A trajectory completion method, device, terminal and computer-readable storage medium

Abstract

The present invention provides a trajectory completion method, device, terminal and computer-readable storage medium. The trajectory completion method includes obtaining an initial trajectory corresponding to a target vehicle in a preset area and a preset time period, wherein the initial trajectory includes a plurality of adjacent and sequentially arranged trajectory points; determining an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle; performing a similarity comparison between all collected vehicles with intermediate trajectory points within a preset time period and the target vehicle to determine the vehicle similarity corresponding to each collected vehicle; in response to the vehicle similarity of the collected vehicle exceeding a threshold, the intermediate trajectory point is supplemented to the initial trajectory of the target vehicle. By determining the intermediate trajectory point between adjacent trajectory points in the initial trajectory of the target vehicle, and then comparing the vehicle collected by the intermediate trajectory point with the target vehicle, the intermediate trajectory point passed by the target vehicle is determined, and the intermediate trajectory point is supplemented into the initial trajectory of the target vehicle, thereby completing the trajectory of the target vehicle.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a trajectory completion method, device, terminal and computer-readable storage medium. Background Art

[0002] With the rapid development and popularization of global positioning technology, mobile Internet technology, and the development of transportation, a large number of video images have been generated. Based on the target objects contained in the video images, the video images are clustered to determine the running trajectory of the target objects. However, due to the problems of shooting angle and image clarity, the running trajectory of the generated target objects is incomplete. Summary of the invention

[0003] The main technical problem solved by the present invention is to provide a trajectory completion method, device, terminal and computer-readable storage medium to solve the problem of incomplete vehicle trajectory in the prior art.

[0004] In order to solve the above technical problems, the first technical solution adopted by the present invention is to provide a trajectory completion method, which includes:

[0005] Acquire an initial trajectory of the target vehicle in a preset area and a preset time period, the initial trajectory including a plurality of adjacent and sequentially arranged trajectory points;

[0006] Determine an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle;

[0007] Compare the similarity of all collected vehicles with intermediate trajectory points within a preset time period with the target vehicle to determine the vehicle similarity corresponding to each collected vehicle;

[0008] In response to the vehicle similarity of the collected vehicle exceeding a threshold, the intermediate trajectory points are supplemented to the initial trajectory of the target vehicle.

[0009] Among them, all collected vehicles with intermediate trajectory points within a preset time period are compared with the target vehicle for similarity, and the vehicle similarity corresponding to each collected vehicle is determined, including:

[0010] Based on the position information and collection time corresponding to the first trajectory point and the second trajectory point respectively and the position information of the intermediate trajectory point, determine the time range in which the target vehicle passes through the intermediate trajectory point;

[0011] All collected vehicles within the time range of the intermediate trajectory points are compared with the target vehicle for similarity, and the vehicle similarity corresponding to each collected vehicle is determined.

[0012] Wherein, obtaining the initial trajectory of the target vehicle in the preset area and the preset time period includes:

[0013] Acquire a captured image set corresponding to a preset area and a preset time period; the captured image set includes a plurality of captured images including a captured vehicle;

[0014] Performing target clustering on the collected images in the captured image set to obtain image subsets corresponding to each collected vehicle;

[0015] Generate an initial trajectory of the collected vehicle using the collection time of the collected images contained in the image subset of the collected vehicle;

[0016] The initial trajectory of each collected vehicle is used as the initial trajectory of the target vehicle in turn.

[0017] Wherein, determining the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle comprises:

[0018] Based on the initial trajectory of the target vehicle, a candidate path between a first trajectory point and a second trajectory point is determined; the candidate path includes candidate trajectory points; the first trajectory point is a previous trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is a subsequent trajectory point;

[0019] Based on the proportion of the candidate trajectory point in all candidate paths corresponding to the first trajectory point and the second trajectory point, it is determined whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0020] Wherein, determining the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle comprises:

[0021] Based on the initial trajectory of the target vehicle, a candidate path between a first trajectory point and a second trajectory point is determined; the candidate path includes candidate trajectory points; the first trajectory point is a previous trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is a subsequent trajectory point;

[0022] Based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point, it is determined whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0023] Wherein, determining the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle comprises:

[0024] Based on the initial trajectories corresponding to all collected vehicles corresponding to the preset area and the preset time period, a candidate path between a first trajectory point and a second trajectory point is determined; the candidate path includes candidate trajectory points; the first trajectory point is the previous trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is the next trajectory point;

[0025] Based on the proportion of the candidate trajectory point in all candidate paths corresponding to the first trajectory point and the second trajectory point, it is determined whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0026] Wherein, determining the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle comprises:

[0027] Based on the initial trajectories corresponding to all collected vehicles corresponding to the preset area and the preset time period, a candidate path between a first trajectory point and a second trajectory point is determined; the candidate path includes candidate trajectory points; the first trajectory point is the previous trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is the next trajectory point;

[0028] Based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point, it is determined whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0029] Wherein, based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point, determining whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point comprises:

[0030] In response to the probability that the candidate trajectory point is after the first trajectory point and the probability that the second trajectory point is after the candidate trajectory point are both greater than a first threshold and the probability that the second trajectory point is after the first trajectory point, the candidate trajectory point is determined to be an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0031] Wherein, determining the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle comprises:

[0032] Based on searching the first trajectory point and the second trajectory point in a preset map corresponding to the preset area, a preset number of candidate paths corresponding to the first trajectory point and the second trajectory point are determined; the candidate paths include candidate trajectory points; the first trajectory point is the previous trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is the next trajectory point;

[0033] Based on the number of candidate trajectory points included in the candidate path and the running time corresponding to the candidate path, it is determined whether each candidate trajectory point is used as an intermediate trajectory point.

[0034] Wherein, based on the number of candidate trajectory points included in the candidate path and the running time corresponding to the candidate path, determining whether each candidate trajectory point is used as an intermediate trajectory point includes:

[0035] determining the shortest running time among a preset number of candidate paths;

[0036] Determine the confidence of each candidate path based on the ratio of the running time of the candidate path to the shortest running time and the number of candidate trajectory points included in the candidate path;

[0037] Based on the confidence of the candidate path, determine the confidence of each candidate trajectory point;

[0038] Based on the confidence of the candidate trajectory point, it is determined whether the candidate trajectory point is used as an intermediate trajectory point.

[0039] Wherein, determining the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle comprises:

[0040] The trajectory completion deep network is used to complete the initial trajectory of the target vehicle to obtain the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle.

[0041] Wherein, based on the position information and the collection time corresponding to the first trajectory point and the second trajectory point respectively and the position information of the intermediate trajectory point, determining the time range of the target vehicle passing through the intermediate trajectory point includes:

[0042] Based on the initial trajectory of each collected vehicle, the running time from the first trajectory point to the intermediate trajectory point is counted;

[0043] Based on the initial trajectory of each collected vehicle, the running time from the middle trajectory point to the second trajectory point is counted;

[0044] Based on the running time from the first trajectory point to the intermediate trajectory point and the running time from the intermediate trajectory point to the second trajectory point, a time range for the target vehicle to pass the intermediate trajectory point is determined.

