Trajectory completion method and related apparatus, device and medium

Abstract

The present application discloses a trajectory completion method and related devices, equipment and media, the trajectory completion method comprises: obtaining the first trajectory of the target object; wherein the first trajectory contains the real point positions successively passed by the target object; displaying the first trajectory on the electronic map; in response to the box selection instruction on the electronic map, determining the point pair related to the box selection region in the first trajectory, and selecting the real point positions successively passed in the point pair as the first point and the second point respectively, and completing the suspected point positions between the first point and the second point in the box selection region to obtain the second trajectory; display the second trajectory on the electronic map. The above scheme can meet the individual needs of users for trajectory completion as much as possible, so as to help improve the application effect of the user in the subsequent downstream task based on the moving trajectory.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a trajectory completion method and related apparatus, equipment and medium. Background Technology

[0002] With the increasing prevalence of data collection devices such as cameras at intersections and toll stations, the analysis of data collected by these devices at various locations to determine the movement trajectory of specific objects has been widely applied in various scenarios.

[0003] However, in practical applications, the movement trajectories obtained through the above data analysis may contain omissions. For example, a specific object may actually pass through a certain point, but this may not be reflected in the movement trajectory, directly affecting the application effect of subsequent user tasks. Therefore, how to meet users' personalized needs for trajectory completion as much as possible to improve the application effect of movement trajectories in subsequent tasks has become an urgent problem to be solved. Summary of the Invention

[0004] The main technical problem addressed by this application is to provide a trajectory completion method and related devices, equipment, and media that can meet users' personalized needs for trajectory completion as much as possible, thereby helping to improve the application effect of users based on movement trajectories in subsequent downstream tasks.

[0005] To address the aforementioned issues, a first aspect of this application provides a trajectory completion method, comprising: acquiring a first trajectory of a target object; wherein the first trajectory includes real points traversed sequentially by the target object; displaying the first trajectory on an electronic map; responding to a selection command on the electronic map, determining point pairs in the first trajectory related to the selected area, selecting sequentially traversed real points in the point pairs as the first point and the second point, respectively, and completing suspected points located between the first point and the second point within the selected area to obtain a second trajectory; and displaying the second trajectory on the electronic map.

[0006] To address the aforementioned problems, a second aspect of this application provides a trajectory completion device, comprising: a trajectory acquisition module, a first display module, a trajectory completion module, and a second display module. The trajectory acquisition module is used to acquire a first trajectory of a target object, wherein the first trajectory includes the actual points traversed sequentially by the target object. The first display module is used to display the first trajectory on an electronic map. The trajectory completion module is used to, in response to a selection command on the electronic map, determine point pairs in the first trajectory related to the selected area, select the actual points traversed sequentially in the point pairs as the first point and the second point, respectively, and complete the suspected points located between the first point and the second point within the selected area to obtain a second trajectory. The second display module is used to display the second trajectory on the electronic map.

[0007] To address the aforementioned issues, a third aspect of this application provides an electronic device including a display, a memory, and a processor. The display and the memory are respectively coupled to the processor. The memory stores program instructions, and the processor executes the program instructions to implement the trajectory completion method described in the first aspect.

[0008] To address the aforementioned problems, a fourth aspect of this application provides a computer-readable storage medium storing program instructions executable by a processor, the program instructions being used to implement the trajectory completion method described in the first aspect.

[0009] The above scheme obtains the first trajectory of the target object, which includes the actual points the target object passes through sequentially. This first trajectory is then displayed on an electronic map. In response to a selection command on the electronic map, it identifies point pairs related to the selected area within the first trajectory, selects the actual points passed sequentially within each pair as the first and second points, and completes the selection with suspected points located between the first and second points to obtain the second trajectory. This second trajectory is then displayed on the electronic map. Since the trajectory completion process is triggered by the user's selection command on the electronic map, and suspected points are completed within the selected area and between point pairs related to the selected area, it can meet the user's personalized needs for trajectory completion as much as possible. This improves the completeness of the target object's movement trajectory within the user's area of ​​interest, and helps improve the application effect of the movement trajectory in subsequent downstream tasks. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating an embodiment of the trajectory completion method of this application; Figure 2a This is a schematic diagram of an embodiment of displaying a trajectory on an electronic map; Figure 2b This is a schematic diagram of another embodiment of displaying a trajectory on an electronic map; Figure 2c This is a schematic diagram of yet another embodiment of displaying a trajectory on an electronic map; Figure 2d This is a schematic diagram of yet another embodiment of displaying a trajectory on an electronic map; Figure 2e This is a schematic diagram of yet another embodiment of displaying a trajectory on an electronic map; Figure 2f This is a schematic diagram of yet another embodiment of displaying a trajectory on an electronic map; Figure 3 This is a flowchart illustrating another embodiment of the trajectory completion method of this application; Figure 4 This is a schematic diagram of the framework of an embodiment of the trajectory completion device of this application; Figure 5 This is a schematic diagram of the framework of an embodiment of the electronic device of this application; Figure 6 This is a schematic diagram of a framework of an embodiment of the computer-readable storage medium of this application. Detailed Implementation

[0011] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0012] In the following description, specific details such as particular system architectures, interfaces, and technologies are presented for illustrative purposes rather than for limiting purposes, in order to provide a thorough understanding of this application.

[0013] In this paper, the terms "system" and "network" are often used interchangeably. The term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "many" in this paper means two or more.

[0014] Please see Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the trajectory completion method of this application. Specifically, it may include the following steps: Step S11: Obtain the first trajectory of the target object.

[0015] In this embodiment, the first trajectory includes the points the target object passes through sequentially. It should be noted that in practical applications, information collection devices such as cameras can be deployed at intersections, toll booths, and other locations. These devices can be used to analyze the information data collected from target objects such as vehicles in downstream tasks such as illegal driving, obtaining the expected analysis results for the downstream task. Furthermore, the geographical locations (e.g., latitude and longitude) of the locations where information collection devices are deployed can be pre-stored. Therefore, when an information collection device collects information data about a target object (e.g., after performing target detection on a camera's captured image and determining that a target object exists in the captured image, it can be determined that the camera has collected information data about the target object, i.e., the target object passed through the camera's location), the geographical location of the information collection device can be recorded as the target object's points of travel. Thus, the first trajectory of the target object can be formed based on the points the target object passes through sequentially.

[0016] In one implementation scenario, after obtaining the first trajectory, any two adjacent points on the trajectory can be analyzed. Specifically, the distance between two adjacent points can be obtained first, along with the difference in travel time between them. The former is then divided by the latter to obtain the average speed of the target object between the two adjacent points. Based on this, the average speed can be checked. If the average speed exceeds a speed threshold, it can be determined that at least one of the two adjacent points on the first trajectory is abnormal. The information data collected from the target object at each of the two adjacent points is then analyzed to identify the abnormal point among the two adjacent points. For example, taking the captured images as information data, the captured images at two adjacent real-time locations can be compared with the preset images of the target object. If the captured image at a certain real-time location fails to match the preset image, the real-time location can be considered an abnormal location and marked in the first trajectory. For example, an abnormal label can be added to the captured image of the real-time location. In response to the selection command of the real-time location, the captured image of the real-time location can be displayed, and the abnormal label, the location name of the real-time location (e.g., intersection name) and the shooting time can be displayed on the captured image, thereby showing the abnormal data evidence chain.

