, the last trajectory point is used as the center point of the new current point cluster to re-execute the point cluster aggregation process.

[0079] S26. If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is less than or equal to the preset dwell time, then the first trajectory point in the current point cluster is marked as a moving point cluster, and the point cluster information is recorded. Then, the second trajectory point in the current point cluster is used as the center point of the new current point cluster to re-execute the point cluster aggregation process.

[0080] S27. After all the target trajectory data has been traversed, determine the cumulative spherical distance between all moving point clusters between adjacent stopping point clusters in the obtained point cluster sequence. If the cumulative spherical distance is less than the preset distance threshold, merge the two stopping point clusters and all moving point clusters between the two stopping point clusters into a new stopping point cluster, and update the point cluster information of the new stopping point cluster.

[0081] S28. Determine the end time of each point cluster in the point cluster sequence to update the point cluster information, and use the updated point cluster sequence as the trajectory point state recognition result of the target object.

[0082] In this embodiment of the invention, the trajectory analysis process for a single target object specifically includes:

[0083] ① Initialize and create an empty current point cluster, and set the first trajectory point in the target trajectory data as the initial center point (latitude and longitude coordinates) of the current point cluster.

[0084] ② Process other trajectory points in the target trajectory data in chronological order. For each current trajectory point to be processed, first determine whether all the trajectory points already included in the current point cluster are located within the target circular area with the center point of the current point cluster as the center and the preset dwell radius as the radius. That is, determine whether all trajectory points in the current point cluster meet the spatial constraints based on the preset dwell radius.

[0085] In one embodiment, determining whether a trajectory point is within the target circular area may specifically include:

[0086] Determine the cosine of the geocentric angle between the trajectory point and the center point of the current point cluster;

[0087] If the cosine value of the geocentric angle is greater than the preset threshold value of the cosine value of the geocentric angle, the trajectory point is determined to be within the target circular area; otherwise, the trajectory point is determined to be outside the target circular area.

[0088] The cosine of the geocentric angle can be the cosine of the angle (geocene angle) formed by the lines connecting the Earth's center to two trajectory points (the trajectory point to be determined and the center point of the point cluster). It is an indirect indicator of the spherical distance between two points on the Earth's surface. The larger the value, the closer the two points are.

[0089] Specifically, the latitude and longitude coordinates of the trajectory point to be judged and the center point of the current point cluster can be extracted. Then, based on the latitude and longitude coordinates of the two points, the cosine value of the geocentric angle between them can be calculated through spherical geometry. For example, for the trajectory point to be judged... and the current cluster center point The cosine of the geocentric angle between them can be expressed as:

[0090]

[0091] In the formula, Represents trajectory points and the current cluster center point The geocentric angle between them.

[0092] Then, the calculated cosine value of the geocentric angle is used... Compared with the preset geocentric angle cosine threshold value If a comparison is made, If the spherical distance between the two points is less than or equal to the preset stopping radius, the trajectory point is determined to be within the target circular area; otherwise, the trajectory point is determined to be outside the target circular area. The preset geocentric angle cosine threshold is used. Can be based on a preset dwell radius Determined with the Earth's radius R, that is .

[0093] It should be understood that in this embodiment, the cosine of the geocentric angle between the trajectory point to be judged and the center point of the current point cluster is used to determine whether the trajectory point satisfies the spatial constraint based on the preset dwell radius. In practical applications, the spherical distance between the two can also be used for spatial constraint judgment, and the spherical distance between the two can be expressed as... If the distance of the sphere is Less than or equal to the preset spherical distance threshold If the coordinates of the geocentric angle are within the target circular area, then the trajectory point is determined to be within the target circular area; otherwise, the trajectory point is determined to be outside the target circular area. As can be seen from the formula for calculating spherical distance, the calculation of the cosine of the geocentric angle is actually an intermediate step in the calculation of spherical distance. This embodiment directly uses the cosine of the geocentric angle for spatial constraint judgment, which can reduce the consumption of computing resources and significantly improve computing efficiency.

