Multi-sensor track fusion method, device, equipment, medium and computer product

Abstract

The invention discloses a multi-sensor track fusion method, device and equipment, a storage medium and a computer product, and relates to the technical field of data processing, and the method comprises the steps: carrying out the time clustering processing and the space clustering processing of a plurality of tracks collected by a plurality of sensors in sequence based on the time information and the space distribution information of the tracks, and obtaining a multi-sensor trajectory cluster, and fusing all trajectories in the multi-sensor trajectory cluster to obtain a fused trajectory of the corresponding target object. According to the method, first clustering processing is carried out on a plurality of tracks collected by each sensor according to time information to form a preliminary track cluster, so that independent analysis processing of each cluster is realized, the complexity of subsequent calculation is reduced, and then the spatial distribution condition of each track is determined according to spatial distribution information, so that the accuracy of track distribution is improved. Therefore, the calculation data volume of sensor track fusion is further reduced, and the technical effect of improving the multi-sensor track fusion efficiency is achieved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a multi-sensor trajectory fusion method, device, equipment, storage medium and computer product. Background Art

[0002] With the development of Internet of Things technology, there are more and more types and specifications of sensors. In a specific application scenario, the trajectory of the same object can be generated by multiple sensors, and for a target object, only one trajectory is needed to represent the activity information of the target object within a period of time. Multiple trajectories generated by multiple sensors need to be fused.

[0003] In the related art, when performing trajectory fusion, usually only spatial dimension information is considered, that is, the trajectory of the target object is determined for fusion based on the trajectory similarity between different trajectories. Since there may be multiple targets in a time period, when only spatial dimension information is used for trajectory fusion, there is a problem of low fusion efficiency due to large amount of calculation. Summary of the invention

[0004] The main purpose of this application is to provide a multi-sensor trajectory fusion method, device, equipment, storage medium and computer product, aiming to solve the technical problem that the related technology has large computational complexity and low fusion efficiency in multi-sensor trajectory fusion.

[0005] To achieve the above objectives, the present application proposes a multi-sensor trajectory fusion method, the method comprising:

[0006] Based on the time information and spatial distribution information of the trajectories, a plurality of trajectories collected by a plurality of sensors are subjected to time clustering processing and space clustering processing in sequence to obtain at least one clustering cluster that satisfies the time clustering condition and the space clustering condition in sequence;

[0007] Any obtained cluster is used as a multi-sensor trajectory cluster for the same target object, and all trajectories in the multi-sensor trajectory cluster are fused to obtain a fused trajectory of the corresponding target object.

[0008] In one embodiment, the time information is determined based on the start time information and the end time information of each track, and the steps of successively performing time clustering processing and space clustering processing on multiple tracks collected by multiple sensors include:

[0009] A first clustering process is performed on any two of the acquired multiple trajectories. If the end time of the first trajectory is greater than or equal to the start time of the second trajectory and the end time of the second trajectory is greater than or equal to the start time of the first trajectory, the first trajectory and the second trajectory are clustered into one cluster.

[0010] In one embodiment, the spatial distribution information includes longitude data and latitude data of each track, and the method further includes the steps of:

[0011] Based on the longitude data of the maximum trajectory point, the longitude data of the minimum trajectory point, the latitude data of the maximum trajectory point, the latitude data of the minimum trajectory point and the spatial resolution in each trajectory, each trajectory point is spatially encoded to obtain a spatially encoded trajectory in the same gridded space; wherein the gridded space is determined based on the spatial resolution.

[0012] In one embodiment, based on the time information and spatial distribution information of the trajectories, the steps of sequentially performing time clustering processing and spatial clustering processing on the multiple trajectories collected by the multiple sensors include:

[0013] Performing a second clustering process on any two spatial coding trajectories in the same cluster cluster after the first clustering process, and determining the grid unit intersection and grid unit union between the two spatial coding trajectories based on the distribution information of the trajectory points of the spatial coding trajectories in the gridded space;

[0014] Determine the trajectory similarity between two spatially encoded trajectories based on the number of grid cells in the grid cell intersection and the number of grid cells in the grid cell union; wherein the trajectory similarity is positively correlated with the number of grid cells in the grid cell intersection;

[0015] If the trajectory similarity is not lower than the preset threshold, the spatial encoding trajectories are clustered into the same cluster.

