Trajectory determination method and apparatus, electronic device, storage medium, and program product

Abstract

The present disclosure provides a trajectory determination method, device, electronic equipment, storage medium and program product. The method comprises: performing image analysis on first video data collected in a first region to obtain first movement data of a target object; the first movement data comprises a movement vector and an object feature; obtaining second movement data of the target object, the second movement data being obtained based on second video data, the second video data being collected in a second region, the second region being adjacent or overlapping with the first region, and a collection interval of the second movement data and the first movement data being less than a first threshold; and determining a movement trajectory of the target object based on a similarity between the first movement data and the second movement data. The present disclosure determines the action trajectory of the target object by means of the movement data, and the movement data can not depend on the face feature information, but consider the movement vector and the object feature of the target object and other aspects, thereby realizing more accurate trajectory determination of the target object.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to a trajectory determination method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] Currently, in most application scenarios, existing methods for determining a trajectory rely on acquiring facial information and then using facial features to determine the movement trajectory of the person being tracked. However, if the face appears at a distance from the surveillance camera, the area occupied by the face is very small, leading to inaccurate acquisition of facial feature information and thus making it impossible to accurately obtain all the trajectory information of the person being tracked. Summary of the Invention

[0003] This disclosure provides a trajectory determination method, apparatus, electronic device, storage medium, and program product to address, to some extent, the problem of being unable to accurately obtain trajectory information provided by tracking personnel.

[0004] According to one aspect of this disclosure, a trajectory determination method is provided, the method comprising: performing image analysis on first video data collected in a first region to obtain first motion data of a target object; the first motion data including: a motion vector and object features; acquiring second motion data of the target object, the second motion data being obtained based on second video data, the second video data being collected in a second region, the second region being adjacent to or overlapping with the first region, and the acquisition interval between the second motion data and the first motion data being less than a first threshold; and determining the motion trajectory of the target object based on the similarity between the first motion data and the second motion data.

[0005] Furthermore, according to one aspect of the method of this disclosure, the movement vector is the displacement vector of the vertex of the smallest rectangle where the target object is located in a collection interval; the object features include at least one of the following: head region position information, body region position information, waist bending degree, arm position information, arm bending degree, leg position information, leg bending degree, and body vertical vector information.

[0006] Furthermore, according to one aspect of the method disclosed herein, determining the movement trajectory of a target object based on the similarity between first movement data and second movement data includes: obtaining movement direction similarity based on the first movement data and second movement data; obtaining object similarity between objects in the first movement data and second movement data; determining the target similarity between the first movement data and second movement data based on the movement direction similarity and object similarity; and when the target similarity reaches a preset first condition, fusing the movement trajectories of the first movement data and the second movement data to obtain the movement trajectory of the target object.

[0007] Furthermore, according to one aspect of the method of this disclosure, based on first movement data and second movement data, the similarity of movement directions is obtained, including: obtaining route information of the first movement data and route information of the second movement data; the route information is used to characterize the overall change of each movement vector; determining the vector distance between the route information of the first movement data and the route information of the second movement data; and determining the similarity of movement directions between the first movement data and the second movement data based on the vector distance.

[0008] Furthermore, according to one aspect of the method of this disclosure, obtaining route information of first movement data includes: determining movement direction information based on the movement vector of the first movement data; the movement direction information is used to characterize the change of at least one movement vector; and the movement direction information is comprehensively processed to obtain route information.

[0009] Furthermore, according to one aspect of the method of this disclosure, determining the target similarity between first moving data and second moving data based on moving direction similarity and object similarity includes: weighting the moving direction similarity and object similarity to obtain the similarity.

[0010] Furthermore, according to one aspect of the method of this disclosure, image analysis is performed on first video data collected in a first region to obtain first motion data of a target object, including: identifying the target object in the first video data based on the shape features of the target object; and extracting the first motion data of the target object.

[0011] Furthermore, according to one aspect of the method of this disclosure, the method further includes: storing the first movement data.