[0094] ③ If all trajectory points within the current point cluster are within the target circular area (i.e., satisfying the spatial constraints), and there are still unprocessed trajectory points in the target trajectory data, then merge the next undetected trajectory point (i.e., the current trajectory point) into the current point cluster, recalculate the center point of the current point cluster, and then return to step S22 to continue determining whether the updated current point cluster satisfies the spatial constraints. The updated center point... It can be represented as: , In the formula, n represents the number of trajectory points within the current point cluster.

[0095] ④ If all trajectory points within the current point cluster are within the target circular area (i.e., spatial constraints are satisfied), and there are no undetected trajectory points in the target trajectory data (i.e., the target trajectory data has been traversed to the end), then the current point cluster is marked as a stopping point cluster (indicating that the target object stops in this area), and the phase point cluster information of this point cluster is recorded, which may include, for example, the following tuple information. ,in, This represents the start time when the j-th point cluster stops at the center point of the cluster, which is also the timestamp corresponding to the first trajectory point in the cluster. and Let these represent the longitude and latitude of the center point of the j-th point cluster, respectively. This represents the number of merged trajectory points contained in the j-th point cluster; This represents the motion state of the j-th point cluster, used to identify the current point cluster. Is it a cluster of resting points or a cluster of moving points, and Indicates a stationary state. Indicates a state of motion.

[0096] ⑤ If at least one trajectory point in the current point cluster is not within the target circular area (i.e., does not meet the spatial constraints), and the duration of the current point cluster (the time difference between the first and last trajectory points in the cluster) is greater than the preset dwell time (i.e., meets the time constraints), then the point cluster after removing the last trajectory point (i.e., the last trajectory point in the current point cluster) is marked as a dwell point cluster and the relevant point cluster information is recorded. Then, the last trajectory point is used as the center point of the new current point cluster, and the point cluster aggregation process is restarted, i.e., returning to execute S22.

[0097] ⑥ If at least one trajectory point in the current point cluster is not within the target circular area (i.e., spatial constraints are not met), and the duration of the current point cluster is less than or equal to the preset dwell time (i.e., time constraints are not met), then the first trajectory point in the current point cluster is marked as a moving point cluster (indicating that the target object is in a moving state) and the relevant point cluster information is recorded. Then, the second trajectory point in the current point cluster is taken as the center point of the new current point cluster, and the point cluster aggregation process is restarted, i.e., return to execute S22.

[0098] ⑦ Once all target trajectory data has been traversed, a point cluster sequence containing several point clusters will be generated; then, this point cluster sequence will be examined, and for any two adjacent stationary point clusters, the cumulative spherical distance between them and all moving point clusters will be calculated. If the cumulative spherical distance Less than the preset distance threshold Then, these two stop point clusters and all intermediate motion point clusters are merged into a new stop point cluster, and the relevant cluster information of the new cluster is updated. For example, if there exists , and For two adjacent clusters of resting points, the corresponding cumulative spherical distance can be expressed as:

[0099]

[0100] After two adjacent stop point clusters are merged, the start time of the new stop point cluster is That is, the previous stop cluster The corresponding start time; the latitude and longitude of the new center point are respectively: , The new number of merged trajectory points is The new state of motion is .

[0101] By merging short-distance movements in step S27, the misjudgment of short-distance movements (such as traffic jams and creeping) is resolved, which can improve the continuity of stay.

[0102] ⑧ Add an end time to each dwell point cluster or moving point cluster in the point cluster sequence. Specifically, if there are subsequent point clusters, the end time of the current point cluster is set as the start time of the next point cluster; if there are no subsequent point clusters, i.e. the current point cluster is the last point cluster in the point cluster sequence, the end time of the current point cluster is set as the right boundary of the aforementioned preset time window; finally, the updated point cluster sequence is used as the trajectory point state recognition result of the corresponding target object.

[0103] This embodiment accurately identifies the motion state of each trajectory point through dual determination based on the spatial constraint of the preset dwell radius and the time constraint of the preset dwell duration. At the same time, parameters such as the preset dwell radius, preset dwell duration, and preset distance threshold can be adjusted according to the actual scenario. For example, a small radius / short duration is used for pedestrian trajectories, and a large radius / long duration is used for vehicle trajectories. This makes the technical solution applicable to the trajectory analysis needs of different target objects (people, vehicles, equipment, etc.).