[0016] In one embodiment, the step of fusing all trajectories in a multi-sensor trajectory cluster includes:

[0017] For each multi-sensor trajectory cluster, a trajectory with the largest number of trajectory points is determined from the multi-sensor trajectory cluster as a reference trajectory;

[0018] For each track other than the reference track, determining the track points that overlap with the reference track and the track points that do not overlap with the reference track;

[0019] For each group of overlapping trajectory points between other trajectories and the reference trajectory, data fusion processing is performed on each overlapping trajectory point to obtain fused trajectory points in the overlapping area;

[0020] Based on the fusion trajectory points of the overlapping area and the non-overlapping trajectory points, a fusion trajectory is determined.

[0021] In one embodiment, the step of performing data fusion processing on each overlapping trajectory point to obtain the fused trajectory points of the overlapping area includes:

[0022] For each group of overlapping trajectory points, the longitude data and latitude data of each overlapping trajectory point are averaged to obtain the longitude data and latitude data of the fused trajectory points in the overlapping area.

[0023] In a second aspect, to achieve the above-mentioned purpose, the present application further provides a multi-sensor trajectory fusion device, the device comprising:

[0024] A clustering module, based on the time information and spatial distribution information of the trajectories, performs time clustering processing and spatial clustering processing on multiple trajectories collected by multiple sensors in turn, and obtains at least one cluster cluster that satisfies the time clustering condition and the spatial clustering condition in turn;

[0025] The trajectory fusion module takes any obtained cluster as a multi-sensor trajectory cluster for the same target object, fuses all trajectories in the multi-sensor trajectory cluster, and obtains a fused trajectory of the corresponding target object.

[0026] In a third aspect, in order to achieve the above-mentioned purpose, the present application continues to provide a multi-sensor trajectory fusion device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the above-mentioned multi-sensor trajectory fusion method.

[0027] In a fourth aspect, in order to achieve the above-mentioned purpose, the present application continues to provide a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the above-mentioned multi-sensor trajectory fusion method are implemented.

[0028] In a fifth aspect, to achieve the above-mentioned purpose, the present application continues to provide a computer program product, the computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned multi-sensor trajectory fusion method are implemented.

[0029] One or more technical solutions proposed in this application have at least the following technical effects:

[0030] The present application performs a first clustering process on multiple trajectories collected by each sensor according to the start time information and end time information of the trajectory data to form a preliminary trajectory grouping, so as to realize independent analysis and processing of each cluster cluster and reduce the complexity of subsequent calculations. Compared with directly comparing the spatial dimension information between each trajectory in all trajectories, it has higher processing efficiency. Subsequently, for the trajectories in each cluster cluster formed after the first clustering process, according to the trajectory similarity between each trajectory in the cluster cluster, the secondary clustering of the trajectory is realized to finely distinguish the trajectories of different target objects. At the same time, the trajectory similarity calculation using the spatial distribution information has a lower data calculation amount than the traditional calculation of the distance information between different trajectory points, which provides a basis for the subsequent trajectory fusion processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0033] Figure 1 Schematic diagram of the process of a multi-sensor trajectory fusion method in one embodiment of the present application.

[0034] Figure 2 This is a flowchart of a multi-sensor trajectory fusion method in a specific implementation example of the present application.

[0035] Figure 3 This is a schematic diagram of the structure of the multi-sensor trajectory fusion device of this application.

[0036] Figure 4 This is a schematic diagram of the structure of the multi-sensor trajectory fusion device of this application.

[0037] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0038] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0039] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0040] With the development of Internet of Things technology, there are more and more types and specifications of sensors. The trajectory of the same target object can be generated by multiple sensors, but only one trajectory is needed to represent the activity information of a target object within a period of time. Multiple trajectories generated by multiple sensors for a target object need to be fused.

[0041] In the related art, when performing trajectory fusion, usually only spatial dimension information is considered, that is, the trajectory of the target object is determined for fusion based on the trajectory similarity between different trajectories. Since there may be multiple targets in a time period, when only spatial dimension information is used for trajectory fusion, there is a problem of low fusion efficiency due to large amount of calculation.