[0045] Wherein, based on the running time from the first trajectory point to the intermediate trajectory point and the running time from the intermediate trajectory point to the second trajectory point, determining the time range of the target vehicle passing the intermediate trajectory point includes:

[0046] Select the shortest time and upper quartile of the first trajectory point running to the middle trajectory point to obtain the first shortest time and the first upper quartile;

[0047] Select the shortest time and upper quartile of the middle trajectory point running to the second trajectory point, and obtain the second shortest time and the second upper quartile;

[0048] Compare the sum of the first shortest time and the second shortest time with the running time from the first trajectory point to the second trajectory point;

[0049] In response to the fact that the sum of the first shortest time and the second shortest time does not exceed the running time from the first trajectory point to the second trajectory point, the time range for the target vehicle to pass the intermediate trajectory point is determined based on the difference between the sum of the collection time and the first upper quartile value of the first trajectory point and the collection time and the second upper quartile value of the second trajectory point.

[0050] Wherein, determining the time range of the target vehicle passing the intermediate trajectory point based on the sum of the acquisition time of the first trajectory point and the first upper quartile value and the difference between the acquisition time of the second trajectory point and the second upper quartile value includes:

[0051] In response to the sum of the collection time of the first trajectory point and the first upper quartile value not exceeding the difference between the collection time of the second trajectory point and the second upper quartile value, it is determined that the collection time corresponding to the target vehicle at the intermediate trajectory point is not less than the sum of the collection time of the first trajectory point and the first shortest time, and does not exceed the sum of the collection time of the first trajectory point and the first upper quartile value, or the collection time corresponding to the target vehicle at the intermediate trajectory point is not less than the difference between the collection time of the second trajectory point and the second upper quartile value, and does not exceed the difference between the collection time of the second trajectory point and the second shortest time;

[0052] In response to the fact that the sum of the collection time of the first trajectory point and the first upper quartile value is not less than the difference between the collection time of the second trajectory point and the second upper quartile value, it is determined that the collection time corresponding to the target vehicle at the intermediate trajectory point is not less than the sum of the collection time of the first trajectory point and the first shortest time, and does not exceed the sum of the collection time of the second trajectory point and the second shortest time.

[0053] Among them, all collected vehicles within the time range of the intermediate trajectory points are compared with the target vehicle for similarity, and the vehicle similarity corresponding to each collected vehicle is determined, including:

[0054] Based on the license plate number information and license plate attribute information respectively corresponding to the collected vehicle and the target vehicle, the vehicle similarity respectively corresponding to each collected vehicle and the target vehicle is determined.

[0055] Wherein, based on the license plate number information and license plate attribute information corresponding to the collected vehicle and the target vehicle respectively, determining the vehicle similarity corresponding to each collected vehicle and the target vehicle respectively includes:

[0056] Determine the similarity of the license plate numbers based on the license plate number information of the collected vehicle and the license plate number information of the target vehicle;

[0057] Based on the license plate attribute information of the collected vehicle and the license plate attribute information of the target vehicle, determining the license plate attribute similarity;

[0058] Based on the similarity of the license plate numbers and the similarity of the license plate attributes, the vehicle similarity is determined.

[0059] Wherein, the license plate number information includes a license plate number character sequence;

[0060] Based on the license plate number information of the collected vehicle and the license plate number information of the target vehicle, the similarity of the license plate numbers is determined, including:

[0061] Compare the similarity of each character in the license plate number character sequence of the collected vehicle with the characters in the corresponding position in the license plate number character sequence of the target vehicle to determine the similarity of the characters in each position;

[0062] Determine whether two characters in the same position are a preset easily confused character pair;

[0063] In response to two characters in the same position being a preset easily confused character pair, assigning a preset weight to the character similarity of the position;

[0064] The similarity of the license plate number is determined based on the sum of the character similarities corresponding to each digit and the preset weights.

[0065] In order to solve the above technical problems, the second technical solution adopted by the present invention is: to provide a trajectory completion device, the trajectory completion device comprising:

[0066] An acquisition module, used to acquire an initial trajectory of the target vehicle corresponding to a preset area and a preset time period, the initial trajectory including a first trajectory point and a second trajectory point that are adjacent and arranged in sequence;

[0067] An analysis module, used for determining an intermediate trajectory point between two adjacent trajectory points in an initial trajectory of the target vehicle;

[0068] A comparison module is used to compare the similarity of all collected vehicles with intermediate trajectory points within a preset time period with the target vehicle to determine the vehicle similarity corresponding to each collected vehicle;

[0069] The restoration module is used to add the intermediate trajectory points to the initial trajectory of the target vehicle in response to the vehicle similarity of the collected vehicle exceeding a threshold.

[0070] To solve the above technical problems, the third technical solution adopted by the present invention is: providing a terminal, the terminal including a memory, a processor, and a computer program stored in the memory and running on the processor, the processor being used to execute program data to implement the steps in the trajectory completion method in the above first technical solution.

[0071] In order to solve the above technical problems, the fourth technical solution adopted by the present invention is: providing a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the trajectory completion method in the above first technical solution are implemented.

[0072] In order to solve the above technical problems, the second technical solution adopted by the present invention is to provide a linkage magnification changing device, which comprises:

[0073] To solve the above technical problems, the third technical solution adopted by the present invention is: providing a terminal, the terminal including a memory, a processor and a computer program stored in the memory and running on the processor, the processor being used to execute program data to implement the steps in the above-mentioned trajectory completion method.

[0074] In order to solve the above technical problems, the fourth technical solution adopted by the present invention is: providing a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned trajectory completion method are implemented.

[0075] The beneficial effects of the present invention are as follows: different from the prior art, a trajectory completion method, device, terminal and computer-readable storage medium are provided, wherein the trajectory completion method includes obtaining an initial trajectory corresponding to a target vehicle in a preset area and a preset time period, wherein the initial trajectory includes a plurality of adjacent and sequentially arranged trajectory points; determining an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle; performing a similarity comparison between all collected vehicles with the intermediate trajectory point within the preset time period and the target vehicle to determine the vehicle similarity corresponding to each collected vehicle; in response to the vehicle similarity of the collected vehicle exceeding a threshold, the intermediate trajectory point is added to the initial trajectory of the target vehicle. The present application determines an intermediate trajectory point between a first trajectory point and a second trajectory point in the initial trajectory of the target vehicle, and then compares the vehicle collected at the intermediate trajectory point with the target vehicle to determine whether the target vehicle passes through the intermediate trajectory point. When the target vehicle passes through the intermediate trajectory point, the intermediate trajectory point is added to the initial trajectory of the target vehicle, thereby completing the trajectory of the target vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0077] Figure 1 It is a schematic diagram of the flow of the trajectory completion method provided by the present invention;

[0078] Figure 2 yes Figure 1 A schematic flow chart of a first specific embodiment of step S2 in the provided trajectory completion method;

[0079] Figure 3 yes Figure 1A schematic flow chart of a second specific embodiment of step S2 in the provided trajectory completion method;

[0080] Figure 4 yes Figure 1 A schematic flow chart of a third specific embodiment of step S2 in the provided trajectory completion method;

[0081] Figure 5 yes Figure 1 A schematic flow chart of a fourth specific embodiment of step S2 in the provided trajectory completion method;

[0082] Figure 6 yes Figure 1 A schematic flow chart of a fifth specific embodiment of step S2 in the provided trajectory completion method;

[0083] Figure 7 yes Figure 1 A schematic flow chart of a specific embodiment of step S3 in the provided trajectory completion method;

[0084] Figure 8 yes Figure 7 A schematic flow chart of a specific embodiment of step S32 in the provided trajectory completion method;

[0085] Fig. 9 It is a schematic diagram of the framework of an embodiment of a trajectory completion device provided by the present invention;

[0086] Fig.10 It is a schematic diagram of the framework of a terminal embodiment provided by the present invention;

[0087] Fig.11 It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0088] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0089] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0090] The term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship. In addition, "many" in this article means two or more than two.