[0104] The technical solution of this invention, by grouping large-scale raw trajectory data into objects and then performing parallel data compression operations on the initial independent trajectory data of each target object to obtain the corresponding target trajectory data, can effectively reduce redundant information, making the obtained target trajectory data more concise. This provides a high-quality, low-noise input foundation for subsequent trajectory analysis and reduces the computational resource consumption of the subsequent analysis process. By adopting a spatiotemporal density constraint condition that combines spatial constraints based on a preset dwell radius and temporal constraints based on a preset dwell duration, parallel trajectory analysis of the target trajectory data of each target object can accurately identify the dwell points and movement points in each trajectory data, improving the accuracy and efficiency of trajectory point state recognition. In addition, the data compression and trajectory analysis processes of each target object are independent of each other, avoiding the confusion of trajectory data of different objects, and can meet the needs of multi-subject (such as multi-user, multi-device, multi-vehicle) trajectory analysis scenarios, thereby improving the universality of this solution.

[0105] Example 3

[0106] Figure 5 This is a flowchart of a trajectory point state recognition method provided in Embodiment 3 of the present invention. Based on the above embodiments, this embodiment uses a vehicle as an example to provide an implementation method for trajectory point state recognition, which can quickly and accurately identify the trajectory point motion state of different target objects. Figure 5 As shown, the trajectory point state recognition method provided in Embodiment 3 of the present invention specifically includes the following steps:

[0107] S310. Obtain the raw trajectory data.

[0108] In this embodiment of the invention, the original trajectory data may contain a large number of trajectory points of target objects (i.e., different vehicles), among which there may be redundant or abnormal trajectory points. Each trajectory point contains the following attribute information: timestamp t, longitude, etc. ,latitude And the target object identifier ID.

[0109] S320. Divide the original trajectory data into initial independent trajectory data corresponding to different target objects according to the target object identifier.

[0110] S330: Assign independent computing units to each target object.

[0111] In this embodiment of the invention, each target object can be assigned an independent computing unit, such as a separate computing node, processor core or thread, and each computing unit will be used to perform data compression and trajectory analysis tasks for its respective target object.

[0112] S340: Control each computing unit to perform the corresponding data compression task according to the initial independent trajectory data of each target object, so as to obtain the target trajectory data corresponding to each target object.

[0113] In this embodiment of the invention, taking a single target object as an example, the corresponding data compression task execution process is as follows:

[0114] ① Time window extraction: Preset time windows can be used to extract initial independent trajectory data. Filtering is performed to obtain trajectory data. This includes the ability to extract data within a daily time window. The setting of the time window needs to carefully consider the region and the off-peak hours of vehicle traffic (e.g., 24-hour billing cycle plus...). The time span is 1 hour (04:00 to 04:00 the next day). If there are sufficient computing resources, the longer the time span, the better, but the deadline must be set during the off-peak hours of vehicle travel.

[0115] ② Latitude and longitude correction: The trajectory data can be corrected using a preset latitude and longitude coordinate range. After further filtering, the trajectory data is obtained. Among these, the effective coordinate span of China can be utilized to retain those that satisfy... and The trajectory points are identified, and invalid trajectory data is excluded, thus obtaining trajectory data filtered by spatiotemporal range. .

[0116] ③ First-time deduplication: Trajectory data typically contains many stationary trajectory points. By merging duplicate points beforehand, the computational workload for subsequent trajectory analysis can be significantly reduced. Specifically, this can be done by first deduplicating... All trajectory points are sorted chronologically, and then the sorted trajectory points are traversed. If multiple consecutive trajectory points have the same latitude and longitude coordinates, for example, trajectory points... and Their latitude and longitude coordinates are the same, that is... , Then only the trajectory point with the earliest timestamp is retained. This yields the trajectory data after deduplication. .