[0017] In one implementation scenario, the first trajectory of the target object can specifically include the actual points that the target object passes through sequentially and continuously. Specifically, to ensure the continuity of the first trajectory as much as possible, after obtaining the original trajectory of the target object, the difference in travel time between any two adjacent points in the original trajectory can be obtained as the actual time for the target object to travel from one of the two adjacent points to the other. The theoretical time for the target object to travel from one of the two adjacent points to the other can also be obtained. If the actual time is much greater than the theoretical time (e.g., the actual time is several times the theoretical time), then it can be considered that the two adjacent points should actually belong to different trajectories, and therefore the original trajectory can be segmented. For example, if the target object passes through points A and B sequentially, but the interval between the travel time of point A and the travel time of point B is greater than a time threshold (e.g., 1 hour), then the original trajectory can be segmented between points A and B to assign points A and B to different first trajectories of the target object.

[0018] In one implementation scenario, after obtaining several first trajectories of the target object, each first trajectory can be deduplicated. For example, if X consecutive path points in a first trajectory are the same, only the first and last of those X consecutive path points can be retained. Furthermore, after deduplication, the total number of different path points in the deduplicated first trajectory can be counted. If the total number is less than a threshold (e.g., 2), the first trajectory can be discarded.

[0019] Step S12: Display the first trajectory on the electronic map.

[0020] Specifically, based on the geographical locations of each point along the first trajectory, these points can be marked on an electronic map. Then, according to the order in which these points are passed along the first trajectory, the marked points on the electronic map are connected to display the first trajectory on the electronic map. It should be noted that the electronic map can include, but is not limited to, road network maps, satellite maps, and 3D maps, etc., and is not limited here. Please refer to [reference needed]. Figure 2a , Figure 2a This is a schematic diagram of an embodiment of displaying a trajectory on an electronic map. For example... Figure 2a As shown in the figure, the dark solid circles represent the various points of the target object, i.e. Figure 2a The actual meridian points A, B, C, and D are connected sequentially to display the first trajectory on the electronic map. Of course, Figure 2a The illustration shows only one possible implementation in practical application and does not limit the specific way the first trajectory is displayed on the electronic map. For example, in addition to solid circles, other point markers such as triangles and squares can also be used.

[0021] Step S13: In response to the box selection command on the electronic map, determine the point pairs related to the boxed area in the first trajectory, select the actual points passed in the point pairs as the first point and the second point respectively, and fill in the suspected points located between the first point and the second point in the boxed area to obtain the second trajectory.

[0022] In one implementation scenario, in response to a box selection command on an electronic map, it can first detect whether the box selection area contains actual points in the first trajectory, and select adjacent actual points in the first trajectory as point pairs based on the total number of actual points in the box selection area.

[0023] In a specific implementation scenario, if the selected area does not contain any actual points, the distances from each actual point in the first trajectory to the boundary of the selected area can be obtained. The total distance from each pair of adjacent actual points in the first trajectory to the boundary of the selected area can then be calculated, and the pair of adjacent actual points with the smallest total distance can be selected as the point pair. Please refer to the relevant documentation. Figure 2b , Figure 2b This is a schematic diagram of another embodiment of displaying a trajectory on an electronic map. For example... Figure 2bAs shown in the figure, the dark solid circles represent the actual meridian points, namely meridian point A, meridian point B, and meridian point C. The solid rectangles represent the selected area. Since the selected area does not contain the actual meridian points, we can obtain the pair of actual meridian points with the minimum total distance from the boundary of the selected area among each group of adjacent actual meridian points in the first trajectory, which is the pair of actual meridian points consisting of actual meridian points B and C. Other cases can be deduced similarly, and will not be listed here.

[0024] In a specific implementation scenario, if the selected area contains only one meridian point, then two sets of adjacent meridian points containing that meridian point can be obtained from the first trajectory. The total distance between each set of adjacent meridian points is then calculated, and the set of adjacent meridian points with the smallest total distance is selected as the point pair. Please refer to [link to relevant documentation]. Figure 2c , Figure 2c This is a schematic diagram of yet another embodiment of displaying a trajectory on an electronic map. For example... Figure 2c As shown in the figure, the dark solid circles represent the actual meridian points, namely meridian point A, meridian point B, and meridian point C. The solid rectangles represent the selected areas. Since the selected area contains one meridian point (i.e., meridian point B), we can obtain two sets of adjacent meridian points containing meridian point B in the first trajectory, namely meridian point A and meridian point B, and meridian point B and meridian point C. Since the latter has a smaller total distance to the boundary of the selected area, the latter can be selected as the point pair. Other cases can be deduced similarly, and will not be listed in detail here.

[0025] In a specific implementation scenario, if the selected area contains two points, it can be checked whether these two points are adjacent in the first trajectory. If so, these two points can be directly treated as a point pair. Please refer to the relevant documentation. Figure 2d , Figure 2d This is a schematic diagram of yet another embodiment of displaying a trajectory on an electronic map. For example... Figure 2d As shown in the diagram, the dark solid circles represent the actual meridian points, namely meridian point A, meridian point B, and meridian point C. The solid rectangle represents the selected area. Since the selected area contains two meridian points (i.e., meridian point B and meridian point C), and they are adjacent to each other on the first trajectory, they can be directly considered as a pair. Conversely, if the two meridian points are not adjacent on the first trajectory, then the former of the two meridian points and its adjacent meridian point on the first trajectory can be considered as a pair, and the latter of the two meridian points and its adjacent meridian point on the first trajectory can be considered as a pair. Please refer to the following references. Figure 2e , Figure 2e This is a schematic diagram of yet another embodiment of displaying a trajectory on an electronic map. For example... Figure 2eAs shown in the diagram, the dark solid circles represent the actual meridian points, namely meridian point A, meridian point B, and meridian point C. The solid rectangle represents the selected area. Since the selected area contains two meridian points (i.e., meridian point A and meridian point C), and they are not adjacent in the first trajectory, the former (merit point A) and its adjacent meridian point B in the first trajectory can be considered as a pair, and the latter (merit point C) and its adjacent meridian point B in the first trajectory can be considered as a pair. Other cases can be deduced similarly, and will not be listed here.

[0026] In a specific implementation scenario, if the selected area contains three or more actual points, similar to the case containing two actual points, it is possible to detect whether these actual points are adjacent to each other. If so, each pair of adjacent actual points can be considered as a point pair. Conversely, if there is at least one pair of actual points that are not adjacent to each other, the former of this pair of actual points and its adjacent actual points in the first trajectory can be considered as a point pair, and the latter of this pair of actual points and its adjacent actual points in the first trajectory can be considered as a point pair.