[0012] Furthermore, according to one aspect of the method disclosed herein, the method further includes: acquiring third motion data of each candidate object in the third video data based on third video data collected in the first region; the acquisition interval between the third motion data and the first motion data is less than a second threshold; determining the object similarity between the target object and each candidate object based on the third motion data and the first motion data; and when the similarity of any object reaches a preset second condition, fusing the motion trajectory of the third motion data and the motion trajectory of the first motion data to obtain the intra-regional trajectory of the target object.

[0013] Furthermore, according to one aspect of the method of this disclosure, determining the object similarity between the target object and each candidate object based on third movement data and first movement data includes: obtaining the degree of difference between the target object and each candidate object in each object feature; and comprehensively processing the degree of difference to obtain the object similarity between the target object and each candidate object.

[0014] Furthermore, according to one aspect of the method disclosed herein, the method further includes: performing anomaly monitoring on the target object based on first movement data and third movement data; and issuing an anomaly alarm when an anomaly occurs.

[0015] Furthermore, according to one aspect of the method disclosed, the method further includes: performing image preprocessing on the first video data and the second video data; the preprocessing includes at least one of the following: image grayscale processing, image enhancement, and image size normalization processing.

[0016] According to another aspect of this disclosure, a trajectory determination apparatus is provided, comprising: a processing unit for performing image analysis on first video data acquired in a first region to obtain first motion data of a target object; the first motion data including: a motion vector and object features; an acquisition unit for acquiring second motion data of the target object, the second motion data being obtained based on second video data acquired in a second region, the second region being adjacent to or overlapping with the first region, and the acquisition interval between the second motion data and the first motion data being less than a first threshold; and a determination unit for determining the motion trajectory of the target object based on the similarity between the first motion data and the second motion data.

[0017] According to another aspect of this disclosure, an electronic device is provided, comprising: a memory for storing computer-readable instructions; and a processor for executing the computer-readable instructions, causing the electronic device to perform the method as described in any embodiment of one aspect.

[0018] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided for storing computer-readable instructions that, when executed by a processor, cause the processor to perform the method as described in any embodiment of one aspect.

[0019] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method as described in any embodiment of one aspect.

[0020] This disclosure provides a trajectory determination method, apparatus, electronic device, storage medium, and program product. This disclosure obtains first motion data of a target object by performing image analysis on first video data collected in a first region. The first motion data is used to characterize the target object's movement vector and object features. Second motion data of the target object is obtained, which is based on second video data collected in a second region adjacent to or overlapping with the first region. The acquisition interval between the second and first motion data is less than a first threshold. The movement trajectory of the target object is determined based on the similarity between the first and second motion data. Thus, by determining the motion data of the target object in multiple monitoring areas, when monitoring areas are adjacent or overlapping, all movement trajectories of the target object are determined based on the similarity of the motion data. In summary, this disclosure uses motion data to determine the movement trajectory of a target object, and the motion data can be determined from multiple aspects, such as the target object's movement vector and object features, without relying on facial feature information. Therefore, even when the target object's face is difficult to accurately identify, the analysis of the similarity of the motion data can accurately determine all movement trajectories of the target object at a distance from the monitoring area or at adjacent / overlapping monitoring areas, thereby achieving more accurate trajectory determination of the target object.

[0021] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further illustration of the claimed technology. Attached Figure Description

[0022] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0023] Figure 1 This is a schematic diagram of a network architecture for trajectory determination provided in an embodiment of the present disclosure;

[0024] Figure 2 A flowchart of a trajectory determination method provided in this embodiment of the disclosure;

[0025] Figure 3 A schematic diagram of a movement vector provided in an embodiment of this disclosure;

[0026] Figure 4 A flowchart of a complete trajectory determination method provided for embodiments of this disclosure;

[0027] Figure 5A structural block diagram of a trajectory determination device provided in an embodiment of this disclosure;

[0028] Figure 6 A hardware block diagram of an electronic device provided in an embodiment of this disclosure;

[0029] Figure 7 This is a schematic diagram of a computer-readable storage medium provided in an embodiment of this disclosure. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this disclosure more apparent, exemplary embodiments according to this disclosure will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this disclosure, and not all embodiments of this disclosure. It should be understood that this disclosure is not limited to the exemplary embodiments described herein.