[0091] In order to enable those skilled in the art to better understand the technical solution of the present invention, a trajectory completion method provided by the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0092] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0093] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations.

[0094] The trajectory completion method provided in the embodiment of the present application can be implemented by the server or the terminal alone, or by the server and the terminal in collaboration. In some embodiments, the terminal or the server can implement the trajectory completion method provided in the embodiment of the present application by running a computer program. For example, the computer program can be a native program or software module in the operating system; it can be a native application (APP, Application), that is, a program that needs to be installed in the operating system to run, such as a client that supports virtual scenes, such as a game APP; it can also be a small program, that is, a program that can be run only by downloading it to a browser environment; it can also be a small program that can be embedded in any APP. In short, the above-mentioned computer program can be an application, module or plug-in in any form.

[0095] The following uses server implementation as an example to illustrate the trajectory completion method provided in the embodiment of the present application.

[0096] See also Figure 1 , Figure 1 It is a flow chart of the trajectory completion method provided by the present invention.

[0097] This embodiment provides a trajectory completion method, which includes the following steps.

[0098] S1: Obtaining an initial trajectory of a target vehicle in a preset area and a preset time period, wherein the initial trajectory includes a plurality of adjacent and sequentially arranged trajectory points.

[0099] S2: Determine an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle.

[0100] S3: performing a similarity comparison between all collected vehicles with intermediate trajectory points within a preset time period and the target vehicle to determine the vehicle similarity corresponding to each collected vehicle.

[0101] S4: In response to the vehicle similarity of the collected vehicle exceeding a threshold, the intermediate trajectory points are added to the initial trajectory of the target vehicle.

[0102] In one embodiment, after the intermediate trajectory points are added to the initial trajectory of the target vehicle, steps S1 to S4 are continued to determine whether there are other intermediate trajectory points between the first trajectory point and the intermediate trajectory point, and whether there are other intermediate trajectory points between the intermediate trajectory point and the second trajectory point, until all the passing trajectory points of the target vehicle are completed.

[0103] In one embodiment, in response to the vehicle similarity of the intermediate trajectory point to all collected vehicles within the preset time period not exceeding the threshold, it is determined that the target vehicle has not passed the intermediate trajectory point, and the intermediate trajectory point does not belong to the trajectory data of the target vehicle. It is continued through steps S1 to S4 to determine whether there is an intermediate trajectory point between the second trajectory point and the adjacent third trajectory point in the initial trajectory of the target vehicle.

[0104] Specifically, the specific implementation method of obtaining the initial trajectory of the target vehicle corresponding to the preset area and the preset time period in step S1 is as follows.

[0105] All arranged sensing devices are obtained. The sensing devices may be image acquisition devices, ETC devices, etc. The sensing devices have corresponding number information and location information. The location information of the specific sensing device is the latitude and longitude information of the sensing device. Each sensing device is used to obtain information of a target object within a preset range. In one embodiment, the target object may be different types of objects such as a vehicle, a vehicle-related object, etc. The information of the target object includes a vehicle license plate or an image containing the target object.

[0106] The target object information collected by all sensing devices arranged in a preset area is obtained, and the target object information of the sensing devices within a preset time period is extracted.

[0107] In a specific embodiment, each image acquisition device in a preset area acquires an image in a corresponding area to obtain an acquired image. All acquired images acquired for the preset area within a preset period are extracted to obtain a captured image set. Each acquired image has a corresponding acquisition time. All acquired images in the captured image set are clustered according to the target object in the acquired image to obtain an image subset of each target object. The acquired images contained in each image subset contain the same target object. According to the acquisition time of the acquired image corresponding to the target object, the acquired images contained in the image subset of the target object are sorted according to the acquisition time to generate an initial trajectory of the target object. Each image acquisition device is respectively regarded as a trajectory point.

[0108] In one embodiment, the target object is a vehicle, and all the collected images in the captured image set are clustered according to the vehicle license plate in the collected image to obtain an image subset of each vehicle license plate. In another embodiment, all the collected images in the captured image set are clustered according to the vehicle license plate and vehicle attributes in the collected image to obtain an image subset of each vehicle license plate.

[0109] In one embodiment, a captured image set corresponding to a preset area and a preset time period is obtained; the captured image set includes multiple captured images containing captured vehicles; target clustering is performed on the captured images in the captured image set to obtain image subsets corresponding to each captured vehicle; and the capture time of the captured images contained in the image subset of the captured vehicle is used to generate an initial trajectory of the captured vehicle.

[0110] Among them, the preset time period can be one month, three days, two hours, etc., which is set according to the actual situation.

[0111] In one embodiment, the initial trajectories of all collected vehicles are traversed, and the initial trajectories of the collected vehicles are sequentially used as the initial trajectory of the target vehicle.

[0112] In one embodiment, any collected vehicle may be selected as a target vehicle, and an ad hoc analysis may be performed on the initial trajectory of the target vehicle and the initial trajectories of other collected vehicles corresponding to a preset area and a preset time period, so as to complete the initial trajectory of the target vehicle.

[0113] Specifically, the specific implementation method of determining the middle trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle in step S2 is as follows.

[0114] In one embodiment, the suspected passing trajectory points of the target vehicle are determined according to the running trajectory of the target vehicle within a preset time period.

[0115] See also Figure 2 , Figure 2 yes Figure 1 A schematic flow chart of a first specific embodiment of step S2 in the provided trajectory completion method.

[0116] S211: Based on the initial trajectory of the target vehicle, determine a candidate path between the first trajectory point and the second trajectory point; the candidate path includes the candidate trajectory point.

[0117] Specifically, the initial trajectory of the target vehicle includes a plurality of continuous trajectory points arranged in sequence according to the acquisition time. The starting trajectory point of the initial trajectory can be used as the first trajectory point, and the next trajectory point adjacent to the starting trajectory point can be used as the second trajectory point.

[0118] Search for multiple candidate paths between the first trajectory point and the second trajectory point in the initial trajectory of the target vehicle. For example, the first trajectory point is trajectory point A, and the second trajectory point is trajectory point C; in the initial trajectory of the target vehicle, the target vehicle runs from trajectory point A through trajectory point B to trajectory point C; the target vehicle runs from trajectory point A through trajectory point D to trajectory point C. That is, the candidate paths between the first trajectory point and the second trajectory point include trajectory point A→trajectory point B→trajectory point C, trajectory point A→trajectory point D→trajectory point C. Among them, the candidate trajectory points are trajectory point B and trajectory point D.

[0119] S212: Determine whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point based on the proportion of the candidate trajectory point in all candidate paths corresponding to the first trajectory point and the second trajectory point.