[0042] In addition, in the related art, when calculating spatial similarity, a distance calculation method between trajectory points such as Euclidean distance calculation is usually used to achieve similarity calculation between trajectories. Since each trajectory contains a large amount of trajectory sampling point data, there are problems of large amount of calculation and low calculation performance when performing similarity calculation.

[0043] In view of the problem of low efficiency of multi-sensor trajectory fusion processing in related technologies, the present application provides a solution. According to the start time information and end time information of the trajectory data, a first clustering processing is performed on the multiple trajectories collected by each sensor to form a preliminary trajectory grouping, so as to realize independent analysis and processing of each cluster cluster and reduce the complexity of subsequent calculations. Compared with directly comparing the spatial dimension information between each trajectory in all trajectories, it has higher processing efficiency. Subsequently, for the trajectories in each cluster cluster formed after the first clustering processing, the secondary clustering of the trajectories is realized according to the trajectory similarity between each trajectory in the cluster cluster, so as to finely distinguish the trajectories of different target objects, and provide a basis for subsequent trajectory fusion processing.

[0044] Furthermore, in terms of trajectory similarity calculation, the present application sets the spatial resolution and spatially encodes each trajectory point in each trajectory to determine the spatial distribution of different trajectories in the grid space of the current spatial resolution, and then determines the grid cell intersection and the grid cell union, and confirms the trajectory similarity based on the number of grid cells in the grid cell intersection and the number of grid cells in the grid cell union. Compared with the method of calculating the similarity based on the distance information of trajectory points between trajectories in the related art, the present application, through a reasonably designed spatial resolution, only needs to confirm the distribution of the trajectory in the grid space (the number of grid cells covered by the trajectory) to achieve the distinction of trajectory data of different target objects, without calculating the distance information between the trajectory points of different trajectories, thereby reducing the amount of calculation and further improving the processing efficiency.

[0045] Based on this, the present application embodiment provides a multi-sensor trajectory fusion method, referring to Figure 1 , Figure 1 This is a flowchart of the first embodiment of the multi-sensor trajectory fusion method of the present application.

[0046] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a multi-sensor trajectory fusion device, etc. The following takes a multi-sensor trajectory fusion device as an example to illustrate this embodiment and the following embodiments.

[0047] In this embodiment, the multi-sensor trajectory fusion method includes steps S10 to S20:

[0048] Step S10, based on the time information and spatial distribution information of the trajectories, a plurality of trajectories collected by a plurality of sensors are subjected to time clustering processing and space clustering processing in sequence to obtain at least one clustering cluster that successively satisfies the time clustering condition and the space clustering condition.

[0049] Step S20 , taking any one of the obtained clusters as a multi-sensor trajectory cluster for the same target object, fusing all trajectories in the multi-sensor trajectory cluster to obtain a fused trajectory corresponding to the target object.

[0050] It should be noted that when multiple sensors are deployed in one application scenario, the sensors may be sensors of different specifications and / or different sampling frequencies. The trajectory collected by the sensor may be the trajectory data of one target object or multiple target objects, and the trajectory data includes at least three attributes: time information, longitude information, and latitude information.

[0051] It should also be noted that the temporal clustering process can be performed based on the temporal information of the trajectory as a temporal clustering condition, and the spatial clustering process can be performed based on the spatial distribution information of the trajectory as a spatial clustering condition.

[0052] Specifically, firstly, based on the time information, the trajectory data collected by each sensor is subjected to the first clustering process (time clustering process) to obtain at least one cluster that meets the time clustering condition. Then, for each cluster obtained after the first clustering process, the second clustering process (spatial clustering process) can be performed according to the spatial distribution information of each trajectory in the cluster. The cluster obtained after the second clustering process can be understood as the trajectory data collected by multiple sensors for the same target object within a period of time. Then, the trajectory fusion process is performed on the multi-sensor trajectory in each cluster after the second clustering process to obtain the fused trajectory of the corresponding target object, and each cluster corresponds to a fused trajectory.

[0053] As a feasible implementation, step S10 may include steps A10 to A30, and step S20 may include step A40:

[0054] Step A10, obtaining multiple trajectories collected by multiple sensors.

[0055] It should be noted that when multiple sensors are deployed in one application scenario, the sensors may be sensors of different specifications and / or different sampling frequencies. The trajectory collected by the sensor may be the trajectory data of one target object or multiple target objects, and the trajectory data includes at least three attributes: time information, longitude information, and latitude information.