[0027] It should be noted that after identifying the point pairs related to the selected area in the first trajectory, the former point in the pair can be selected as the first point, and the latter point as the second point. For example, using... Figure 2b For example, we can choose point B as the first point and point C as the second point; or, using Figure 2c For example, we can choose point B as the first point and point C as the second point; or, using Figure 2d For example, we can choose point B as the first point and point C as the second point; or, using Figure 2e For example, we can identify two point pairs. In one point pair, we can choose point A as the first point and point B as the second point. In the other point pair, we can choose point B as the first point and point C as the second point. Other cases can be deduced similarly, and will not be listed here.

[0028] In one implementation scenario, the first trajectory of a target object can be completed based on the historical trajectories of a preset group, where the preset group includes at least the target object. Specifically, after obtaining the historical trajectories of the preset group, a first probability distribution representing the target object passing through other points between any given points can be obtained based on the historical trajectory of the target object. A second probability distribution representing the preset group passing through other points between any given points can also be obtained based on the historical trajectories of the preset group. It should be noted that in the above statement "passing through other points between any given points," "other points" is different from either of the two points referred to in "between any given points." Based on this, a first probability value representing the target object passing through candidate points within the selected area between the first and second points can be obtained from the first probability distribution, and a second probability value representing the preset group passing through candidate points between the first and second points can be obtained from the second probability distribution. Then, based on the first and second probability values, it is determined whether to select candidate points as potential points, thus obtaining the second trajectory. It should be noted that each point within the selected area can be selected as a candidate point. The above method first obtains the probability distribution of the target object and the preset group passing through other points between points based on their historical trajectories, which at least include the target object. Based on this, the probability values ​​of the target object and the preset group passing through candidate points between the first point and the second point are obtained. Then, based on the probability values, it is determined whether to select candidate points as suspected points to add to the first trajectory. This method can infer and complete the trajectory based on historical data, which helps to improve the accuracy of trajectory completion.

[0029] In a specific implementation scenario, to obtain the probability distribution as accurately as possible based on historical trajectories, preprocessing such as anomaly screening, trajectory segmentation, and trajectory deduplication can be performed on the historical trajectories. For details of the preprocessing, please refer to the aforementioned processing procedures related to the first trajectory; they will not be repeated here.

[0030] In a specific implementation scenario, after obtaining the historical trajectory of the target object, we can count the first number of times the target object passes through any two points consecutively, and the second number of times it passes through other points between the two points. It should be noted that the "other points" in the above statement "passing through other points between the two points" are different from any one of the "two points". Taking the historical trajectory of the target object "G1, G2, G3, G4, G5, G4, G1, G2" as an example, where G1~G5 represent different points, we can calculate the first number of consecutive passages through points G1 and G2: C(G1, G2) = 2; the first number of consecutive passages through points G2 and G3: C(G2, G3) = 1; the first number of consecutive passages through points G3 and G4: C(G3, G4) = 1; the first number of consecutive passages through points G4 and G5: C(G4, G5) = 1; the first number of consecutive passages through points G5 and G4: C(G5, G4) = 1; and the first number of consecutive passages through points G4 and G1: C(G4, G1) = 1. Other historical trajectories can be calculated similarly to obtain the first number of passages. Similarly, we can calculate the second number of times the target object passes through point G2 between points G1 and G3, D((G1, G3), G2) = 1; the second number of times it passes through point G3 between points G2 and G4, D((G2, G4), G3) = 1; the second number of times it passes through point G4 between points G3 and G5, D((G3, G5), G4) = 1; the second number of times it passes through point G4 between points G5 and G1, D((G5, G1), G4) = 1; and the second number of times it passes through point G1 between points G4 and G2, D((G4, G2), G1) = 1. Other historical trajectories can be calculated similarly to obtain the second number. Based on this, we can obtain the sum of the second number of times the target object passes through each of the other points between two points, which is taken as the first total number of times. For example, for points G1 and G3, if the second number of times the target object passes through point G2 is 1, the second number of times it passes through point G4 is 1, and the second number of times it passes through point G5 is 1, then the first total number of times the target object passes through each of the other points between points G1 and G3 is 3. Other cases can be deduced similarly, and will not be listed here. Based on this, the ratio of the second number to the first total number of times can be obtained based on the comparison between the first number and the first total number of times, and used as the first probability value of the target object passing through the corresponding other points between the two points, or 0 can be selected as the first probability value of the target object passing through any other point between the two points.For example, if the comparison result indicates that the first total number of occurrences is significantly greater than the first number, the ratio of the second number of occurrences at other points to the first total number of occurrences can be obtained as the first probability value of the target object passing through other corresponding points between the two points; or, if the comparison result indicates that the first total number of occurrences does not satisfy the condition of being significantly greater than the first number, 0 can be selected as the first probability value of the target object passing through any other point between the two points. It should be noted that the specific definition of "significantly greater than" can be set according to the actual application situation. For example, when the first total number of occurrences is greater than a preset multiple of the first number (e.g., 5 times, 10 times), it can be considered that the first total number of occurrences is significantly greater than the first number; conversely, it can be considered that the first total number of occurrences does not satisfy the condition of being significantly greater than the first number. For ease of understanding, taking four points G1 to G4 as an example, the first number of times the target object passes through any point consecutively can be represented by a matrix as follows:

[0031] Similarly, the second number of times a target object traverses other points between two points can be represented by a matrix as follows:

[0032] Taking the target object passing through points G2 and G3 in succession as an example, the first number of times the target object passes through points G2 and G3 in succession is P23. The second number of times the target object passes through point G2 first and then point G3, and passes through point G1 again during the period between passing through points G2 and G3 is P213. The second number of times the target object passes through point G4 again during the period between passing through points G2 and G3 is P243. Based on this, the total number of times the target object passes through other points during the period between passing through points G2 and G3 can be calculated as P213 + P243. Therefore, when P213 + P243 >> P23, the first probability value of the target object passing through point G1 during the period between passing through points G2 and G3 is ProS213 = P213 / (P213 + P243), and the first probability value of the target object passing through point G4 during the period between passing through points G2 and G3 is ProS243 = P213 / (P213 + P243). Conversely, if the above condition of "much greater than" is not met, the first probability value of the target object passing through any other point during the period between passing through points G2 and G3 is 0. Other cases can be deduced by analogy, and will not be listed here. The above method, based on the historical trajectory of the target object, counts the first number of times the target object passes through any two points consecutively and the second number of times it passes through other points between the two points. The sum of the second number of times the target object passes through each other point between the two points is taken as the first total number. Based on the comparison between the first number and the first total number, the ratio of the second number for each other point to the first total number is obtained as the first probability value of the target object passing through the corresponding other point between the two points. Alternatively, 0 can be chosen as the first probability value of the target object passing through any other point between the two points. Therefore, this method helps improve the accuracy of the first probability distribution. Of course, the above example only uses passing through one other point as an example. In practical applications, it is not limited to this; it can also pass through two, three, or more other points. Furthermore, when the number of other points passed through is two or more, the process of obtaining the first and second probability distributions can be deduced similarly, which will not be illustrated here.