[0120] In one embodiment, the proportion of each candidate trajectory point in all candidate trajectory points is calculated to determine the passing probability of the target vehicle passing each candidate trajectory point. For example, the passing probability of the target vehicle passing through trajectory point B when running from trajectory point A to trajectory point C is p(B|AC)=1 / 2; the passing probability of the target vehicle passing through trajectory point D when running from trajectory point A to trajectory point C is p(D|AC)=1 / 2.

[0121] The passing probability of the candidate trajectory point is compared with the preset probability to determine whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point. When the passing probability of the candidate trajectory point exceeds the preset probability, the candidate trajectory point is determined to be an intermediate trajectory point between the first trajectory point and the second trajectory point. When the passing probability of the candidate trajectory point does not exceed the preset probability, the candidate trajectory point is determined not to be an intermediate trajectory point between the first trajectory point and the second trajectory point. For example, the preset probability can be 30%, 45%, 50%, etc., which is set according to the actual situation.

[0122] In another embodiment, the suspected passing trajectory points of the target vehicle may also be determined according to the running trajectory corresponding to the target vehicle through another embodiment.

[0123] See also Figure 3 , Figure 3 yes Figure 1 A schematic flow chart of a second specific embodiment of step S2 in the provided trajectory completion method.

[0124] S221: Based on the initial trajectory of the target vehicle, determine a candidate path between the first trajectory point and the second trajectory point; the candidate path includes the candidate trajectory point.

[0125] Specifically, the initial trajectory of the target vehicle includes a plurality of continuous trajectory points arranged in sequence according to the acquisition time. The starting trajectory point of the initial trajectory can be used as the first trajectory point, and the next trajectory point adjacent to the starting trajectory point can be used as the second trajectory point.

[0126] Search for multiple candidate paths between the first trajectory point and the second trajectory point in the initial trajectory of each target vehicle. Among them, the candidate trajectory is a trajectory running from the first trajectory point to the second trajectory point. For example, the first trajectory point is trajectory point A, and the second trajectory point is trajectory point C. In the initial trajectory of the acquisition vehicle, the acquisition vehicle runs from trajectory point A to trajectory point C through trajectory point B; and the acquisition vehicle runs from trajectory point A to trajectory point C through trajectory point D. That is, the candidate paths between the first trajectory point and the second trajectory point include trajectory point A→trajectory point B→trajectory point C, trajectory point A→trajectory point D→trajectory point C. Among them, the candidate trajectory points are trajectory point B and trajectory point D.

[0127] S222: Determine whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point.

[0128] In one embodiment, in response to the probability that the candidate trajectory point is after the first trajectory point and the probability that the second trajectory point is after the candidate trajectory point are both greater than a first threshold and the probability that the second trajectory point is after the first trajectory point, the candidate trajectory point is determined to be an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0129] For example, in the initial trajectory of the target vehicle, the initial trajectory of the target vehicle includes continuous trajectory points A, B, C, D and E. Then the probability that the next trajectory point after the target vehicle passes trajectory point A is B is p(B|A)=1, the probability that the next trajectory point after the target vehicle passes trajectory point B is C is p(C|B)=1, the probability that the next trajectory point after the target vehicle passes trajectory point C is D is p(D|C)=1, and the probability that the next trajectory point after the target vehicle passes trajectory point D is E is p(E|D)=1.

[0130] For example, the initial trajectory of the target vehicle includes continuous trajectory points A, B, C, A and E. The probability that the next trajectory point after the target vehicle passes trajectory point A is B is p(B|A)=1 / 2, the probability that the next trajectory point after the target vehicle passes trajectory point B is C is p(C|B)=1, the probability that the next trajectory point after the target vehicle passes trajectory point C is A is p(A|C)=1, and the probability that the next trajectory point after the target vehicle passes trajectory point A is E is p(E|D)=1 / 2.

[0131] For example, when it is determined that the first track point in the initial track of the target vehicle is track point A, and the second track point is track point C, by searching in the initial track of the target vehicle, it is determined that the probability p(B|A)=2 / 3 that the next track point after track point A is B, and the probability p(C|B)=2 / 3 that the next track point after track point B is C, and p(B|A)=2 / 3 and p(C|B)=2 / 3 both meet the corresponding preset requirements, that is, the probability that the next track point after track point A is B and the probability that the next track point after track point B is C are both greater than the first threshold. And the probability that the next track point after track point A is B is greater than the probability that the next track point after track point A is C, and the probability that the next track point after track point B is C is greater than the probability that the next track point after track point A is C.

[0132] In one embodiment, in response to the probability that the candidate trajectory point is after the first trajectory point or the probability that the second trajectory point is after the candidate trajectory point is not greater than a first threshold and the probability that the second trajectory point is after the first trajectory point, it is determined that the candidate trajectory point is not an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0133] The intermediate track points in the above embodiment are suspected passing track points of the target vehicle.

[0134] In one embodiment, the suspected passing trajectory points of the target vehicle are determined according to the running trajectories corresponding to all collected vehicles.

[0135] See also Figure 4 , Figure 4 yes Figure 1 A schematic flow chart of a third specific embodiment of step S2 in the provided trajectory completion method.

[0136] S231: Determine a candidate path between a first trajectory point and a second trajectory point based on initial trajectories corresponding to all collected vehicles in a preset area and a preset time period; the candidate path includes the candidate trajectory point.

[0137] Specifically, the initial trajectory of the target vehicle includes a plurality of continuous trajectory points arranged in sequence according to the acquisition time. The starting trajectory point of the initial trajectory can be used as the first trajectory point, and the next trajectory point adjacent to the starting trajectory point can be used as the second trajectory point.

[0138] A plurality of candidate paths between the first trajectory point and the second trajectory point are searched in the initial trajectory of each collected vehicle, wherein the candidate trajectory is a trajectory running from the first trajectory point to the second trajectory point.

[0139] S232: Determine whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point based on the proportion of the candidate trajectory point in all candidate paths corresponding to the first trajectory point and the second trajectory point.

[0140] In one embodiment, the proportion of each candidate trajectory point in all candidate trajectory points is calculated to determine the probability of the target vehicle passing through each candidate trajectory point. For example, the probability of all collected vehicles passing through trajectory point B from trajectory point A to trajectory point C is p(B|AC)=1 / 2; the probability of all collected vehicles passing through trajectory point D from trajectory point A to trajectory point C is p(D|AC)=1 / 2.

[0141] The passing probability of the candidate trajectory point is compared with the preset probability to determine whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point. When the passing probability of the candidate trajectory point exceeds the preset probability, the candidate trajectory point is determined to be an intermediate trajectory point between the first trajectory point and the second trajectory point. When the passing probability of the candidate trajectory point does not exceed the preset probability, the candidate trajectory point is determined not to be an intermediate trajectory point between the first trajectory point and the second trajectory point. For example, the preset probability can be 30%, 45%, 50%, etc., which is set according to the actual situation.

[0142] In another embodiment, the suspected passing trajectory points of the target vehicle may also be determined according to the running trajectories corresponding to all collected vehicles through another embodiment.

[0143] See also Figure 5 , Figure 5 yes Figure 1 A schematic flow chart of a fourth specific embodiment of step S2 in the provided trajectory completion method.

[0144] S241: Determine a candidate path between a first trajectory point and a second trajectory point based on initial trajectories corresponding to all collected vehicles in a preset area and a preset time period; the candidate path includes the candidate trajectory point.