[0056] Exemplarily, trajectory data collected within a certain period of time may be obtained from multiple sensors of the target scene, and necessary preprocessing may be performed on the received trajectory data to ensure the quality and consistency of the trajectory data.

[0057] Step A20: performing a first clustering process on each trajectory based on the time information of the trajectory.

[0058] In a feasible implementation manner, the time information is determined based on the start time information and the end time information of each track, and step A20 may include step A21:

[0059] A first clustering process is performed on any two of the acquired multiple trajectories. If the end time of the first trajectory is greater than or equal to the start time of the second trajectory and the end time of the second trajectory is greater than or equal to the start time of the first trajectory, the first trajectory and the second trajectory are clustered into one cluster.

[0060] It should be noted that the start time information can be the start time data of a certain trajectory collected within a time period, for example, the timestamp of the first trajectory point in a certain trajectory collected within a time period, and the end time information can be the end time data of a certain trajectory collected within a time period, for example, the timestamp of the last trajectory point in a certain trajectory collected within a time period.

[0061] It should also be noted that the first trajectory is one of any two trajectories among a plurality of trajectories obtained from a plurality of sensors, and the second trajectory is the other one of the two trajectories.

[0062] Specifically, a first clustering process is performed on all trajectories acquired from a plurality of sensors, so that each trajectory forms at least one cluster according to the start time information and the end time information.

[0063] This embodiment can be understood as follows: if there is an overlap in time between any two trajectories within a certain time period, that is, the end time of one of the two trajectories is not earlier than the start time of the other trajectory and the end time of the other trajectory is not earlier than the start time of the first trajectory, then the two trajectories are clustered together. On the contrary, if there is no overlap in time between any two trajectories, that is, the end time of one of the two trajectories is earlier than the start time of the other trajectory or the end time of the other trajectory is earlier than the start time of the first trajectory, then the two trajectories will not be clustered together.

[0064] It can be understood that preliminary clustering based on the start time information and end time information of each trajectory can quickly screen out trajectories that overlap or are adjacent in time. This may be the performance of the same target object under different sensors. Using the start time information and end time information to design clustering standards and preliminary clustering of each trajectory can achieve independent analysis and processing of each cluster cluster, reduce the complexity of subsequent calculations, and improve calculation efficiency.

[0065] Step A30 , for the trajectories in the same cluster after the first clustering process, based on the trajectory similarities between the trajectories, a second clustering process is performed on the trajectories in the same cluster.

[0066] It should be noted that the trajectory similarity refers to the similarity between any two trajectories in each cluster with two or more trajectories after the first clustering process.

[0067] For example, the Euclidean distances of all corresponding trajectory points between any two trajectories may be calculated, and the distance difference between the two trajectories may be considered as a whole to determine the distance similarity between the two trajectories. According to the trajectory similarity, the trajectories under the same clustering cluster may be further clustered.

[0068] It can be understood that after the first clustering processing in the time dimension, the second clustering processing can consider the difference between any two trajectories in a cluster in the spatial dimension, determine the trajectory similarity, and perform the second clustering processing based on the trajectory similarity to achieve further clustering of the trajectories. The trajectories in the same cluster after the second clustering processing can be confirmed as the collection trajectories of different sensors for the same target object within a time period.

[0069] In a feasible implementation manner, the trajectory similarity is determined according to the spatial distribution information of each trajectory, and the spatial distribution information at least includes the longitude data and the dimension data of each trajectory. Step A30 may also include steps A31 to A34:

[0070] Step A31, based on the longitude data of the maximum trajectory point, the longitude data of the minimum trajectory point, the latitude data of the maximum trajectory point, the latitude data of the minimum trajectory point and the spatial resolution in each of the trajectories, spatially encode each of the trajectory points to obtain a spatially encoded trajectory in the same gridded space; wherein the gridded space is determined based on the spatial resolution.

[0071] Step A32, performing a second clustering process on any two spatial coding trajectories in the same cluster after the first clustering process, and determining the grid unit intersection and grid unit union between the two spatial coding trajectories based on the distribution information of the trajectory points of the spatial coding trajectories in the gridded space.