[0033] In a specific implementation scenario, similar to the first probability distribution of the target object mentioned above, after obtaining the historical trajectory of the preset group, the third number of times the preset group passes through any two points consecutively and the fourth number of times it passes through other points between the two points can be counted based on the historical trajectory of the preset group. The sum of the fourth numbers of the preset group passing through each other point between the two points is obtained as the second total number. Based on the comparison between the third number and the second total number, the ratio of the fourth number corresponding to other points to the second total number can be obtained as the second probability value of the preset group passing through the corresponding other points between the two points. Alternatively, 0 can be selected as the second probability value of the preset group passing through any other point between the two points. The specific process is similar to the specific process of the first probability distribution mentioned above, and will not be repeated here.

[0034] In a specific implementation scenario, after obtaining the first and second probability distributions, a completion detection result can be obtained based on these distributions. This result includes whether completion is needed between the first and second points and the degree of suspicion that completion is required. Based on this, in response to the completion detection result indicating that completion is needed between the first and second points, a probability threshold negatively correlated with the degree of suspicion can be obtained. That is, the higher the degree of suspicion, the smaller the probability threshold, and vice versa. For each candidate point, a first probability value for the target object passing through the candidate point between the first and second points, and a second probability value for the preset group passing through the candidate point between the first and second points, can be obtained. If either the first or second probability value is not less than the probability threshold, the candidate point can be selected as a suspected point to obtain the second trajectory; otherwise, if either the first or second probability value is less than the probability threshold, the candidate point can be discarded as a suspected point. The same judgment process can then be performed on the next candidate point until all candidate points have been judged. For example, taking the case where a target object passes through a first point GX and a second point GY, if there exists a candidate point GZ such that the first probability value ProSXZY of the target object passing through the candidate point GZ between the first point GX and the second point GY is not less than a probability threshold, or the second probability value ProGXZY of the preset group passing through the candidate point GZ between the first point GX and the second point GY is not less than a probability threshold, then the candidate point GZ can be selected as a suspected point; otherwise, the candidate point GZ can be discarded. Furthermore, as mentioned earlier, based on historical trajectories, we can statistically obtain the first number of times the target object has consecutively passed through the first and second points, the first total number of times the target object has historically passed through any other point between the first and second points, the third number of times the preset group has consecutively passed through the first and second points, and the second total number of times the preset group has historically passed through any other point between the first and second points.Based on this, if the comparison result of the first probability distribution representing the first number and the first total number of times satisfies the first condition and the comparison result of the second probability distribution representing the third number and the second total number of times satisfies the second condition, it can be determined that the completion detection result, including the first point and the second point, does not need to be completed; or, if the comparison result of the first probability distribution representing the first number and the first total number of times does not satisfy the first condition or the comparison result of the second probability distribution representing the third number and the second total number of times does not satisfy the second condition, it can be determined that the completion detection result, including the first point and the second point, needs to be completed and the degree of suspicion of needing to be completed is the first degree; or, if the comparison result of the first probability distribution representing the first number and the first total number of times does not satisfy the first condition and the comparison result of the second probability distribution representing the third number and the second total number of times does not satisfy the second condition, it can be determined that the completion detection result, including the first point and the second point, needs to be completed and the degree of suspicion of needing to be completed is the second degree. It should be noted that the first condition can include a comparison between the first count and the first total count showing that the first total count is much greater than the first count. The second condition can include a comparison between the third count and the second total count showing that the second total count is much greater than the third count, and the first degree is lower than the second degree. Furthermore, as mentioned earlier, if the first total number of times the target object passes through all other points between any two points is much greater than the first number of times the target object passes through those two points consecutively, then the first probability value can be determined based on the ratio of the second count to the first total count; otherwise, it is directly set to 0. The second probability distribution follows the same logic. Therefore, if the first probability distribution shows that the sum of the first probability values ​​of the target object passing through any other point between the first and second points is not 0, then the target object should satisfy the condition that the total number of times it has passed through other points between the first and second points is much greater than the number of times it has passed through the first and second points consecutively. Conversely, if the first probability distribution shows that the sum of the first probability values ​​of the target object passing through any other point between the first and second points is 0, then the target object should satisfy the condition that the total number of times it has passed through other points between the first and second points is not much greater than the number of times it has passed through the first and second points consecutively. Similarly, if the second probability distribution shows that the sum of the first probability values ​​of the preset group passing through any other point between the first and second points is not 0, then the preset group should satisfy that the second total number of times it has passed through other points between the first and second points is much greater than the third number of times it has passed through the first and second points consecutively. Conversely, if the second probability distribution shows that the sum of the first probability values ​​of the preset group passing through any other point between the first and second points is 0, then the preset group should satisfy that the second total number of times it has passed through other points between the first and second points is not much greater than the third number of times it has passed through the first and second points consecutively.The above method obtains completion detection results based on a first probability distribution and a second probability distribution. The completion detection results include whether completion is needed between the first point and the second point and the degree of suspicion that completion is needed. In response to the completion detection results indicating that completion is needed between the first point and the second point, a probability threshold negatively correlated with the degree of suspicion is obtained. If either the first probability value or the second probability value is not less than the probability threshold, a candidate point is determined as a suspected point to obtain the second trajectory. Conversely, if neither the first probability value nor the second probability value is less than the probability threshold, the candidate point is discarded as a suspected point. Therefore, it is possible to detect candidate points in a targeted manner according to whether completion is needed and the degree of suspicion that completion is needed, which helps to improve the accuracy of completing the first trajectory based on historical trajectories.

[0035] In another implementation scenario, unlike the aforementioned completion of the first trajectory based on historical trajectories, the first trajectory can also be completed based on an electronic map. Specifically, candidate paths from the first point to the second point within a selected area can be obtained based on the electronic map. Based on a comparison between the lateral travel time of each candidate path and the actual travel time from the first point to the second point, candidate paths are selected as suspected paths. Furthermore, points within the selected area of ​​the suspected paths can be obtained and added as suspected points between the first and second points of the first trajectory to obtain the second trajectory. This method, by obtaining candidate paths from the first point to the second point within a selected area and determining the reliability of the candidate paths based on their predicted travel time and the actual travel time, helps improve the reliability of completing the first trajectory based on an electronic map.

[0036] In a specific implementation scenario, the path length of candidate paths on an electronic map can be obtained, as well as the movement speed of the target object (e.g., maximum movement speed, average movement speed, etc.). The ratio of path length to movement speed can then be used as the predicted travel time of the candidate path.

[0037] In a specific implementation scenario, please refer to the relevant documents. Figure 2bCandidate paths BDC and BEC can be obtained. For candidate path BDC, the path length on the electronic map and the target object's movement speed (e.g., maximum speed, average speed) can be obtained. The ratio of path length to movement speed can then be used as the predicted travel time of candidate path BDC. Simultaneously, the travel time of the target object at the second point can be subtracted from the travel time at the first point to obtain the measured travel time from point B to point C. Based on this, if the predicted travel time is not greater than the measured travel time, candidate path BDC is considered relatively reliable, and point D can be added as a potential point between point B and point C to obtain the second trajectory ABDC. Conversely, if the predicted travel time is greater than the measured travel time, candidate path BDC is considered unreliable and can be discarded. The same logic applies to candidate path BEC, and will not be elaborated further here.