[0145] Specifically, the initial trajectory of the target vehicle includes a plurality of continuous trajectory points arranged in sequence according to the acquisition time. The starting trajectory point of the initial trajectory can be used as the first trajectory point, and the next trajectory point adjacent to the starting trajectory point can be used as the second trajectory point.

[0146] A plurality of candidate paths between the first trajectory point and the second trajectory point are searched in the initial trajectory of each collected vehicle, wherein the candidate trajectory is a trajectory running from the first trajectory point to the second trajectory point.

[0147] S242: Determine whether the candidate trajectory point is an intermediate trajectory point between the first trajectory point and the second trajectory point based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point.

[0148] In a specific embodiment, in response to the probability that the candidate trajectory point is after the first trajectory point and the probability that the second trajectory point is after the candidate trajectory point are both greater than a first threshold and the probability that the second trajectory point is after the first trajectory point, the candidate trajectory point is determined to be an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0149] For example, when it is determined that the first track point in the initial track of the target vehicle is track point A and the second track point is track point C, by searching in the initial tracks of all collected vehicles, it is determined that the probability p(B|A)=2 / 3 that the next track point after track point A is B and the probability p(C|B)=2 / 3 that the next track point after track point B is C both meet the corresponding preset requirements, that is, the probability that the next track point after track point A is B and the probability that the next track point after track point B is C are both greater than the first threshold. And the probability that the next track point after track point A is B is greater than the probability that the next track point after track point A is C, and the probability that the next track point after track point B is C is greater than the probability that the next track point after track point A is C.

[0150] In one embodiment, in response to the probability that the candidate trajectory point is after the first trajectory point or the probability that the second trajectory point is after the candidate trajectory point is not greater than a first threshold and the probability that the second trajectory point is after the first trajectory point, it is determined that the candidate trajectory point is not an intermediate trajectory point between the first trajectory point and the second trajectory point.

[0151] The intermediate track points in the above embodiment are suspected passing track points of the target vehicle.

[0152] In one embodiment, the suspected passing track points of the target vehicle are determined according to the path marked in the preset map.

[0153] See also Figure 6 , Figure 6 yes Figure 1 A schematic flow chart of a fifth specific embodiment of step S2 in the provided trajectory completion method.

[0154] S251: Based on the first trajectory point and the second trajectory point, a preset map corresponding to a preset area is searched to determine a preset number of candidate paths corresponding to the first trajectory point and the second trajectory point; the candidate paths include the candidate trajectory points.

[0155] In one embodiment, the initial trajectory of the target vehicle includes a plurality of continuous trajectory points arranged in sequence according to the acquisition time. The starting trajectory point of the initial trajectory can be used as the first trajectory point, and the next trajectory point adjacent to the starting trajectory point can be used as the second trajectory point.

[0156] All candidate paths corresponding to the first trajectory point and the second trajectory point are searched in a preset map according to the latitude and longitude information of the first trajectory point and the second trajectory point. According to the running time of each candidate path, a preset number of candidate paths with the shortest running time are selected. The candidate path may include a candidate trajectory point, that is, the candidate path includes the first trajectory point, at least one candidate trajectory point, and the second trajectory point that are adjacent in sequence; or it may not include a candidate trajectory point, that is, the candidate path includes the first trajectory point and the second trajectory point that are adjacent in sequence.

[0157] S252: Determine whether each candidate trajectory point is used as an intermediate trajectory point based on the number of candidate trajectory points included in the candidate path and the running time corresponding to the candidate path.

[0158] In a specific embodiment, the shortest running time among a preset number of candidate paths is determined. The confidence of each candidate path is determined based on the ratio of the running time of the candidate path to the shortest running time and the number of candidate trajectory points included in the candidate path. For example, when the ratio of the running time of the candidate path to the shortest running time is small and the number of candidate trajectory points included in the candidate path is small, the confidence of each candidate path is determined to be higher.

[0159] Based on the confidence of the candidate paths, the confidence of each candidate trajectory point is determined. When multiple candidate trajectories contain the same candidate trajectory point, the confidence of the candidate trajectory point is determined based on the confidence of all candidate paths containing the candidate trajectory point. The more candidate paths a candidate trajectory point belongs to, the higher the confidence of the candidate trajectory point.

[0160] Based on the confidence of the candidate trajectory point, it is determined whether the candidate trajectory point is used as an intermediate trajectory point.

[0161] In response to the confidence of the candidate trajectory point being greater than a preset confidence, it is determined that the candidate trajectory point can be used as an intermediate trajectory point; in response to the confidence of the candidate trajectory point being not greater than the preset confidence, it is determined that the candidate trajectory point is not an intermediate trajectory point of the target vehicle.

[0162] The intermediate track points between two adjacent track points are determined in sequence by the above method, and the intermediate track points between the track points in the initial track and the adjacent intermediate track points are determined, until all the intermediate track points suspected to have been passed by the target vehicle are obtained.

[0163] In one embodiment, a trajectory completion deep network is used to complete the initial trajectory of the target vehicle to obtain an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle. The trajectory completion deep network can be a deep learning algorithm such as Transformer, LSTM (Long Short-Term Memory), etc.

[0164] In one embodiment, the trajectory completion deep network is pre-trained, wherein the trajectory completion deep network includes an individual trajectory completion deep network, or may be a whole trajectory completion deep network.

[0165] In a specific embodiment, the complete trajectory and defect trajectory corresponding to the training target in a preset time period are obtained. That is, the trajectory data formed by the training target passing through each trajectory point is obtained, and the trajectory data of the training target is segmented based on the speed and / or time between two adjacent trajectory points to obtain the complete trajectory of the training target, and some trajectory points in the complete trajectory are blocked to form the defect trajectory of the training target. The input trajectory of the defect trajectory is completed in the deep network to obtain all the preset intermediate trajectory points corresponding to the defect trajectory. The predicted intermediate trajectory points are added to the defect trajectory to form a preset trajectory, and the trajectory completion deep network is trained based on the error between the preset trajectory of the training target and the complete trajectory.

[0166] Specifically, in step S3, all collected vehicles with intermediate trajectory points within a preset time period are compared with the target vehicle for similarity, and a specific implementation method for determining the vehicle similarity corresponding to each collected vehicle is as follows.

[0167] See also Figure 7 , Figure 7 yes Figure 1 A flowchart of a specific embodiment of step S3 in the provided trajectory completion method is provided.

[0168] S31: Based on the position information and collection time corresponding to the first trajectory point and the second trajectory point respectively and the position information of the intermediate trajectory point, determine the time range for the target vehicle to pass the intermediate trajectory point.

[0169] In one embodiment, based on the initial trajectory of each collected vehicle, the running time from the first trajectory point W1 to the intermediate trajectory point S1 is counted; based on the initial trajectory of each collected vehicle, the running time from the intermediate trajectory point S1 to the second trajectory point W2 is counted; based on the running time from the first trajectory point W1 to the intermediate trajectory point S1 and the running time from the intermediate trajectory point S1 to the second trajectory point W2, the time range of the target vehicle passing the intermediate trajectory point S1 is determined.

[0170] In one embodiment, the shortest time from the first trajectory point to the middle trajectory point and the upper quartile are selected to obtain the first shortest time and the first upper quartile Select the shortest time from the middle trajectory point to the second trajectory point and the upper quartile to get the second shortest time and the second upper quartile .