[0072] Step A33: determining the trajectory similarity between two of the spatially encoded trajectories based on the number of grid cells in the intersection of the grid cells and the number of grid cells in the union of the grid cells; wherein the trajectory similarity is positively correlated with the number of grid cells in the intersection of the grid cells.

[0073] Step A34: if the trajectory similarity is not lower than a preset threshold, clustering the spatially encoded trajectories into the same cluster.

[0074] In this embodiment, based on the latitude and longitude information and spatial resolution of the trajectory points, the trajectory data acquired from each sensor is standardized into a discrete spatially encoded trajectory to facilitate the subsequent second clustering process.

[0075] Specifically, the longitude and latitude data of each track point are confirmed, and the longitude data of the maximum track point, the longitude data of the minimum track point, the latitude data of the maximum track point and the latitude data of the minimum track point in the track are confirmed, and the spatial resolution parameter is confirmed. Then, according to the spatial resolution parameter, the longitude data of the maximum track point, the longitude data of the minimum track point, the latitude data of the maximum track point and the latitude data of the minimum track point, the track points of each track are mapped to the grid space to achieve independent spatial coding of each track point and form a standardized spatial coding track.

[0076] In this embodiment, the spatial resolution is designed to improve the robustness of the second clustering process, that is, the spatial resolution represents the tolerance to the errors of each sensor. For ease of understanding, a reference gridded space can be set. The gridded space can be understood as a space composed of grid cells. The size of the grid cell is used to describe the spatial resolution. Each grid in the gridded space is a grid cell. The smaller the grid cell size, the lower the tolerance to sensor errors. Conversely, the larger the grid cell size, the higher the tolerance to sensor errors. The trajectory points of each trajectory can be spatially encoded and mapped to the gridded space.

[0077] Exemplarily, the spatial encoding method of each trajectory point is shown in Formula 1:

[0078] Formula 1

[0079] Among them, encode ijk represents the spatial encoding of the jth track point of the ith track at the kth time point, lon ijk Indicates the longitude data of the jth track point of the ith track at the kth time point, lat ijkrepresents the latitude data of the jth track point of the ith track at the kth time point, Lon_Min represents the minimum track point longitude data, Lon_Max represents the maximum track point longitude data, Lat_Min represents the minimum track point latitude data, Lat_Max represents the maximum track point latitude data, S resolusion Indicates the spatial resolution.

[0080] It can be understood that, through Formula 1, the trajectory points of each trajectory can be mapped to the gridded space under the set spatial resolution to achieve spatial encoding of each trajectory point, so that the subsequent trajectory similarity calculation can be performed. Each trajectory point after spatial encoding can cover a grid unit. Furthermore, after the trajectory points of each trajectory are spatially encoded to obtain the spatially encoded trajectory, for any two trajectories in each cluster after the first clustering process, the distribution of their spatially encoded trajectory points in the gridded space is checked.

[0081] Specifically, we can traverse the grid cells that are jointly covered by the spatially encoded trajectory points of the two trajectories (i.e., the intersection of the grid cells), as well as the grid cells that they each cover (i.e., the union of the grid cells). The trajectory similarity is measured by comparing the number of grid cells in the intersection of the grid cells with the number of grid cells in the union of the grid cells. Generally speaking, the larger the size of the grid cells, the more likely the trajectory points of different trajectories are to cover the same grid cell. The more grid cells there are in the intersection of the grid cells, the closer the spatial positions of the two trajectories are, and the higher the trajectory similarity.

[0082] For example, the calculation method of trajectory similarity is shown in Formula 2:

[0083] Formula 2: Similarity = k*I n,m / U n,m

[0084] Among them, Similarity represents the trajectory similarity between the nth trajectory and the mth trajectory in the cluster, I n,m represents the grid cell intersection of the nth trajectory and the mth trajectory in the cluster, U n,m represents the union of the grid cells of the nth trajectory and the mth trajectory in the cluster, and k represents the similarity coefficient.

[0085] After determining the trajectory similarity between two trajectories, if the similarity reaches a preset threshold, it is considered that the two trajectories are likely to represent trajectory data of the same target object under different sensors, and they are classified into the same cluster.