[0117] ④ Drift Point Removal: Due to signal noise, GPS data can drift. This step identifies and removes trajectory points that are abnormally far from their neighbors. Specifically, iterates through the trajectory data. For every three consecutive trajectory points: the previous trajectory point Current trajectory point and the next trajectory point Perform distance analysis, that is, calculate the current trajectory points respectively. Compared with the previous trajectory point The first square Euclidean distance between them and the previous trajectory point Between the next trajectory point The second square Euclidean distance From the perspective of plane geometry, let the previous trajectory point... Current trajectory point and the next trajectory point Given points A, B, and C respectively, a normal trajectory curve should be a smooth forward curve. If... ,Right now According to the Law of Sines, we can obtain The current trajectory point can be seen. The distance to its adjacent trajectory points is abnormally large, therefore the current trajectory point is determined to be... This step identifies and removes drift points. This process removes most drift noise at various locations, resulting in trajectory data after drift point removal. .

[0118] ⑤ Secondary deduplication: Since new adjacent duplicate points may appear after the drift points are removed, the trajectory data can be deduplicated as in step ③. Perform another deduplication operation to obtain further simplified target trajectory data. .

[0119] S350: Control each computing unit to perform the corresponding trajectory analysis task according to the target trajectory data of their respective target objects, so as to obtain the trajectory point status recognition results corresponding to each target object.

[0120] In this embodiment of the invention, taking a single target object as an example, the execution flow of its corresponding trajectory analysis task is as follows:

[0121] ① Initialize by creating an empty current point cluster and setting the first trajectory point in the target trajectory data. Set as the initial center point of the current point cluster. (Latitude and longitude coordinates).

[0122] ② Traverse subsequent trajectory points in chronological order, and determine whether all trajectory points in the current point cluster are located at the center point. Centered on the circle, with a preset dwell radius Within the target circular area with a radius of , it is determined whether all trajectory points within the current point cluster satisfy the spatial constraints based on the preset dwell radius.

[0123] This step can use the geocentric angle to approximate the spherical distance on the Earth's surface, that is, compare the points of the trajectory to be judged. and the current cluster center point The cosine of the geocentric angle between the two points is used to determine whether the trajectory point is within the target circular area, thereby improving computational efficiency and avoiding redundant calculations. Specifically, the trajectory point to be determined... and the current cluster center point The cosine of the geocentric angle between them can be expressed as: The spherical distance between the two can be expressed as Where the Earth's radius R = 6,371,000 meters; if the following conditions are met: Then determine the trajectory point. If the target circular area is within the specified spatial constraints, then the trajectory point is determined to be outside the specified circular area. It is not within the target circular area (i.e., it does not meet the spatial constraints).

[0124] ③ If the current point cluster All trajectory points within the target circular area (i.e., satisfying spatial constraints) are within the target trajectory data, and there are still unprocessed trajectory points in the target trajectory data. Then the next undetected trajectory point Merge the points into the current point cluster, recalculate the center point of the current point cluster, and then return to step ② to continue to determine whether the updated current point cluster satisfies the spatial constraints.

[0125] Among them, the current cluster center point It can be updated to the Euclidean center of the current point cluster, i.e. , .

[0126] ④ If the current point cluster If all trajectory points within the target circular area (i.e., satisfying spatial constraints) and there are no undetected trajectory points in the target trajectory data (i.e., traversing to the end of the target trajectory data), then the current point cluster is marked as a dwell point cluster, and the phase point cluster information of this point cluster, i.e., tuple information, is recorded. ;in, For the current cluster of points The start time, i.e., the point cluster The timestamp corresponding to the first trajectory point in the middle; and Each represents the current point cluster. Longitude and latitude; Indicates the current point cluster The number of merged trajectory points included; Setting it to 0 indicates the current point cluster. For a cluster of resting points.

[0127] ⑤ If the current point cluster If at least one trajectory point is outside the target circular region (i.e., spatial constraints are not met), then the duration of the current point cluster is calculated. If the duration (i.e., satisfying the time constraint), then the point cluster will be... Mark them as stop point clusters and record tuple information. ;in, Number of merged trajectory points Then, the trajectory points As the new center point of the current point cluster, restart the point cluster aggregation process, that is, return to step ②.