[0031] Currently, in most application scenarios, existing methods for determining a trajectory rely on acquiring facial information and then using facial features to determine the movement trajectory of the person being tracked. However, if the face appears at a distance from the surveillance camera, the area occupied by the face is very small, leading to inaccurate acquisition of facial feature information and thus making it impossible to accurately obtain all the trajectory information of the person being tracked.

[0032] Therefore, this disclosure provides a trajectory determination method that can determine the movement trajectory of a target object using motion data. This motion data can be independent of facial feature information, considering factors such as the target object's movement vector and other object characteristics. Thus, even when the target object's face is difficult to accurately identify, the analysis of motion data similarity can accurately determine all movement trajectories of the target object at a distance from the monitored area or in adjacent / overlapping areas, thereby achieving more precise trajectory determination. First, please refer to... Figure 1 , Figure 1 This is a schematic diagram of a network architecture for trajectory determination provided in an embodiment of this disclosure. Figure 1 As shown, the network architecture includes: multiple surveillance cameras, a computing unit, and a trajectory storage unit.

[0033] Specifically, surveillance cameras can acquire image data in real time. Surveillance camera 1 and surveillance camera 2 are adjacent cameras; the first area corresponding to surveillance camera 1 and the second area corresponding to surveillance camera 2 can be adjacent or overlap. For example, the edges of the first and second areas are adjacent, ensuring that the monitoring ranges of the two cameras can be effectively connected, avoiding blind spots. Alternatively, the first and second areas may partially overlap; the area of ​​the overlapping region is not limited, and image analysis of the overlapping area can achieve more accurate target tracking and identification. Similarly, surveillance camera 2 and all subsequent surveillance cameras are adjacent cameras, which will not be elaborated further.

[0034] The computing unit is primarily responsible for calculating and processing the video stream data collected from the surveillance cameras within each acquisition cycle. There is no specific limit to the acquisition cycle; it is typically once per second.

[0035] The trajectory storage unit is mainly responsible for storing the information calculated by the computing unit, and can be queried when there is an external need to access it.

[0036] Next, please refer to Figure 2 , Figure 2 This is a flowchart illustrating a trajectory determination method provided in an embodiment of this disclosure. Figure 2 As shown, the method includes:

[0037] In step S201, image analysis is performed on the first video data collected in the first region to obtain the first motion data of the target object; the first motion data includes: motion vector and object features.

[0038] In this disclosure, the first movement data can be understood as a data set of the movement status of the target object within a first area, which is multi-dimensional comprehensive data. The first movement data can be a currently collected data set of movement status, meaning it can be collected and analyzed in real time, and is applicable to scenarios such as monitoring and protecting nursing homes; or it can be a data set of historically collected movement status, meaning it can be analyzed and judged retrospectively using historical movement data, which is not applicable to nursing home monitoring and protection scenarios, but can be applied to other trajectory determination scenarios. The first movement data of this disclosure can include the movement vector of the target object and object features. The movement vector can be understood as the direction and speed of the target object's movement within the first video data of the first area, typically represented as (dx, dy). It helps determine the movement trajectory and trend of the target object. Object features can be understood as describing the shape characteristics of the target object, such as its shape, size, and local features, which help in identifying and tracking the target object.

[0039] Specifically, image analysis can be performed on the first video data collected in the first region using computer vision techniques (such as deep learning models, traditional image processing algorithms, etc.) to obtain first motion data containing the target object's movement vector and object features. A more detailed image analysis process might involve extracting the target object's features; analyzing the target object's motion within the video data sampling interval; and calculating its movement vector. The image analysis process does not have a strict order and can be flexibly adjusted according to needs or actual circumstances.