[0038] In another implementation scenario, unlike the two methods mentioned above for completing the first trajectory, the first trajectory can also be completed by combining historical trajectories and electronic maps. Specifically, suspected points between the first and second points can be obtained based on historical trajectories, and suspected points between the first and second points can also be obtained based on electronic maps. In other words, suspected points between the first and second points can be obtained by combining the two methods mentioned above.

[0039] Step S14: Display the second trajectory on the electronic map.

[0040] Specifically, after obtaining the second trajectory, it can be displayed on the electronic map. For example, the first trajectory previously displayed on the electronic map can be updated to the second trajectory; or, both the first and second trajectories can be displayed simultaneously on the electronic map. Figure 2a As shown, the electronic map can simultaneously display the first trajectory ABCD and the second trajectory ABECFD. Of course, to facilitate differentiation between the first and second trajectories, the first trajectory before completion and the second trajectory after completion can be displayed on the electronic map in different styles. For example, the first trajectory before completion can be displayed as a dashed line on the electronic map, while the second trajectory after completion can be displayed as a solid line. Furthermore, actual and suspected locations can also be displayed on the electronic map in different styles, such as using different colors or grayscale levels for differentiation; this is not limited here.

[0041] The above scheme obtains the first trajectory of the target object, which includes the actual points the target object passes through sequentially. This first trajectory is then displayed on an electronic map. In response to a selection command on the electronic map, it identifies point pairs related to the selected area within the first trajectory, selects the actual points passed sequentially within each pair as the first and second points, and completes the selection of suspected points between the first and second points within the selected area to obtain the second trajectory. This second trajectory is then displayed on the electronic map. Since the trajectory completion process is triggered by the user's selection command on the electronic map, and suspected points are completed within the selected area and between point pairs related to the selected area, it can meet the user's personalized needs for trajectory completion as much as possible. This improves the completeness of the target object's movement trajectory within the user's area of ​​interest, and helps improve the application effect of the movement trajectory in subsequent downstream tasks.

[0042] Please see Figure 3 , Figure 3 This is a flowchart illustrating another embodiment of the trajectory completion method of this application. Specifically, it may include the following steps: Step S31: Obtain the first trajectory of the target object.

[0043] In this embodiment of the disclosure, the first trajectory includes the actual points traversed by the target object in sequence. For details, please refer to the relevant descriptions in the foregoing embodiments of the disclosure, which will not be repeated here.

[0044] Step S32: Display the first trajectory on the electronic map.

[0045] For details, please refer to the relevant descriptions in the foregoing disclosed embodiments, which will not be repeated here.

[0046] Step S33: In response to the box selection command on the electronic map, determine the point pairs related to the boxed area in the first trajectory, select the actual points passed in the point pairs as the first point and the second point respectively, and fill in the suspected points located between the first point and the second point in the boxed area to obtain the second trajectory.

[0047] For details, please refer to the relevant descriptions in the foregoing disclosed embodiments, which will not be repeated here.

[0048] Step S34: Display the second trajectory on the electronic map.

[0049] For details, please refer to the relevant descriptions in the foregoing disclosed embodiments, which will not be repeated here.

[0050] Step S35: In response to the selection instruction for suspected points in the second trajectory, based on the first time when the target object passes through the first point and the second time when it passes through the second point, determine the time range of the target object passing through the suspected points and acquire the video data captured by the suspected points within the time range.

[0051] Specifically, the first time it takes for the target object to travel from the first point to the suspected point can be predicted, and the second time it takes for the target object to travel from the suspected point to the second point can be predicted. Based on this, the sum of the first time and the first duration can be obtained as the lower limit of the time range, and the difference between the second time and the second duration can be obtained as the lower limit of the time range. It should be noted that the path length from the first point to the suspected point can be obtained from an electronic map, and the movement speed of the target object from the first point to the suspected point (e.g., average movement speed, fastest movement speed, etc.) can be obtained. The ratio of the path length to the movement speed is used as the first time it takes for the target object to travel from the first point to the suspected point. Similarly, the path length from the suspected point to the second point can be obtained from an electronic map, and the movement speed of the target object from the suspected point to the second point (e.g., average movement speed, fastest movement speed, etc.) can be obtained. The ratio of the path length to the movement speed is used as the second time it takes for the target object to travel from the suspected point to the second point. For ease of description, the first time the target object passes through the first point GX can be denoted as... The second time when passing through the second point GY can be recorded as The first time it takes for the target to travel from the first point GX to the suspected point GZ can be any time. The second time it takes for the target to travel from the suspected location GZ to the second location GY can be denoted as... Therefore, the lower limit of the time range can be denoted as... The upper limit of the time range can be denoted as: The above method predicts the first time it takes for the target object to travel from the first location to the suspected location, and predicts the second time it takes for the target object to travel from the suspected location to the second location. Based on this, the sum of the first time and the first duration is obtained as the lower limit of the time range, and the difference between the second time and the second duration is obtained as the upper limit of the time range, which helps to improve the accuracy of the time range.

[0052] In addition, after determining the time range, video data taken at the suspected locations within the determined time range can be obtained for subsequent analysis.

[0053] Step S36: Based on the captured image of the target object at any real point in the first trajectory, extract at least one frame from the video data as a suspected image of the target object at the suspected point.

[0054] In one implementation scenario, the similarity between captured images and each frame of video data can be obtained. Based on at least one of the following: a first probability value of the target object passing through a suspected location between a first and second location; and a second probability value of a pre-defined group passing through a suspected location between the first and second locations, the reliability of the target object passing through the suspected location is obtained, where the pre-defined group includes at least the target object. Furthermore, the reliability can be fused with the similarity corresponding to each frame to obtain the confidence level that the corresponding frame belongs to the target object. Therefore, based on the confidence level of each frame, at least one frame can be selected as a suspected image. This method, by using similarity measured from the image dimension and reliability measured from the probability dimension to filter suspected images in video data, helps improve the accuracy of suspected images.

[0055] In a specific implementation scenario, image features of the captured image and image features of each frame in the video data can be extracted separately, and similarity measures such as cosine similarity can be used to measure the similarity between the two image features to obtain the similarity between the captured image and each frame in the video data.

[0056] In a specific implementation scenario, the specific process of obtaining the first probability value and the second probability value can be referred to the aforementioned disclosed embodiments, and will not be repeated here. Furthermore, when the reliability is obtained based solely on the first probability value, the first probability value can be directly used as the reliability. Alternatively, when the reliability is obtained based solely on the second probability value, the second probability value can be directly used as the reliability. Or, when the reliability is obtained based on both the first and second probability values, the larger of the two can be used as the reliability. Of course, this is not a limitation; for example, the average of the two can be used as the reliability, or the weighted average of the two can be used as the reliability. No further limitation is made here.