[0128] ⑥ If the current point cluster There exists at least one trajectory point that is not within the target circular area (i.e., does not satisfy the spatial constraints), and the duration of this current point cluster is less than or equal to the preset dwell time. (i.e., the time constraint is not met), then the current point cluster will be... The first trajectory point in This is marked as a cluster of moving points (in which there is actually only one moving point). ), and record tuple information ;in, Number of merged trajectory points Then, the current cluster of points... The second trajectory point As the new center point of the current point cluster, restart the point cluster aggregation process, that is, return to step ②.

[0129] ⑦ Merge brief movements: Repeat steps ② to ⑥ until all trajectory points have been traversed, generating a point cluster sequence containing several point clusters. Then, for two adjacent stop point clusters in the point cluster sequence, for example, if there exists , and For two adjacent stop clusters ( (For a continuous set of moving points), calculate the cumulative spherical distance between all moving point sets. :

[0130]

[0131] If the cumulative spherical distance Less than the preset distance threshold Then, these two rest point clusters and all the motion point clusters in between will be merged into a new rest point cluster. and for the new point cluster The relevant point cluster information is updated, including: , , , , .

[0132] ⑧ Add an end time to each dwell point cluster or moving point cluster in the point cluster sequence. Specifically, if subsequent point clusters exist, the end time of the current point cluster is set to the start time of the next point cluster; if no subsequent point clusters exist, meaning the current point cluster is the last point cluster in the point cluster sequence, the end time of the current point cluster is set to the right boundary of the aforementioned preset time window; finally, the updated point cluster sequence is used as the trajectory point state recognition result for the corresponding target object. Each point cluster in the point cluster sequence contains the following fields: .

[0133] Figure 6 This is a flowchart of another trajectory point state recognition method provided in Embodiment 3 of the present invention. Figure 6 As shown, the scheme mainly includes: the original GPS input process, the data preprocessing process, and the trajectory point status recognition process. The specific implementation process can be referred to the above embodiments, and will not be repeated here.

[0134] The technical solution of this invention, by grouping large-scale raw trajectory data into objects and then performing parallel data compression operations on the initial independent trajectory data of each target object, can effectively reduce redundant information, making the obtained target trajectory data more concise. This provides a high-quality, low-noise input foundation for subsequent trajectory analysis and reduces the computational resource consumption of the subsequent analysis process. By using time-space dual constraints to perform parallel trajectory analysis on the target trajectory data of each target object, the accuracy and efficiency of identifying stop points in the trajectory are improved. At the same time, this solution is applicable to the trajectory analysis needs of different target objects (people, vehicles, equipment, etc.) and has strong universality.

[0135] Example 4

[0136] Figure 7 This is a schematic diagram of the structure of a trajectory point state recognition device provided in Embodiment 4 of the present invention. Figure 7 As shown, the device includes:

[0137] Compression processing module 41 is used to compress the acquired raw trajectory data to obtain target trajectory data of different target objects;

[0138] The trajectory analysis module 42 is used to perform trajectory analysis on the target trajectory data of each target object according to the preset spatiotemporal density constraints, and generate the trajectory point status recognition results of each target object; wherein, the preset spatiotemporal density constraints include spatial constraints based on preset dwell radius and temporal constraints based on preset dwell time.

[0139] Furthermore, based on the above embodiments of the invention, the compression processing module 41 includes:

[0140] The object identifier extraction unit is used to extract the target object identifier of each trajectory point in the original trajectory data;

[0141] The object grouping unit is used to divide the original trajectory data into initial independent trajectory data corresponding to different target objects according to the target object identifier;

[0142] The data compression parallel execution unit is used to perform data compression operations on each initial independent trajectory data in parallel to obtain the corresponding target trajectory data.

[0143] Furthermore, based on the above embodiments of the invention, a data compression operation is performed on the initial independent trajectory data to obtain the corresponding target trajectory data, including:

[0144] According to the preset time window and preset latitude and longitude coordinate range, the initial independent trajectory data is filtered for trajectory points to obtain the first trajectory data;

[0145] Arrange the trajectory points in the first trajectory data in chronological order, merge trajectory points with consecutive latitude and longitude coordinates and retain the earliest timestamp to obtain the second trajectory data;

[0146] Traverse the second trajectory data, determine the first distance between the current trajectory point and the previous trajectory point, and the second distance between the previous trajectory point and the next trajectory point of the current trajectory point. If the first distance and the second distance meet the preset drift point determination conditions, then determine the current trajectory point as a drift point and remove it.