[0040] In step S202, second motion data of the target object is obtained. The second motion data is obtained based on second video data. The second video data is acquired in a second region. The second region is adjacent to or overlaps with the first region. The acquisition interval between the second motion data and the first motion data is less than a first threshold.

[0041] In this disclosure, the second motion data can be understood as obtained by image analysis of the second video data for the second region, and its specific meaning is consistent with that of the first motion data, which will not be repeated here.

[0042] It is important to note that the second area disclosed herein is adjacent to or overlaps with the first area. That is, the video data acquisition ranges of the first and second areas can be closely adjacent or partially overlap, as long as it allows for comprehensive monitoring of the target object as it moves from the first area to the second area, minimizing blind spots. This enables seamless tracking and comprehensive monitoring of the target object, ensuring its movement trajectory is completely recorded as it moves between different areas.

[0043] Meanwhile, the sampling interval between the second and first motion data is less than the first threshold. This ensures high timeliness in tracking the target object's motion state, enabling more accurate analysis of the target object's movement between different areas within a short period. The first threshold has no specific limitation and is typically set to once per second, ensuring that the acquisition times of the first and second motion data are sufficiently close for continuous monitoring. This allows for real-time tracking and precise analysis of the target object's motion state. Furthermore, continuous monitoring effectively reduces the risk of information loss and inaccurate tracking due to excessively long time intervals, providing a solid foundation for subsequent accurate trajectory determination.

[0044] Specifically, image analysis can be performed on the second video data collected in the second region using a method similar to that used to analyze the first video data. It's important to note that this sampling interval is less than a first threshold. This provides a solid foundation for subsequently and accurately determining the trajectory of the target object between different regions.

[0045] In step S203, the movement trajectory of the target object is determined based on the similarity between the first movement data and the second movement data.

[0046] In this disclosure, the similarity between first and second movement data can be determined by comparing two aspects: object features and movement vectors. If the two data identify the same target object based on object features, and the changes in the direction and speed of the target object as it moves from the first region to the second region are within a reasonable range in terms of movement vectors, then it can be determined that the trajectory of the target object in the first region and the trajectory in the second region are the same trajectory, thereby obtaining all movement trajectories of the target object.

[0047] In summary, this disclosure utilizes mobile data to determine the movement trajectory of a target object. This mobile data can be independent of facial feature information, considering factors such as the target object's movement vector and other object characteristics. Thus, even when the target object's face is difficult to accurately identify, the analysis of mobile data similarity can accurately determine all movement trajectories of the target object at a distance from the monitored area or in adjacent / overlapping areas, thereby achieving more precise trajectory determination.

[0048] The following section will elaborate on the content of the first movement data. The first movement data includes: object features and movement vectors;

[0049] Wherein, the movement vector is the displacement vector of the vertex of the smallest rectangle where the target object is located within a collection interval;

[0050] The object features include at least one of the following: head region location information, body region location information, waist curvature, arm location information, arm curvature, leg location information, leg curvature, and body vertical vector information.

[0051] In this disclosure, the first movement data may include a movement vector, which can be understood as the displacement vector of the vertices of the smallest rectangle containing the target object within a certain acquisition interval, i.e., the movement vectors of the four vertices of the rectangle containing the human body in the image. The acquisition interval is defined as one acquisition period, and is not limited thereto. (Example follows) Figure 3 A schematic diagram of the movement vector provided in the embodiments of this disclosure, such as Figure 3 As shown: The smallest rectangle where the human body is located is the dashed rectangle where the human body is located in the figure. When the human body moves from image 1 to image 2, a displacement vector can be generated, which is the four arrows a, b, c, and d with directions in the figure. At least one of the four arrows a, b, c, and d can be regarded as a displacement vector.

[0052] In this disclosure, the first movement data may further include object features, which can be understood as characteristics describing the object. In this disclosure, object features may be local features of the human body, including but not limited to at least one of the following: head region position information, body region position information, waist flexion degree, arm position information, arm flexion degree, leg position information, leg flexion degree, and body vertical vector information. Specifically, head region position information helps determine the observation direction and focus of attention of the target object; body region position information enables an overall grasp of the target object's position; the flexion degree of the waist, arms, and legs reflects the target object's posture and movement state; and body vertical vector information can be understood as the deviation value between the head region position information and the body region position information, which can assist in judging the target object's tilt state, etc.