[0171] The sum of the first shortest time and the second shortest time is compared with the running time from the first trajectory point to the second trajectory point.

[0172] In one embodiment, in response to the sum of the first shortest time and the second shortest time exceeding the running time from the first trajectory point to the second trajectory point, it is determined that the target vehicle has not passed the corresponding intermediate trajectory point. , it is determined that the target vehicle has not passed the corresponding intermediate trajectory point.

[0173] In one embodiment, in response to the sum of the first shortest time and the second shortest time not exceeding the running time from the first trajectory point to the second trajectory point, the time range of the target vehicle passing the intermediate trajectory point is determined based on the difference between the sum of the collection time of the first trajectory point and the first upper quartile value and the collection time of the second trajectory point and the second upper quartile value. The time range of the target vehicle passing the intermediate trajectory point is determined based on the sum of the acquisition time and the first upper quartile value of the first trajectory point and the difference between the acquisition time and the second upper quartile value of the second trajectory point.

[0174] In one embodiment, in response to the sum of the acquisition time of the first trajectory point and the first upper quartile value being less than the difference between the acquisition time of the second trajectory point and the second upper quartile value, that is, , then determine that the target vehicle's corresponding collection time T at the middle trajectory point is not less than the sum of the collection time of the first trajectory point and the first shortest time, and does not exceed the sum of the collection time of the first trajectory point and the first upper quartile value, that is, T∈ ; or the target vehicle's corresponding collection time T at the middle trajectory point is not less than the difference between the collection time of the second trajectory point and the second upper quartile value, and does not exceed the difference between the collection time of the second trajectory point and the second shortest time, that is, T∈ .

[0175] In one embodiment, in response to the sum of the acquisition time of the first trajectory point and the first upper quartile value being not less than the difference between the acquisition time of the second trajectory point and the second upper quartile value, that is, ; Then determine that the acquisition time of the target vehicle at the intermediate trajectory point is not less than the sum of the acquisition time of the first trajectory point and the first shortest time, and does not exceed the sum of the acquisition time of the second trajectory point and the second shortest time, that is, T∈ .

[0176] S32: performing a similarity comparison between all collected vehicles within the time range of the intermediate trajectory points and the target vehicle to determine the vehicle similarity corresponding to each collected vehicle.

[0177] In one embodiment, based on the license plate number information and license plate attribute information corresponding to the acquisition vehicle and the target vehicle respectively, the vehicle similarity corresponding to each acquisition vehicle and the target vehicle is determined.

[0178] Please refer to Figure 8 , Figure 8 which Figure 7 is a schematic flowchart of a specific embodiment of step S32 in the provided trajectory completion method.

[0179] S321: Based on the license plate number information of the acquisition vehicle and the license plate number information of the target vehicle, the license plate number similarity is determined.

[0180] Among them, the license plate number information includes the license plate number character sequence.

[0181] Specifically, each character in the license plate number character sequence of the acquisition vehicle is compared with the characters in the corresponding positions in the license plate number character sequence of the target vehicle to determine the character similarity of each position; it is determined whether two characters in the same position are a preset pair of easily confused characters; in response to two characters in the same position being a preset pair of easily confused characters, a preset weight is assigned to the character similarity of the position; based on the sum of the character similarities corresponding to each position and the preset weight, the license plate number similarity is determined. Among them, the preset pair of easily confused characters can be "8 - B", "1 - I", "2 - Z", "5 - S", "7 - T", "3 - B", "4 - A", etc. For example, if the license plate number of the target vehicle is "Zhe A12145" and the license plate number of the acquisition vehicle is "Zhe A12T45", then a preset weight is assigned to the character similarity corresponding to "1" and "T" in the same position to improve the license plate number similarity between the target vehicle and the acquisition vehicle.

[0182] In another embodiment, considering the influence of the inversion between the characters corresponding to the license plate number on the license plate similarity, that is, when there is an inversion between the license plate number of the target vehicle and the license plate number of the acquisition vehicle, a higher weight is assigned to the license plate number similarity corresponding to the target vehicle and the acquisition vehicle. For example, if the license plate number of the target vehicle is "Zhe A12145" and the license plate number of the acquisition vehicle is "Zhe A12154", then a higher weight is assigned to the license plate number similarity corresponding to the target vehicle and the acquisition vehicle to improve the license plate number similarity between the target vehicle and the acquisition vehicle.

[0183] S322: Based on the license plate attribute information of the acquisition vehicle and the license plate attribute information of the target vehicle, the license plate attribute similarity is determined.

[0184] Specifically, the license plate attribute information includes at least one of the number of annual inspection marks, body color, vehicle type, and vehicle brand.

[0185] S323: Determine vehicle similarity based on the license plate number similarity and the license plate attribute similarity.

[0186] Specifically, the vehicle similarity is determined based on the weighted sum of the license plate number similarity and the license plate attribute similarity.

[0187] Specifically, in step S4, in response to the vehicle similarity of the collected vehicle exceeding a threshold, a specific implementation method of adding the intermediate trajectory points to the initial trajectory of the target vehicle is as follows.

[0188] In one embodiment, in response to the vehicle similarity of the captured vehicle exceeding a threshold, the intermediate trajectory points are added to the initial trajectory of the target object, and the initial trajectory of the captured vehicle is fused with the initial trajectory of the target object to obtain an updated trajectory of the target object.

[0189] In one embodiment, in response to the vehicle similarities of multiple collected vehicles exceeding a threshold, the vehicles are sorted in descending order according to their vehicle similarities and a preset number of collected images with the highest sorting order are selected for display, so that the collected vehicle that is most similar to the target vehicle can be selected from the preset number of collected vehicles according to actual conditions, so that the initial trajectory of the collected vehicle and the initial trajectory of the target object are fused to obtain an updated trajectory of the target object.

[0190] In another embodiment, all abnormal license plate data within a preset area and a preset time period are counted. The abnormal license plate data includes the total number of abnormal license plates, the total number of processed license plates, the total number of confirmed abnormal license plates, etc.

[0191] In one embodiment, in response to the vehicle similarity of the collected vehicle not exceeding a threshold, it is determined that the target vehicle has not passed the intermediate trajectory point.

[0192] In the trajectory completion method provided in this embodiment, the initial trajectory corresponding to the target vehicle in the preset area and the preset time period is obtained, and the initial trajectory includes a plurality of adjacent and sequentially arranged trajectory points; the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle is determined; all collected vehicles with the intermediate trajectory point within the preset time period are compared with the target vehicle for similarity, and the vehicle similarity corresponding to each collected vehicle is determined; in response to the vehicle similarity of the collected vehicle exceeding the threshold, the intermediate trajectory point is added to the initial trajectory of the target vehicle. The present application determines the intermediate trajectory point between the first trajectory point and the second trajectory point in the initial trajectory of the target vehicle, and then compares the vehicle collected at the intermediate trajectory point with the target vehicle to determine whether the target vehicle passes through the intermediate trajectory point. When the target vehicle passes through the intermediate trajectory point, the intermediate trajectory point is added to the initial trajectory of the target vehicle, thereby completing the trajectory of the target vehicle.

[0193] See also Fig. 9 , Fig. 9It is a schematic diagram of the framework of an embodiment of a trajectory completion device provided by the present invention.