[0057] In a specific implementation scenario, after obtaining the similarity and reliability, fusion operations such as addition, averaging, and weighting can be performed on the two to obtain the confidence level of each frame image belonging to the target object. Based on this, the frames can be sorted in descending order of confidence level, and the frames before a preset order (e.g., the first 5 images) or before a preset proportion (e.g., the first 10%) can be selected as suspected images.

[0058] In one implementation scenario, after obtaining a suspected image, each frame of suspected images belonging to the target object at the suspected point can be displayed. In response to a confirmation command for the suspected image, the selected suspected image is fixed as the image of the target object captured by the suspected point in the second trajectory. Then, based on the analysis of the image of the target object captured by the suspected point, the direction of travel of the target object at the suspected point is obtained. Based on the direction of travel, the connecting line of the suspected point in the second trajectory is optimized. Therefore, the image of the target object captured at the suspected point can be determined through user interaction, and the connecting line of the suspected point in the second trajectory can be optimized accordingly, which helps to further improve the accuracy of the second trajectory.

[0059] In a specific implementation scenario, after the selected suspected image is fixed as an image of the target object taken from a suspected point in the second trajectory, an image tag representing fusion heterogeneity can be added to the image. This allows the evidence chain to be presented, indicating that the image was determined through fusion heterogeneity analysis. Please refer to [link to relevant documentation]. Figure 2f , Figure 2f This is a schematic diagram of yet another embodiment of electronic map trajectory display. For example... Figure 2f As shown, the text "Fusion Heterogeneity" can be directly displayed on the captured image as an image label representing "fusion heterogeneity." In addition to the image label representing "fusion heterogeneity," the captured image can also display the name of the suspected location and the time the image was captured. Alternatively, after selecting a suspected location, a video clip of the target object passing through that location can be played.

[0060] In a specific implementation scenario, after fixing the selected suspected image as the image of the target object taken from the suspected point in the second trajectory, the target object in the image can be analyzed to obtain the target object's direction of travel. This allows for the optimization of the connecting line of the second trajectory at the suspected point based on the direction of travel. Please refer to further details. Figure 2f For the suspected point F, analysis of the image of the target object at suspected point F reveals that the target object continues to move forward from suspected point F, rather than turning right directly. Therefore, the connection line between suspected point F and actual point D can be further optimized: starting from suspected point F, moving forward and passing through the intersection where suspected point F and actual point D are located, then connecting with actual point D. This makes the second trajectory more accurate.

[0061] The above scheme, after displaying the second trajectory on the electronic map, responds to the selection instruction for suspected points in the second trajectory, determines the time range of the target object passing through the suspected points based on the first time of the target object passing through the first point and the second time of passing through the second point, and obtains the video data captured by the suspected points within the time range. Thus, based on the image of the target object captured by any actual passing point in the first trajectory, at least one frame is extracted from the video data as the suspected image of the target object at the suspected point. In turn, it can respond to user interaction and obtain the captured image of the target object through heterogeneous fusion analysis, which helps to improve and perfect the passage evidence chain and restore the real trajectory as much as possible.

[0062] Please see Figure 4 , Figure 4 This is a schematic diagram of an embodiment of the trajectory completion device 40 of this application. The trajectory completion device 40 includes: a trajectory acquisition module 41, a first display module 42, a trajectory completion module 43, and a second display module 44. The trajectory acquisition module 41 is used to acquire a first trajectory of a target object; wherein, the first trajectory includes the actual points passed by the target object in sequence. The first display module 42 is used to display the first trajectory on an electronic map. The trajectory completion module 43 is used to respond to a box selection command on the electronic map, determine the point pairs related to the boxed area in the first trajectory, select the actual points passed by in sequence in the point pairs as the first point and the second point, respectively, and complete the suspected points located between the first point and the second point in the boxed area to obtain the second trajectory. The second display module 44 is used to display the second trajectory on the electronic map.

[0063] The above scheme obtains the first trajectory of the target object, which includes the actual points the target object passes through sequentially. This first trajectory is then displayed on an electronic map. In response to a selection command on the electronic map, it identifies point pairs related to the selected area within the first trajectory, selects the actual points passed sequentially within each pair as the first and second points, and completes the selection of suspected points between the first and second points within the selected area to obtain the second trajectory. This second trajectory is then displayed on the electronic map. Since the trajectory completion process is triggered by the user's selection command on the electronic map, and suspected points are completed within the selected area and between point pairs related to the selected area, it can meet the user's personalized needs for trajectory completion as much as possible. This improves the completeness of the target object's movement trajectory within the user's area of ​​interest, and helps improve the application effect of the movement trajectory in subsequent downstream tasks.

[0064] In some disclosed embodiments, the trajectory completion module 43 includes a historical trajectory acquisition submodule for acquiring the historical trajectory of a preset group, wherein the preset group at least includes the target object; the trajectory completion module 43 includes a first distribution acquisition submodule for obtaining a first probability distribution representing the target object passing through other points between any points based on the historical trajectory of the target object; the trajectory completion module 43 includes a second distribution acquisition submodule for obtaining a second probability distribution representing the preset group passing through other points between any points based on the historical trajectory of the preset group; the trajectory completion module 43 includes a passage probability acquisition submodule for acquiring a first probability value in the first probability distribution representing the target object passing through candidate points located within the selected area between the first and second points, and acquiring a second probability value in the second probability distribution representing the preset group passing through candidate points between the first and second points; the trajectory completion module 43 includes a suspected point determination submodule for determining whether to select candidate points as suspected points based on the first and second probability values ​​to obtain the second trajectory.

[0065] In some disclosed embodiments, the first distribution acquisition submodule includes a first statistics unit, used to count the first number of times the target object passes through any two points consecutively and the second number of times it passes through other points between the two points, based on the historical trajectory of the target object; the first distribution acquisition submodule includes a first summation unit, used to obtain the sum of the second number of times the target object passes through each other point between the two points, as the first total number; the first distribution acquisition submodule includes a first calculation unit, used to obtain the ratio of the second number corresponding to other points to the first total number based on the comparison result of the first number and the first total number, as the first probability value of the target object passing through the corresponding other points between the two points, or to select 0 as the first probability value of the target object passing through any other point between the two points.

[0066] In some disclosed embodiments, the second distribution acquisition submodule includes a second statistical unit, used to count the third number of times the preset group passes through any two points consecutively and the fourth number of times it passes through other points between the two points, based on the historical trajectory of the preset group; the second distribution acquisition submodule includes a second summation unit, used to obtain the sum of the fourth numbers of the preset group passing through each other point between the two points, as the second total number; the second distribution acquisition submodule includes a second calculation unit, used to obtain the ratio of the fourth number corresponding to other points to the second total number based on the comparison result of the third number and the second total number, as the second probability value of the preset group passing through the corresponding other points between the two points, or to select 0 as the second probability value of the preset group passing through any other point between the two points.