[0053] These rich object features, combined with movement vectors, make the movement analysis of target objects more accurate and comprehensive.

[0054] As mentioned above, this disclosure can determine the movement trajectory of a target object by using the similarity between the first movement data and the second movement data. The following describes in detail the method for determining the movement trajectory, including:

[0055] Based on the first and second movement data, the similarity of movement directions is obtained;

[0056] Obtain the object similarity between each object in the first and second mobile data;

[0057] Based on the similarity of movement direction and object similarity, the target similarity between the first moving data and the second moving data is determined;

[0058] When the target similarity reaches the preset first condition, the movement trajectory of the first movement data and the movement trajectory of the second movement data are fused to obtain the movement trajectory of the target object.

[0059] In this disclosure, movement direction similarity can be understood as the degree of similarity between the movement direction represented by the first movement data and the movement direction represented by the second movement data. This can be determined by comparing the movement vectors of the first movement data and the movement vectors of the second movement data. For example, similarity can be determined by calculating the dot product of vectors, vector distance, etc. The value of movement direction similarity is usually between 0 and 1, with the value closer to 1 indicating greater similarity in movement directions.

[0060] In this disclosure, the first movement data includes not only movement vectors but also object features. When determining target similarity, it is necessary to determine not only the similarity of movement directions represented by the movement vectors but also the object similarity represented by the object features. That is, it is necessary to compare the object features in the first movement data with the object data in the second movement data to determine the object similarity between objects. This can be achieved using appropriate similarity algorithms, such as Euclidean distance or cosine similarity, to calculate the similarity between object features. The value of object similarity is typically between 0 and 1, with values ​​closer to 1 indicating greater object similarity.

[0061] Specifically, the similarity of movement directions can be determined using the movement vectors of the first and second moving data. Next, the object similarity can be determined using the object features of the first and second moving data. Finally, by combining the movement direction similarity and object similarity, the target similarity can be calculated using at least one method, such as weighted average or weighted summation, to determine whether the first and second moving data represent the same movement trajectory. This can be determined by checking if the target similarity meets a preset first condition. The preset first condition can be a preset threshold (e.g., meeting 60% or 70%) or a preset range (e.g., within 50%-60% or 60%-70%), with no specific limitation. When the target similarity meets the first condition, it can be determined that the first and second moving data represent the same trajectory. Therefore, the trajectories of both can be merged to obtain all the movement trajectories of the target object within the first and second regions. In other words, the complete movement trajectory of the target object can be obtained by combining images from various regions; there is no specific limit to the number of regions, and it can be applied to multiple monitoring areas.

[0062] The following will elaborate on each step in obtaining target similarity:

[0063] The first step is to determine the similarity of the movement directions, which can be achieved through methods such as:

[0064] Obtain the route information of the first movement data and the route information of the second movement data; the route information is used to characterize the overall change of each movement vector.

[0065] Determine the vector distance between the route information of the first movement data and the route information of the second movement data;

[0066] Based on vector distance, the similarity of the movement directions between the first and second moving data is determined.

[0067] In this disclosure, distance information can be understood as the overall change of the movement vector in space, reflecting the path length traversed by the target object during its movement. That is, distance information is the sum of the changes in the movement vector in all directions. The unit of distance information is usually also related to the unit of the movement vector.

[0068] In this disclosure, vector distance can be understood as the degree of difference between the vectors corresponding to two path information. It can be determined by calculating the difference value of the two vectors using at least one distance calculation method such as Euclidean distance or Manhattan distance.

[0069] In this disclosure, movement direction similarity can be understood as an indicator measuring the degree to which the movement directions of a target object are similar across different time periods, and it can be represented by the magnitude of the vector distance. The smaller the vector distance, the higher the movement direction similarity, indicating that the movement directions of the target object are closer in the two time periods; the larger the vector distance, the lower the movement direction similarity, indicating that the movement directions of the target object are more different in the two time periods.