[0194] This embodiment provides a trajectory completion device 60 , which includes an acquisition module 61 , an analysis module 62 , a comparison module 63 and a restoration module 64 .

[0195] The acquisition module 61 is used to acquire an initial trajectory of the target vehicle in a preset area and a preset time period, and the initial trajectory includes a plurality of adjacent and sequentially arranged trajectory points.

[0196] The analysis module 62 is used to determine an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle.

[0197] The comparison module 63 is used to compare the similarity of all collected vehicles with intermediate trajectory points within a preset time period with the target vehicle to determine the vehicle similarity corresponding to each collected vehicle.

[0198] The restoration module 64 is configured to add the intermediate trajectory points to the initial trajectory of the target vehicle in response to the vehicle similarity of the collected vehicle exceeding a threshold.

[0199] The trajectory completion device provided in this embodiment determines the intermediate trajectory point between the first trajectory point and the second trajectory point in the initial trajectory of the target vehicle, and then compares the vehicle collected at the intermediate trajectory point with the target vehicle to determine whether the target vehicle passes through the intermediate trajectory point. When the target vehicle passes through the intermediate trajectory point, the intermediate trajectory point is added to the initial trajectory of the target vehicle, thereby completing the trajectory of the target vehicle.

[0200] See also Fig.10 , Fig.10 It is a schematic diagram of the framework of a terminal embodiment provided by the present invention.

[0201] This embodiment provides a terminal 80, which includes a memory 81 and a processor 82 coupled to each other, and the processor 82 is used to execute program instructions stored in the memory 81 to implement the steps of any of the above trajectory completion method embodiments. In a specific implementation scenario, the terminal 80 may include but is not limited to: a microcomputer, a server, and in addition, the terminal 80 may also include a mobile device such as a laptop computer and a tablet computer, which is not limited here.

[0202] Specifically, the processor 82 is used to control itself and the memory 81 to implement the steps of any of the above-mentioned trajectory completion method embodiments. The processor 82 can also be called a CPU (Central Processing Unit). The processor 82 may be an integrated circuit chip with signal processing capabilities. The processor 82 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 82 can be implemented by an integrated circuit chip.

[0203] In the above scheme, the trajectory completion method includes obtaining an initial trajectory corresponding to the target vehicle in a preset area and a preset time period, the initial trajectory including a plurality of adjacent and sequentially arranged trajectory points; determining an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle; performing a similarity comparison between all collected vehicles within the preset time period and the target vehicle to determine the vehicle similarity corresponding to each collected vehicle; in response to the vehicle similarity of the collected vehicle exceeding a threshold, the intermediate trajectory point is supplemented to the initial trajectory of the target vehicle.

[0204] See also Fig.11 , Fig.11 It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium provided by the present invention.

[0205] This embodiment provides a computer-readable storage medium 90 , which stores program instructions 901 that can be executed by a processor, and the program instructions 901 are used to implement the steps of any of the above-mentioned trajectory completion method embodiments.

[0206] In the above scheme, the trajectory completion method includes obtaining an initial trajectory corresponding to the target vehicle in a preset area and a preset time period, the initial trajectory including a plurality of adjacent and sequentially arranged trajectory points; determining an intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle; performing a similarity comparison between all collected vehicles within the preset time period and the target vehicle to determine the vehicle similarity corresponding to each collected vehicle; in response to the vehicle similarity of the collected vehicle exceeding a threshold, the intermediate trajectory point is supplemented to the initial trajectory of the target vehicle.

[0207] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0208] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0209] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0210] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0211] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

[0212] The above are only implementation modes of the present invention, and are not intended to limit the patent protection scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A trajectory completion method, characterized in that: The trajectory completion method comprises: Acquire an initial trajectory of the target vehicle in a preset area and a preset time period, wherein the initial trajectory includes a plurality of adjacent and sequentially arranged trajectory points; Determine an intermediate track point between two adjacent track points in the initial track of the target vehicle; the two adjacent track points include an adjacent first track point and a second track point; Compare the similarity of all collected vehicles within the preset time period of the intermediate track point with the target vehicle to determine the vehicle similarity corresponding to each collected vehicle; In response to the vehicle similarity of the collected vehicle exceeding a threshold, the intermediate trajectory point is added to the initial trajectory of the target vehicle; The step of comparing the similarity of all collected vehicles of the intermediate track points within the preset time period with the target vehicle to determine the vehicle similarity corresponding to each collected vehicle includes: Based on the initial trajectory of each collected vehicle, counting the running time from the first trajectory point to the intermediate trajectory point; based on the initial trajectory of each collected vehicle, counting the running time from the intermediate trajectory point to the second trajectory point; Select the shortest time and the upper quartile from the first trajectory point to the middle trajectory point to obtain the first shortest time and the first upper quartile; select the shortest time and the upper quartile from the middle trajectory point to the second trajectory point to obtain the second shortest time and the second upper quartile; comparing the sum of the first shortest time and the second shortest time with the running time from the first trajectory point to the second trajectory point; In response to the sum of the first shortest time and the second shortest time not exceeding the running time from the first trajectory point to the second trajectory point, determining the time range for the target vehicle to pass the intermediate trajectory point based on the difference between the sum of the collection time and the first upper quartile value of the first trajectory point and the collection time and the second upper quartile value of the second trajectory point; A similarity comparison is performed between all the collected vehicles within the time range of the intermediate trajectory points and the target vehicle to determine the vehicle similarity corresponding to each of the collected vehicles.

2. The trajectory completion method according to claim 1, characterized in that: The obtaining of the initial trajectory of the target vehicle in the preset area and the preset time period includes: Acquire a captured image set corresponding to the preset area and the preset time period; the captured image set includes a plurality of captured images including the captured vehicle; Performing target clustering on the collected images in the captured image set to obtain image subsets corresponding to each of the collected vehicles; generating an initial trajectory of the acquisition vehicle based on the acquisition time of the acquisition image contained in the image subset of the acquisition vehicle; The initial trajectories of the collected vehicles are sequentially used as the initial trajectories of the target vehicle.

3. The trajectory completion method according to claim 1, characterized in that: The determining of an intermediate track point between two adjacent track points in the initial track of the target vehicle comprises: Based on the initial trajectory of the target vehicle, determine a candidate path between a first trajectory point and a second trajectory point; the candidate path includes a candidate trajectory point; the first trajectory point is a previous trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is a later trajectory point; based on the proportion of the candidate trajectory point in all the candidate paths corresponding to the first trajectory point and the second trajectory point, determine whether the candidate trajectory point is the intermediate trajectory point between the first trajectory point and the second trajectory point; Or, based on the initial trajectory of the target vehicle, determine a candidate path between a first trajectory point and a second trajectory point; the candidate path includes a candidate trajectory point; the first trajectory point is the first trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is the second trajectory point; based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point, determine whether the candidate trajectory point is the intermediate trajectory point between the first trajectory point and the second trajectory point; Or, based on the initial trajectories respectively corresponding to all the collected vehicles corresponding to the preset area and the preset time period, determine a candidate path between a first trajectory point and a second trajectory point; the candidate path includes a candidate trajectory point; the first trajectory point is the first trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is the second trajectory point; based on the proportion of the candidate trajectory point in all the candidate paths corresponding to the candidate trajectory point between the first trajectory point and the second trajectory point, determine whether the candidate trajectory point is the intermediate trajectory point between the first trajectory point and the second trajectory point; Or, based on the initial trajectories respectively corresponding to all the collected vehicles corresponding to the preset area and the preset time period, determine a candidate path between a first trajectory point and a second trajectory point; the candidate path includes candidate trajectory points; the first trajectory point is the first trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is the second trajectory point; based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point, determine whether the candidate trajectory point is the intermediate trajectory point between the first trajectory point and the second trajectory point; Or, based on searching the first trajectory point and the second trajectory point in the preset map corresponding to the preset area, determining a preset number of candidate paths corresponding to the first trajectory point and the second trajectory point; the candidate path includes candidate trajectory points; the first trajectory point is the first trajectory point of two adjacent trajectory points in the initial trajectory of the target vehicle, and the second trajectory point is the second trajectory point; based on the number of the candidate trajectory points included in the candidate path and the running time corresponding to the candidate path, determining whether each candidate trajectory point is used as the intermediate trajectory point; Alternatively, a trajectory completion deep network is used to complete the initial trajectory of the target vehicle to obtain the intermediate trajectory point between two adjacent trajectory points in the initial trajectory of the target vehicle.