[0147] The second trajectory data, after the drift points are removed, is used as the third trajectory data. Trajectory points with consecutive latitude and longitude coordinates in the third trajectory data are merged to obtain the target trajectory data.

[0148] Furthermore, based on the above embodiments of the invention, the trajectory analysis module 42 includes:

[0149] The parallel execution unit for state recognition is used to perform trajectory point state recognition operations in parallel on each target trajectory data and generate corresponding trajectory point state recognition results. The trajectory point state recognition operation is used to identify trajectory points in the target trajectory data as stationary points or moving points based on preset spatiotemporal density constraints.

[0150] Furthermore, based on the above embodiments of the invention, a trajectory point state recognition operation is performed on the target trajectory data to generate a corresponding trajectory point state recognition result, including:

[0151] Initialize the first trajectory point in the target trajectory data as the center point of the current point cluster;

[0152] Traverse subsequent trajectory points in chronological order and determine whether all trajectory points in the current point cluster are within the target circular area with the center point as the center and the preset dwell radius as the radius.

[0153] If all trajectory points are within the target circular area, and there are undetected trajectory points in the target trajectory data, then merge the current trajectory points into the current point cluster, update the center point of the current point cluster, and then return to the step of traversing subsequent trajectory points in chronological order to determine whether all trajectory points in the current point cluster are within the target circular area with the center point as the center and the preset dwell radius as the radius.

[0154] If all trajectory points are within the target circular area and there are no undetected trajectory points in the target trajectory data, then mark the current point cluster as a stationary point cluster and record the point cluster information;

[0155] If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is longer than the preset dwell time, then the current point cluster after removing the last trajectory point is marked as a dwell point cluster, and the point cluster information is recorded. Then, the last trajectory point is used as the center point of the new current point cluster to re-execute the point cluster aggregation process.

[0156] If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is less than or equal to the preset dwell time, then the first trajectory point in the current point cluster is marked as a moving point cluster, and the point cluster information is recorded. Then, the second trajectory point in the current point cluster is used as the center point of the new current point cluster to re-execute the point cluster aggregation process.

[0157] After all the target trajectory data has been traversed, the cumulative spherical distance between all moving point clusters between adjacent stopping point clusters in the obtained point cluster sequence is determined. If the cumulative spherical distance is less than a preset distance threshold, the two stopping point clusters and all moving point clusters between the two stopping point clusters are merged into a new stopping point cluster, and the point cluster information of the new stopping point cluster is updated.

[0158] The end time of each point cluster in the point cluster sequence is determined to update the point cluster information, and the updated point cluster sequence is used as the trajectory point state recognition result of the target object.

[0159] Furthermore, based on the above embodiments of the invention, determining whether the trajectory point is within the target circular area includes:

[0160] Determine the cosine of the geocentric angle between the trajectory point and the center point of the current point cluster;

[0161] If the cosine value of the geocentric angle is greater than the preset threshold value of the cosine value of the geocentric angle, the trajectory point is determined to be within the target circular area; otherwise, the trajectory point is determined to be outside the target circular area.

[0162] The trajectory point state recognition device provided in this embodiment of the invention can execute the trajectory point state recognition method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0163] Example 5

[0164] Figure 8 A schematic diagram of an electronic device 50 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0165] like Figure 8 As shown, the electronic device 50 includes at least one processor 51 and a memory, such as a read-only memory (ROM) 52 and a random access memory (RAM) 53, communicatively connected to the at least one processor 51. The memory stores computer programs executable by the at least one processor. The processor 51 can perform various appropriate actions and processes based on the computer program stored in the ROM 52 or loaded from storage unit 58 into the RAM 53. The RAM 53 can also store various programs and data required for the operation of the electronic device 50. The processor 51, ROM 52, and RAM 53 are interconnected via a bus 54. An input / output (I / O) interface 55 is also connected to the bus 54.