[0086] It can be understood that the present embodiment only needs to confirm the distribution of the trajectory in the grid space (the number of grid units covered by the trajectory) to distinguish the trajectory data of different target objects, without calculating the distance information between the trajectory points of different trajectories, thereby reducing the amount of calculation and further improving the processing efficiency. At the same time, through the design of reasonable spatial resolution, the influence of sensor error on the calculation accuracy can be reduced. The reasonable design of the similarity coefficient and the preset threshold can take into account the influence of sensors with different sampling frequencies on the number of trajectory points, further improving the robustness of the second clustering processing and ensuring a good clustering effect.

[0087] Step A40 , performing trajectory fusion processing on the trajectories under the same multi-sensor trajectory cluster after the second clustering processing to obtain a fused trajectory.

[0088] In this embodiment, the trajectories in each multi-sensor trajectory cluster after the second clustering process are considered to be trajectory data of the same target object collected by different sensors, and these trajectory data need to be fused to form one trajectory data of the target object.

[0089] In a feasible implementation, step A40 may include steps A41 to A44:

[0090] Step A41 : for each of the multi-sensor trajectory clusters, determine a trajectory with the largest number of trajectory points from the multi-sensor trajectory cluster as a reference trajectory.

[0091] Step A42: for each track other than the reference track in the same cluster, determine the overlapping track points and non-overlapping track points of the track other than the reference track.

[0092] Step A43: for each group of overlapping trajectory points between other trajectories and the reference trajectory, data fusion processing is performed on each overlapping trajectory point to obtain fused trajectory points in the overlapping area.

[0093] Step A44, determining a fused trajectory based on the fused trajectory points in the overlapping area and the non-overlapping trajectory points.

[0094] It should be noted that the trajectory with the largest number of trajectory points can also be considered as the trajectory data collected by the sensor with the highest sampling frequency. When there are sensors with the same sampling frequency, a trajectory can be randomly selected as the reference trajectory.

[0095] It should be noted that the overlapping trajectory points are the trajectory points of the same cluster that overlap in time, and the non-overlapping trajectory points are the trajectory points of the same cluster that do not overlap in time. For example, if the timestamp of a trajectory point of one of any two trajectories in a cluster is the same as the timestamp of a trajectory point of the other trajectory, then these two trajectory points are the overlapping trajectory points of the two trajectories, and the two trajectory points are confirmed as a set of overlapping trajectory points.

[0096] Specifically, the trajectory with the largest number of trajectory points (or the highest sampling frequency) is selected from the same multi-sensor trajectory cluster as the baseline trajectory, and then the overlapping and non-overlapping trajectory points between other trajectories and the baseline trajectory are analyzed. The trajectory points that overlap in time are fused to generate fused trajectory points in the overlapping area. The trajectory points that do not overlap in time are directly included in the fused trajectory to form a complete fused trajectory.

[0097] It is understandable that by selecting the trajectory with the most trajectory points as the reference trajectory, the accuracy of the data with high sampling frequency can be taken into account while ensuring the integrity of the trajectory fusion result. By directly incorporating non-overlapping trajectory points into the fusion trajectory, the resulting fusion trajectory can not only more comprehensively reflect the motion characteristics of the target object, but also improve the efficiency and robustness of multi-sensor data fusion, providing a higher quality data foundation for subsequent target analysis.

[0098] In a specific implementation, step A43 may include step B10:

[0099] Step B10: for each group of overlapping trajectory points, the longitude data and latitude data of each overlapping trajectory point are averaged to obtain the longitude data and latitude data of the fused trajectory points in the overlapping area.

[0100] It can be understood that by taking the average, the accuracy of different sensor data can be comprehensively considered to obtain relatively accurate fusion trajectory point data.

[0101] For example, to help understand the implementation process of the multi-sensor trajectory fusion method obtained in this embodiment, please refer to Figure 2 , Figure 2 A brief structural diagram of the multi-sensor trajectory fusion method in this implementation example is provided, specifically:

[0102] In this example, the multi-sensor trajectory data fusion method may include the following specific steps

[0103] The method of multi-sensor trajectory data fusion can be as follows: access multiple sensors (the number of sensors is ≥ 2) trajectory data (the number of trajectory data is ≥ 2), and the attributes of the accessed trajectory data include at least three attributes: time, longitude and latitude. Get the timestamp T of the start time of each trajectory s The timestamp T of the end time e , for the i-th and j-th trajectories, if T ei ≥T sj And T ej ≥T si , then the two trajectories are divided into the same time cluster, otherwise the two trajectories are divided into different time clusters until all trajectories are aggregated. Set the spatial resolution S resolution The maximum longitude of the access trajectory range is recorded as LON_MAX, the minimum longitude is LON_MIN, the maximum latitude is recorded as LAT_MAX, and the minimum latitude is recorded as LAT_MIN. Each longitude and latitude point of all trajectories is spatially encoded.