4. The trajectory completion method according to claim 3, characterized in that: The determining whether the candidate trajectory point is the intermediate trajectory point between the first trajectory point and the second trajectory point based on the probability that the candidate trajectory point is after the first trajectory point, the probability that the second trajectory point is after the candidate trajectory point, and the probability that the second trajectory point is after the first trajectory point includes: In response to the probability that the candidate trajectory point is after the first trajectory point and the probability that the second trajectory point is after the candidate trajectory point are both greater than a first threshold and the probability that the second trajectory point is after the first trajectory point, determining that the candidate trajectory point is the intermediate trajectory point between the first trajectory point and the second trajectory point.

5. The trajectory completion method according to claim 3, characterized in that: The determining, based on the number of the candidate trajectory points included in the candidate path and the running time corresponding to the candidate path, whether each candidate trajectory point is used as the intermediate trajectory point includes: Determining the shortest running time among the preset number of candidate paths; Determining the confidence of each candidate path based on the ratio of the running time of the candidate path to the shortest running time and the number of the candidate trajectory points included in the candidate path; Determining the confidence of each of the candidate trajectory points based on the confidence of the candidate path; Based on the confidence of the candidate trajectory point, it is determined whether the candidate trajectory point is used as the intermediate trajectory point.

6. The trajectory completion method according to claim 1, characterized in that: The step of determining the time range of the target vehicle passing through the intermediate trajectory point based on the sum of the acquisition time and the first upper quartile value of the first trajectory point and the difference between the acquisition time and the second upper quartile value of the second trajectory point includes: In response to the sum of the collection time of the first trajectory point and the first upper quartile value being less than the difference between the collection time of the second trajectory point and the second upper quartile value, it is determined that the collection time corresponding to the target vehicle at the intermediate trajectory point is not less than the sum of the collection time of the first trajectory point and the first shortest time, and does not exceed the sum of the collection time of the first trajectory point and the first upper quartile value, or the collection time corresponding to the target vehicle at the intermediate trajectory point is not less than the difference between the collection time of the second trajectory point and the second upper quartile value, and does not exceed the difference between the collection time of the second trajectory point and the second shortest time; In response to the fact that the sum of the collection time of the first trajectory point and the first upper quartile value is not less than the difference between the collection time of the second trajectory point and the second upper quartile value, it is determined that the collection time of the target vehicle corresponding to the intermediate trajectory point is not less than the sum of the collection time of the first trajectory point and the first shortest time, and does not exceed the sum of the collection time of the second trajectory point and the second shortest time.

7. The trajectory completion method according to claim 1, characterized in that: The comparing the similarity of all the collected vehicles within the time range of the intermediate trajectory points with the target vehicle to determine the vehicle similarity corresponding to each collected vehicle includes: Based on the license plate number information and license plate attribute information respectively corresponding to the collected vehicles and the target vehicles, the vehicle similarity respectively corresponding to each of the collected vehicles and the target vehicles is determined.

8. The trajectory completion method according to claim 7, characterized in that: The determining the vehicle similarity corresponding to each of the collected vehicles and the target vehicle based on the license plate number information and the license plate attribute information corresponding to the collected vehicles and the target vehicle respectively includes: Determining the license plate number similarity based on the license plate number information of the collected vehicle and the license plate number information of the target vehicle; Determining the license plate attribute similarity based on the license plate attribute information of the collected vehicle and the license plate attribute information of the target vehicle; The vehicle similarity is determined based on the license plate number similarity and the license plate attribute similarity.

9. The trajectory completion method according to claim 8, characterized in that: The license plate number information includes a license plate number character sequence; The determining of the license plate number similarity based on the license plate number information of the collected vehicle and the license plate number information of the target vehicle includes: Comparing the similarity of each character in the license plate number character sequence of the collected vehicle with each character in the corresponding position in the license plate number character sequence of the target vehicle to determine the similarity of the characters in each position; Determining whether the two characters in the same position are a preset easily confused character pair; In response to the two characters in the same position being the preset easily confused character pair, assigning a preset weight to the similarity of the characters in the position; The similarity of the license plate number is determined based on the sum of the character similarities corresponding to each digit and the preset weight.

10. A trajectory completion device, characterized in that: The trajectory completion device comprises: An acquisition module, used to acquire an initial trajectory of the target vehicle in a preset area and a preset time period, wherein the initial trajectory includes a plurality of trajectory points that are adjacent and arranged in sequence; An analysis module, used to determine an intermediate track point between two adjacent track points in the initial track of the target vehicle; the two adjacent track points include an adjacent first track point and a second track point; The comparison module is used to compare the similarity of all the collected vehicles within the preset time period of the intermediate trajectory point with the target vehicle to determine the vehicle similarity corresponding to each of the collected vehicles; it is also used to count the running time from the first trajectory point to the intermediate trajectory point based on the initial trajectory of each of the collected vehicles; based on the initial trajectory of each of the collected vehicles, count the running time from the intermediate trajectory point to the second trajectory point; select the shortest time and the upper quartile from the first trajectory point to the intermediate trajectory point to obtain the first shortest time and the first upper quartile; select the shortest time and the upper quartile from the intermediate trajectory point to the second trajectory point to obtain the second shortest time and the second upper quartile; comparing the sum of the first shortest time and the second shortest time with the running time from the first trajectory point to the second trajectory point; in response to the sum of the first shortest time and the second shortest time not exceeding the running time from the first trajectory point to the second trajectory point, determining the time range of the target vehicle passing the intermediate trajectory point based on the sum of the collection time of the first trajectory point and the first upper quartile value and the difference between the collection time and the second upper quartile value of the second trajectory point; performing a similarity comparison between all the collected vehicles within the time range of the intermediate trajectory point and the target vehicle, and determining the vehicle similarity corresponding to each of the collected vehicles; The restoration module is used for adding the intermediate trajectory points to the initial trajectory of the target vehicle in response to the vehicle similarity of the collected vehicle exceeding a threshold.

11. A terminal, characterized in that: The terminal includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor is used to execute program data to implement the steps in the trajectory completion method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps in the trajectory completion method according to any one of claims 1 to 9 are implemented.

Images