[0067] In some disclosed embodiments, the trajectory completion module 43 includes a completion requirement detection submodule, used to obtain a completion detection result based on a first probability distribution and a second probability distribution; wherein, the completion detection result includes whether completion is needed between the first point and the second point and the degree of suspicion that completion is needed; the suspected point determination submodule includes a probability threshold acquisition unit, used to obtain a probability threshold negatively correlated with the degree of suspicion in response to the completion detection result indicating that completion is needed between the first point and the second point; the suspected point determination submodule includes a first determination unit, used to determine that if either the first probability value or the second probability value is not less than the probability threshold, a candidate point is selected as a suspected point to obtain the second trajectory; the suspected point determination submodule includes a second determination unit, used to determine that if both the first probability value and the second probability value are less than the probability threshold, a candidate point is discarded as a suspected point.

[0068] In some disclosed embodiments, the completion requirement detection submodule includes a first response unit, configured to determine that the completion detection result, including the first point and the second point, does not require completion in response to a first probability distribution representing the comparison result of the first number and the first total number of times satisfying a first condition and a second probability distribution representing the comparison result of the third number and the second total number of times satisfying a second condition; the completion requirement detection submodule includes a second response unit, configured to determine that the completion detection result, including the first point and the second point, requires completion in response to a first probability distribution representing the comparison result of the first number and the first total number of times not satisfying the first condition or a second probability distribution representing the comparison result of the third number and the second total number of times not satisfying the second condition, and that the degree of suspicion of needing completion is a first degree; the completion requirement detection submodule includes a third response unit. The unit is configured to, in response to a first probability distribution representing the comparison result of the first number and the first total number of times not satisfying the first condition and a second probability distribution representing the comparison result of the third number and the second total number of times not satisfying the second condition, determine that the completion detection result includes the need for completion between the first point and the second point, and the degree of suspicion of needing completion is the second degree; wherein, the first number represents the number of times the target object has continuously passed through the first point and the second point in history, the first total number of times represents the total number of times the target object has historically passed through any other point between the first point and the second point, the third number represents the number of times the preset group has historically continuously passed through the first point and the second point, and the second total number of times represents the total number of times the preset group has historically passed through any other point between the first point and the second point, and the first degree is lower than the second degree.

[0069] In some disclosed embodiments, the trajectory completion module 43 includes a candidate path acquisition submodule, used to acquire candidate paths from the first point to the second point within the selected area based on an electronic map; the trajectory completion module 43 includes a suspected path selection submodule, used to select candidate paths as suspected paths based on the comparison results between the predicted travel time of each candidate path and the measured travel time from the first point to the second point; the trajectory completion module 43 includes a suspected point insertion submodule, used to acquire each point located within the selected area in the suspected path and insert it as a suspected point between the first point and the second point to obtain a second trajectory.

[0070] In some disclosed embodiments, the trajectory completion device 40 includes a range determination module, used to determine the time range of the target object passing through the suspected point in the second trajectory based on the first time of the target object passing through the first point and the second time of passing through the second point in response to the selection instruction of the suspected point in the second trajectory; the trajectory completion device 40 includes a video acquisition module, used to acquire video data captured at the suspected point within the time range; the trajectory completion device 40 includes an image selection module, used to extract at least one frame of image from the video data as the suspected image of the target object at the suspected point based on the captured image of the target object at any actual passing point in the first trajectory.

[0071] In some disclosed embodiments, the image selection module includes a similarity measurement submodule for obtaining the similarity between the captured image and each frame of the video data; the image selection module includes a reliability measurement submodule for obtaining the reliability of the target object passing through the suspected point based on at least one of a first probability value of the target object passing through the suspected point between the first point and the second point, and a second probability value of a preset group passing through the suspected point between the first point and the second point; wherein the preset group includes at least the target object; the image selection module includes a confidence determination submodule for fusing the reliability with the similarity corresponding to each frame of the image to obtain the confidence that the corresponding frame of the image belongs to the target object; the image selection module includes an image selection submodule for selecting at least one frame of the image as a suspected image based on the confidence of each frame of the image.

[0072] In some disclosed embodiments, the trajectory completion device 40 includes an image confirmation module, used to fix the selected suspected image as a captured image of the target object from a suspected point in the second trajectory in response to a confirmation command for a suspected image; the trajectory completion device 40 includes a direction analysis module, used to analyze the captured image of the target object from the suspected point to obtain the direction of travel of the target object at the suspected point; the trajectory completion device 40 includes a trajectory optimization module, used to optimize the connecting lines of the suspected points in the second trajectory based on the direction of travel.

[0073] In some disclosed embodiments, the range determination module includes a duration prediction submodule, used to predict the first duration of the target object traveling from the first point to the suspected point, and to predict the second duration of the target object traveling from the suspected point to the second point; the range determination module includes a boundary determination submodule, used to obtain the sum of the first time and the first duration as the lower limit of the time range, and to obtain the difference between the second time and the second duration as the upper limit of the time range.

[0074] Please see Figure 5 , Figure 5 This is a schematic diagram of an embodiment of the electronic device 50 of this application. The electronic device 50 includes a memory 51, a processor 52, and a display 53. The memory 51 and the display 53 are respectively coupled to the processor 52. The memory 51 stores program instructions, and the processor 52 is used to execute the program instructions to implement the steps in any of the above-described trajectory completion method embodiments. For details, please refer to the foregoing disclosed embodiments, which will not be repeated here. It should be noted that the electronic device 50 may include, but is not limited to, servers, desktop computers, laptops, etc., and is not limited here.

[0075] Specifically, processor 52 can also be referred to as a CPU (Central Processing Unit). Processor 52 may be an integrated circuit chip with signal processing capabilities. Processor 52 can also be a general-purpose processor, digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component. A general-purpose processor can be a microprocessor or any conventional processor. In addition, processor 52 can be implemented by multiple integrated circuit chips.

[0076] In the above scheme, the processor 52 in the electronic device 50 implements the steps in any of the above trajectory completion method embodiments. Since the trajectory completion process is triggered by the user's box selection command on the electronic map, and the suspected points are completed between the point pairs related to the box selection area, it can meet the user's personalized needs for trajectory completion as much as possible, so as to improve the completeness of the target object's movement trajectory in the user's attention area, and help improve the user's application effect based on the movement trajectory in subsequent downstream tasks.

[0077] Please see Figure 6 , Figure 6 This is a schematic diagram of a framework of an embodiment of the computer-readable storage medium 60 of this application. The computer-readable storage medium 60 stores processor-executable program instructions 61, which can be executed to implement the steps in any of the above-described trajectory completion method embodiments.

[0078] It should be noted that the computer-readable storage medium 60 can be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or a medium that can store program instructions 61. Alternatively, it can be a server that stores the program instructions 61. The server can send the stored program instructions 61 to other devices for execution, or it can execute the stored program instructions 61 itself.

[0079] The above scheme, implemented by computer-readable storage medium 60, achieves the steps in any of the above trajectory completion method embodiments. Since the trajectory completion process is triggered by the user's box selection command on the electronic map, and suspected points are completed within the box selection area and between point pairs related to the box selection area, it can meet the user's personalized needs for trajectory completion as much as possible, thereby improving the completeness of the target object's movement trajectory in the user's attention area and helping to improve the user's application effect based on movement trajectory in subsequent downstream tasks.