[0166] Multiple components in electronic device 50 are connected to I / O interface 55, including: input unit 56, such as keyboard, mouse, etc.; output unit 57, such as various types of monitors, speakers, etc.; storage unit 58, such as disk, optical disk, etc.; and communication unit 59, such as network card, modem, wireless transceiver, etc. Communication unit 59 allows electronic device 50 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0167] Processor 51 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 51 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 51 performs the various methods and processes described above, such as trajectory point state recognition methods.

[0168] In some embodiments, the trajectory point state recognition method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 58. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 50 via ROM 52 and / or communication unit 59. When the computer program is loaded into RAM 53 and executed by processor 51, one or more steps of the trajectory point state recognition method described above may be performed. Alternatively, in other embodiments, processor 51 may be configured to perform the trajectory point state recognition method by any other suitable means (e.g., by means of firmware).

[0169] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0170] In some embodiments, the trajectory point state recognition method may be implemented as a computer program, which is implicitly included in a computer program product. When executed by a processor, the computer program implements the trajectory point state recognition method of the present invention. The computer program product can be understood as a software product that primarily implements its solution through a computer program. The computer program used to implement the method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer program causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer program may be executed entirely on a machine, partially on a machine, partially on a remote machine as a standalone software package, or entirely on a remote machine or server.

[0171] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0172] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0173] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0174] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0175] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0176] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for recognizing the state of trajectory points, characterized in that, The method includes: The acquired raw trajectory data is compressed to obtain target trajectory data for different target objects; Based on preset spatiotemporal density constraints, trajectory analysis is performed on the target trajectory data of each target object to generate trajectory point state recognition results for each target object; wherein, the preset spatiotemporal density constraints include spatial constraints based on preset dwell radius and temporal constraints based on preset dwell duration; The process includes trajectory analysis of the target trajectory data to generate corresponding trajectory point state recognition results, including: The first trajectory point in the target trajectory data is initialized as the center point of the current point cluster; Traverse subsequent trajectory points in chronological order and determine whether all trajectory points in the current point cluster are within the target circular area centered at the center point and with the preset dwell radius as the radius. If all trajectory points are within the target circular area, and there are undetected trajectory points in the target trajectory data, then the current trajectory points are merged into the current point cluster, the center point of the current point cluster is updated, and then the process of traversing subsequent trajectory points in chronological order and determining whether all trajectory points in the current point cluster are within the target circular area with the center point as the center and the preset dwell radius as the radius is returned. If all trajectory points are within the target circular area and there are no undetected trajectory points in the target trajectory data, then the current point cluster is marked as a stationary point cluster, and the point cluster information is recorded; If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is greater than the preset dwell time, then the current point cluster after removing the last trajectory point is marked as a dwell point cluster, and the point cluster information is recorded. Then, the last trajectory point is used as the center point of the new current point cluster to re-execute the point cluster aggregation process. If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is less than or equal to the preset dwell time, then the first trajectory point in the current point cluster is marked as a moving point cluster, and the point cluster information is recorded. Then, the second trajectory point in the current point cluster is used as the new center point of the current point cluster to re-execute the point cluster aggregation process. After all the target trajectory data has been traversed, the cumulative spherical distance between all moving point clusters between adjacent stopping point clusters in the obtained point cluster sequence is determined. If the cumulative spherical distance is less than a preset distance threshold, the two stopping point clusters and all moving point clusters between the two stopping point clusters are merged into a new stopping point cluster, and the point cluster information of the new stopping point cluster is updated. The end time of each point cluster in the point cluster sequence is determined to update the point cluster information, and the updated point cluster sequence is used as the trajectory point state recognition result of the target object.

2. The method according to claim 1, characterized in that, The process of compressing the acquired raw trajectory data to obtain target trajectory data for different target objects includes: Extract the target object identifiers of each trajectory point from the original trajectory data; The original trajectory data is divided into initial independent trajectory data corresponding to different target objects according to the target object identifier; Data compression operations are performed in parallel on each of the initial independent trajectory data to obtain the corresponding target trajectory data.