[0104] Set the spatial similarity threshold to S threshold For each time cluster, the nth and mth trajectories are spatially encoded as n and encode m , whose spatial encoding intersection is I n,m =encode n ∩encode m , the union of the spatial encodings is U n,m =encode n ∪encode m , calculate the similarity of two trajectories according to the similarity calculation formula, if Similarity ≥ S threshold , then the two trajectories belong to the same spatial cluster, otherwise they belong to two spatial clusters. The trajectories of each time cluster are aggregated in turn to obtain the spatial cluster trajectory. For all the trajectories of each spatial cluster, the trajectory with the largest number of trajectory points is selected as the reference trajectory. The trajectory points of the remaining trajectories that overlap with the reference trajectory are averaged and merged into one trajectory point, while the trajectory points of the non-overlapping parts are retained. Finally, the trajectory clusters of the same space are merged into one trajectory. The trajectories of all spatial clusters are calculated in turn to obtain the fused trajectory of the multi-sensor trajectory data.

[0105] This application also provides a multi-sensor trajectory fusion device, please refer to Figure 3 , the multi-sensor trajectory fusion device includes:

[0106] A clustering module is used to perform time clustering processing and space clustering processing on multiple trajectories collected by multiple sensors based on the time information and spatial distribution information of the trajectories, and obtain at least one cluster cluster that satisfies the time clustering condition and the space clustering condition in turn;

[0107] The trajectory fusion module is used to use any obtained cluster as a multi-sensor trajectory cluster for the same target object, fuse all trajectories in the multi-sensor trajectory cluster, and obtain a fused trajectory of the corresponding target object.

[0108] The multi-sensor trajectory fusion device provided by the present application adopts the multi-sensor trajectory fusion method in the above embodiment, which can solve the technical problem of low fusion efficiency. Compared with the related art, the beneficial effects of the multi-sensor trajectory fusion device provided by the present application are the same as the beneficial effects of the multi-sensor trajectory fusion method provided by the above embodiment, and other technical features in the multi-sensor trajectory fusion device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0109] The present application provides a multi-sensor trajectory fusion device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the multi-sensor trajectory fusion method in the above-mentioned embodiment.

[0110] Reference below Figure 4 , which shows a schematic diagram of the structure of a multi-sensor trajectory fusion device suitable for implementing the embodiment of the present application. The multi-sensor trajectory fusion device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The multi-sensor trajectory fusion device shown is only an example and should not bring any limitations to the functions and scope of use of the embodiments of the present application.

[0111] The multi-sensor trajectory fusion device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the multi-sensor trajectory fusion device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the multi-sensor trajectory fusion device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a multi-sensor trajectory fusion device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.

[0112] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0113] The multi-sensor trajectory fusion device provided by the present application adopts the multi-sensor trajectory fusion method in the above embodiment, which can solve the technical problem of low fusion efficiency. Compared with the related art, the beneficial effects of the multi-sensor trajectory fusion device provided by the present application are the same as the beneficial effects of the multi-sensor trajectory fusion method provided by the above embodiment, and the other technical features in the multi-sensor trajectory fusion device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0114] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0115] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0116] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the multi-sensor trajectory fusion method in the above-mentioned embodiment.

[0117] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0118] The computer-readable storage medium may be included in the multi-sensor trajectory fusion device; or may exist independently without being assembled into the multi-sensor trajectory fusion device.

[0119] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the multi-sensor trajectory fusion device, the multi-sensor trajectory fusion device implements the multi-sensor trajectory fusion method.

[0120] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0121] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0122] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0123] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned multi-sensor trajectory fusion method, and can solve the technical problem of low fusion efficiency. Compared with the related art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the multi-sensor trajectory fusion method provided in the above-mentioned embodiment, and will not be repeated here.

[0124] The present application also provides a computer program product, including a computer program, which implements the steps of the multi-sensor trajectory fusion method as described above when the computer program is executed by a processor.