[0080] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus implementations described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0081] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0082] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0083] If the integrated unit is implemented as 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 this application, in essence, 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. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0084] If the technical solution of this application involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, with clear signs / information informing users of the personal information processing rules, authorization is obtained from the individual through pop-up information or by asking the individual to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

Claims

1. A trajectory completion method, characterized in that, include: Obtain the first trajectory of the target object; wherein, the first trajectory includes the actual points that the target object passes through in sequence; Display the first trajectory on the electronic map; In response to the box selection command on the electronic map, the point pairs related to the boxed area in the first trajectory are determined, and the actual points passed in the point pairs are selected as the first point and the second point, respectively. Suspected points located between the first point and the second point are filled in the boxed area to obtain the second trajectory. The second trajectory is displayed on the electronic map; In response to the selection instruction for the suspected point in the second trajectory, based on the first time when the target object passes through the first point and the second time when it passes through the second point, the time range of the target object passing through the suspected point is determined, and video data captured by the suspected point within the time range is acquired; The similarity between the captured image of the target object at any of the real-time points in the first trajectory and the images of each frame in the video data is obtained. Based on at least one of the first probability value of the target object passing through the suspected point between the first point and the second point, and the second probability value of a preset group passing through the suspected point between the first point and the second point, the reliability of the target object passing through the suspected point is obtained; wherein, the preset group includes at least the target object. Based on the reliability and the similarity of each frame image, the confidence that the corresponding frame image belongs to the target object is obtained; Based on the confidence level of each frame of image, at least one frame of image is selected as the suspected image of the target object at the suspected location.

2. The method according to claim 1, characterized in that, The step of completing the suspected points located between the first point and the second point within the selected area to obtain the second trajectory includes: Obtain the historical trajectory of a preset group, wherein the preset group contains at least the target object; Based on the historical trajectory of the target object, a first probability distribution representing the target object passing through other points between any points is obtained, and based on the historical trajectory of the preset group, a second probability distribution representing the preset group passing through other points between any points is obtained. Obtain a first probability value from the first probability distribution representing the target object passing through a candidate point located within the selected area between the first point and the second point, and obtain a second probability value from the second probability distribution representing the preset group passing through the candidate point between the first point and the second point. Based on the first probability value and the second probability value, determine whether to select the candidate point as the suspected point to obtain the second trajectory.

3. The method according to claim 2, characterized in that, The process of obtaining a first probability distribution representing the target object's passage through other points between any given points, based on the target object's historical trajectory, includes: Based on the historical trajectory of the target object, the first number of times the target object passes through any two points consecutively and the second number of times it passes through other points between the two points are counted. The sum of the second number of times the target object passes through each other point between the two points is obtained as the first total number of times; Based on the comparison result of the first number and the first total number, the ratio of the second number corresponding to the other point to the first total number is obtained, which is used as the first probability value of the target object passing through the other point between the two points, or 0 is selected as the first probability value of the target object passing through any of the other points between the two points.

4. The method according to claim 2, characterized in that, The process of obtaining a second probability distribution representing the passage of the preset group through other points between any given points, based on the historical trajectory of the preset group, includes: Based on the historical trajectory of the preset group, count the third time the preset group passes through any two points consecutively and the fourth time it passes through other points between the two points. The sum of the fourth number of times the preset group passes through each other point between the two points is obtained as the second total number; Based on the comparison result of the third count and the second total count, the ratio of the fourth count corresponding to the other points to the second total count is obtained, which is used as the second probability value of the preset group passing through the other points between the two points, or 0 is selected as the second probability value of the preset group passing through any of the other points between the two points.

5. The method according to any one of claims 2 to 4, characterized in that, Before determining whether to select the candidate point as the suspected point based on the first probability value and the second probability value to obtain the second trajectory, the method further includes: Based on the first probability distribution and the second probability distribution, a completion detection result is obtained; wherein, the completion detection result includes whether the first point and the second point need to be completed and the degree of suspicion that completion is needed; The step of determining whether to select the candidate point as the suspected point based on the first probability value and the second probability value to obtain the second trajectory includes: In response to the completion detection result indicating that completion is needed between the first point and the second point, a probability threshold negatively correlated with the degree of suspicion is obtained; If either the first probability value or the second probability value is not less than the probability threshold, then the candidate point is selected as the suspected point to obtain the second trajectory. If both the first probability value and the second probability value are less than the probability threshold, then the candidate point is discarded as the suspected point.

6. The method according to claim 1, characterized in that, The step of completing the suspected points located between the first point and the second point within the selected area to obtain the second trajectory includes: Based on the electronic map, candidate paths from the first point to the second point are obtained within the selected area; Based on the comparison results between the predicted travel time of each candidate path and the measured travel time from the first point to the second point, the candidate path is selected as the suspected path. Each point located within the selected area in the suspected path is obtained and inserted between the first point and the second point as a suspected point to obtain the second trajectory.

7. The method according to claim 1, characterized in that, After selecting at least one frame as the suspected image of the target object at the suspected location based on the confidence level of each frame, the method further includes: In response to the confirmation command for the suspected image, the selected suspected image is fixed as the image of the target object captured by the suspected point in the second trajectory; Based on the suspected location, the captured image of the target object is analyzed to obtain the direction of travel of the target object at the suspected location; Based on the direction of travel, optimize the connecting lines of the suspected points in the second trajectory.

8. A trajectory completion device, characterized in that, include: The trajectory acquisition module is used to acquire the first trajectory of the target object; wherein, the first trajectory includes the actual points that the target object passes through in sequence; The first display module is used to display the first trajectory on an electronic map; The trajectory completion module is used to respond to the box selection command on the electronic map, determine the point pairs related to the boxed area in the first trajectory, select the actual points passed in the point pairs as the first point and the second point respectively, and complete the suspected points located between the first point and the second point in the boxed area to obtain the second trajectory. The second display module is used to display the second trajectory on the electronic map; The range determination module is used to respond to the selection instruction of the suspected point in the second trajectory and determine the time range of the target object passing through the suspected point based on the first time of the target object passing through the first point and the second time of the target object passing through the second point; The video acquisition module is used to acquire video data of the suspected location within the time range. An image selection module is used to obtain the similarity between an image of the target object taken from any of the real-time points in the first trajectory and each frame of the video data, and to obtain the reliability of the target object passing through the suspected point based on at least one of a first probability value of the target object passing through the suspected point between the first point and the second point, and a second probability value of a preset group passing through the suspected point between the first point and the second point; wherein the preset group includes at least the target object; the reliability is fused with the similarity corresponding to each frame of the image to obtain the confidence that the corresponding frame of the image belongs to the target object; based on the confidence of each frame of the image, at least one frame of the image is selected as the suspected image of the target object at the suspected point.

9. An electronic device, characterized in that, The device includes a display, a memory, and a processor. The display and the memory are respectively coupled to the processor. The memory stores program instructions, and the processor is used to execute the program instructions to implement the trajectory completion method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The system stores program instructions that can be executed by a processor, the program instructions being used to implement the trajectory completion method according to any one of claims 1 to 7.

Images