[0070] Specifically, the changes in each direction of the movement vector can be obtained to determine the total distance information between the first and second movement data. Then, the vector distance between the distance information of the first and second movement data can be determined. This can be calculated using at least one method, such as Euclidean distance or Manhattan distance. Finally, based on the magnitude of the vector distance, the similarity of the movement directions between the first and second movement data is determined. The smaller the vector distance, the closer the vectors corresponding to the two distance information sets are, meaning the higher the similarity of the target object's movement direction within the time period corresponding to the first and second movement data sets; conversely, the larger the vector distance, the lower the similarity of the movement directions.

[0071] As mentioned above, the similarity of movement directions can be determined using the route information from mobile data. This disclosure also provides specific methods for determining route information, including:

[0072] Based on the movement vector of the first movement data, movement direction information is determined; the movement direction information is used to characterize the change of at least one movement vector.

[0073] The movement direction information is processed to obtain the distance information.

[0074] In this disclosure, comprehensive processing can be understood as at least one processing such as accumulation or integration. That is, the changes of at least one movement vector are obtained, and at least one processing such as accumulation or integration is performed on it to obtain distance information representing the total change.

[0075] Specifically, by analyzing at least one movement vector in the first movement data, the specific change in the target object's movement direction can be determined, thus obtaining movement direction information. Movement direction information can include the direction angle of the movement vector, the trend of direction change, etc. For example, the direction change can be determined by calculating the difference in direction change between two adjacent movement vectors within the same trajectory, or the trend of direction change can be obtained by statistically analyzing the main directions of movement vectors over a period of time. Next, the obtained movement direction information can be comprehensively processed to determine the distance information. For example, all movement direction information can be integrated or accumulated to obtain the distance information of the target object throughout the entire movement process. For example, for movement vectors in two-dimensional space, the displacement components in the horizontal and vertical directions can be accumulated separately to obtain the total distance information.

[0076] The second step is to determine the object similarity of the mobile data, using methods including:

[0077] Based on the feature data of each object, the similarity between the objects is determined.

[0078] In this disclosure, a suitable similarity calculation method can be selected to determine the similarity of the same object feature corresponding to the first and second moving data. The similarity calculation method may include, but is not limited to, at least one of the following: cosine similarity, relative proportion difference. Then, the similarity of each object feature can be combined to obtain the overall object similarity. For example, a weighted average can be applied to the similarity of each object feature, where the weights can be set according to the importance of different object features. This allows for a more comprehensive and accurate measurement of the similarity between the target objects corresponding to the first and second moving data, providing a strong basis for subsequent target object tracking and analysis.

[0079] The third step involves comprehensively considering the determined similarity of movement directions and object similarity to jointly determine the target similarity. At this stage, the methods include:

[0080] The similarity is obtained by weighting the similarity of the movement direction and the similarity of the objects.

[0081] In this disclosure, the weighting process may include, but is not limited to, at least one of the following: weighted average, weighted summation.

[0082] Specifically, different weights can be assigned to movement direction similarity and object similarity based on the actual situation. For example, if the accuracy of movement direction is more important in a specific application scenario, then a larger weight can be assigned to movement direction similarity; if the discriminative power of object features is crucial for judging the consistency of movement trajectories, then a larger weight can be assigned to object similarity. Next, the target similarity is calculated using at least one method such as weighted summation or weighted average. This makes the calculation of target similarity more flexible and can accurately reflect the degree of similarity between two movement data in different aspects. The target similarity obtained in this way can more comprehensively measure whether the first and second movement data represent the same movement trajectory, thus providing a reliable basis for subsequent trajectory fusion.

[0083] In one exemplary embodiment, the relationship between target similarity, movement direction similarity, and object similarity satisfies the following formula:

[0084] Target similarity = movement direction similarity + (1 - movement direction similarity) × object similarity.