[0125] The computer program product provided by the present application can solve the technical problem of low fusion efficiency. Compared with the related art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the multi-sensor trajectory fusion method provided by the above embodiment, which will not be repeated here.

[0126] The above are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A multi-sensor trajectory fusion method, characterized in that: The method includes: Based on the time information and spatial distribution information of the trajectories, a plurality of trajectories collected by a plurality of sensors are subjected to time clustering processing and space clustering processing in sequence to obtain at least one clustering cluster that satisfies the time clustering condition and the space clustering condition in sequence; Any obtained cluster is used as a multi-sensor trajectory cluster for the same target object, and all trajectories in the multi-sensor trajectory cluster are fused to obtain a fused trajectory of the corresponding target object.

2. The method according to claim 1, characterized in that The time information is determined based on the start time information and the end time information of each of the trajectories, and the step of successively performing time clustering processing and space clustering processing on the multiple trajectories collected by the multiple sensors includes: A first clustering process is performed on any two of the acquired multiple trajectories. If the end time of the first trajectory is greater than or equal to the start time of the second trajectory and the end time of the second trajectory is greater than or equal to the start time of the first trajectory, the first trajectory and the second trajectory are clustered into one cluster.

3. The method according to claim 2, characterized in that The spatial distribution information includes longitude data and latitude data of each of the trajectories, and the method further includes the steps of: Based on the longitude data of the maximum trajectory point, the longitude data of the minimum trajectory point, the latitude data of the maximum trajectory point, the latitude data of the minimum trajectory point and the spatial resolution in each of the trajectories, spatial encoding processing is performed on each of the trajectory points to obtain a spatially encoded trajectory in the same gridded space; wherein the gridded space is determined based on the spatial resolution.

4. The method according to claim 3, characterized in that The step of sequentially performing time clustering processing and space clustering processing on multiple trajectories collected by multiple sensors based on the time information and spatial distribution information of the trajectories includes: Performing a second clustering process on any two spatially coded trajectories in the same cluster after the first clustering process, and determining a grid unit intersection and a grid unit union between the two spatially coded trajectories based on distribution information of the trajectory points of the spatially coded trajectories in the gridded space; Determining the trajectory similarity between two of the spatially encoded trajectories based on the number of grid cells in the intersection of the grid cells and the number of grid cells in the union of the grid cells; wherein the trajectory similarity is positively correlated with the number of grid cells in the intersection of the grid cells; If the trajectory similarity is not lower than a preset threshold, the spatially encoded trajectories are clustered into the same cluster.

5. The method according to claim 1, characterized in that The step of fusing all trajectories in the multi-sensor trajectory cluster comprises: For each of the multi-sensor trajectory clusters, determining a trajectory with the largest number of trajectory points from the multi-sensor trajectory clusters as a reference trajectory; For each track other than the reference track, determining the overlapping track points and non-overlapping track points of the other track and the reference track; For each group of overlapping trajectory points between the other trajectory and the reference trajectory, performing data fusion processing on each of the overlapping trajectory points to obtain overlapping area fusion trajectory points; The fusion trajectory is determined based on the overlapped area fusion trajectory points and the non-overlapped trajectory points.

6. The method according to claim 5, characterized in that The step of performing data fusion processing on each of the overlapping trajectory points to obtain the fused trajectory points of the overlapping area comprises: For each group of overlapping trajectory points, the longitude data and latitude data of each of the overlapping trajectory points are averaged to obtain the longitude data and latitude data of the fused trajectory points in the overlapping area.

7. A multi-sensor trajectory fusion device, characterized in that: The device comprises: A clustering module is used to perform time clustering processing and space clustering processing on multiple trajectories collected by multiple sensors based on the time information and spatial distribution information of the trajectories, and obtain at least one cluster cluster that satisfies the time clustering condition and the space clustering condition in turn; The trajectory fusion module is used to use any obtained cluster as a multi-sensor trajectory cluster for the same target object, fuse all trajectories in the multi-sensor trajectory cluster, and obtain a fused trajectory of the corresponding target object.

8. A multi-sensor trajectory fusion device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the multi-sensor trajectory fusion method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the multi-sensor trajectory fusion method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the multi-sensor trajectory fusion method according to any one of claims 1 to 6 are implemented.

Images