3. The method according to claim 2, characterized in that, Perform data compression on the initial independent trajectory data to obtain the corresponding target trajectory data, including: According to a preset time window and a preset latitude and longitude coordinate range, the initial independent trajectory data is filtered for trajectory points to obtain the first trajectory data; Arrange the trajectory points in the first trajectory data in chronological order, merge trajectory points with consecutive latitude and longitude coordinates and retain the earliest timestamp to obtain the second trajectory data; Traverse the second trajectory data to determine the first distance between the current trajectory point and the previous trajectory point, and the second distance between the previous trajectory point and the next trajectory point of the current trajectory point. If the first distance and the second distance meet the preset drift point determination conditions, then the current trajectory point is determined as a drift point and removed. The second trajectory data after removing the drift points is used as the third trajectory data, and the trajectory points with consecutive latitude and longitude coordinates in the third trajectory data are merged to obtain the target trajectory data.

4. The method according to claim 1, characterized in that, The step of performing trajectory analysis on the target trajectory data of each target object according to preset spatiotemporal density constraints to generate trajectory point state recognition results for each target object includes: The trajectory point state recognition operation is performed in parallel on each of the target trajectory data to generate the corresponding trajectory point state recognition result; wherein, the trajectory point state recognition operation is used to identify the trajectory points in the target trajectory data as stationary points or moving points based on the preset spatiotemporal density constraint conditions.

5. The method according to claim 1, characterized in that, Determining whether the trajectory point is within the target circular area includes: Determine the cosine of the geocentric angle between the trajectory point and the center point of the current point cluster; If the cosine value of the geocentric angle is greater than a preset threshold value for the cosine value of the geocentric angle, then the trajectory point is determined to be within the target circular area; otherwise, the trajectory point is determined not to be within the target circular area.

6. A trajectory point state recognition device, characterized in that, The device includes: The compression processing module is used to compress the acquired raw trajectory data to obtain target trajectory data for different target objects; The trajectory analysis module is used to perform trajectory analysis on the target trajectory data of each target object according to preset spatiotemporal density constraints, and generate trajectory point state recognition results for each target object; wherein, the preset spatiotemporal density constraints include spatial constraints based on preset dwell radius and temporal constraints based on preset dwell duration. The trajectory analysis module is further configured to: The first trajectory point in the target trajectory data is initialized as the center point of the current point cluster; Traverse subsequent trajectory points in chronological order and determine whether all trajectory points in the current point cluster are within the target circular area centered at the center point and with the preset dwell radius as the radius. If all trajectory points are within the target circular area, and there are undetected trajectory points in the target trajectory data, then the current trajectory points are merged into the current point cluster, the center point of the current point cluster is updated, and then the process of traversing subsequent trajectory points in chronological order and determining whether all trajectory points in the current point cluster are within the target circular area with the center point as the center and the preset dwell radius as the radius is returned. If all trajectory points are within the target circular area and there are no undetected trajectory points in the target trajectory data, then the current point cluster is marked as a stationary point cluster, and the point cluster information is recorded; If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is greater than the preset dwell time, then the current point cluster after removing the last trajectory point is marked as a dwell point cluster, and the point cluster information is recorded. Then, the last trajectory point is used as the center point of the new current point cluster to re-execute the point cluster aggregation process. If at least one trajectory point is not within the target circular area, and the duration of the current point cluster is less than or equal to the preset dwell time, then the first trajectory point in the current point cluster is marked as a moving point cluster, and the point cluster information is recorded. Then, the second trajectory point in the current point cluster is used as the new center point of the current point cluster to re-execute the point cluster aggregation process. After all the target trajectory data has been traversed, the cumulative spherical distance between all moving point clusters between adjacent stopping point clusters in the obtained point cluster sequence is determined. If the cumulative spherical distance is less than a preset distance threshold, the two stopping point clusters and all moving point clusters between the two stopping point clusters are merged into a new stopping point cluster, and the point cluster information of the new stopping point cluster is updated. The end time of each point cluster in the point cluster sequence is determined to update the point cluster information, and the updated point cluster sequence is used as the trajectory point state recognition result of the target object.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the trajectory point state recognition method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the trajectory point state recognition method according to any one of claims 1-5.

9. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the trajectory point state recognition method according to any one of claims 1-5.

Images