[0085] It's important to note that the first and second steps described above do not have a strict order and can be flexibly adjusted as needed. In summary, both movement direction similarity and object similarity can be used to comprehensively determine the target similarity.

[0086] The following section details how to determine the first motion data for video data. The methods include:

[0087] Based on the shape features of the target object, target object recognition is performed in the first video data;

[0088] Extract the first movement data of the target object.

[0089] In this disclosure, shape features can be understood as denotative features that describe at least one of the following: the external outline, size, local features, gait features, etc. of the target object. Appropriate shape features can be selected for target object recognition based on different application scenarios and requirements.

[0090] Specifically, various computer vision techniques, such as deep learning detection and traditional image recognition methods, can be used to accurately identify target objects based on their shape features, based on the initial video data. After identifying the target object, its initial movement data can be extracted by analyzing changes in its position within the video data acquisition interval. For example, the coordinates of the target object in different frames can be recorded, and then the displacement vector and corresponding object feature set within the acquisition interval can be calculated to obtain the target object's movement data.

[0091] It is important to note that in this disclosure, when performing image analysis on the first video data to obtain the first motion data, this first motion data can also be stored. This facilitates subsequent data retrieval and analysis. The storage method can be saving the first motion data to a database or storing it as a file on a disk. During storage, relevant metadata such as time information and video source can be recorded simultaneously for better management and utilization of this data. When motion data similarity calculation or trajectory fusion is required, the first motion data and other relevant data can be quickly retrieved from storage, improving processing efficiency.

[0092] The following will explain in detail how, while determining the first movement data, its trajectory within the area can also be determined. In this case, the method includes:

[0093] Based on the third video data collected in the first region, the third motion data of each candidate object in the third video data is obtained; the acquisition interval between the third motion data and the first motion data is less than the second threshold.

[0094] Based on the third movement data and the first movement data, determine the object similarity between the target object and each candidate object;

[0095] When the similarity of any object reaches the preset second condition, the movement trajectory of the third movement data and the movement trajectory of the first movement data are fused to obtain the trajectory of the target object within the region.

[0096] In this disclosure, the preset second condition may include, but is not limited to, at least one of the following: a preset threshold (e.g., meeting a threshold of 60% or 70%), a preset range (e.g., within the range of 50%-60% or within the range of 60%-70%), and there is no specific limitation.

[0097] In this disclosure, when acquiring the first video data of the first region, third video data, namely historical video data, can also be acquired in the same first region. By analyzing the third motion data in the historical video data and the first motion data in the current first video data, the trajectory of the target object in the first region can be determined more accurately.

[0098] It is important to note that the acquisition interval between the third and first movement data sets needs to be less than the second threshold. This is because ensuring that these two sets of movement data are relatively close in time makes it more likely that they belong to the movement of the same target object within a similar time period. If it is uncertain whether the acquisition interval is less than the second threshold, it may lead to erroneous comparisons of movement data over a long period, reducing the accuracy of determining the target object's movement trajectory.

[0099] Specifically, third-party motion data can be acquired, and then the object features of each candidate object in the third-party motion data can be compared with the object features of the target object in the first-party motion data to obtain object similarity. This can be determined by comparing the target object with each candidate object separately. If the object similarity of any candidate object reaches a preset second condition, it indicates that the candidate object and the target object are highly likely to be the same object. At this time, the motion trajectory of the third-party motion data and the motion trajectory of the first-party motion data can be fused to obtain the trajectory of the target object in the first region. This can make full use of historical video data and current video data, improving the accuracy and reliability of target object trajectory determination.

[0100] The following will elaborate on the specific methods for determining object similarity, including:

[0101] Obtain the degree of difference between the target object and each candidate object in terms of their respective features;

[0102] The degree of difference is comprehensively processed to obtain the object similarity between the target object and each candidate object.

[0103] In this disclosure, object features can be understood as multiple features that can describe a target object or candidate object. These object features can reflect the characteristics of the target object or candidate object from different perspectives. Therefore, for each feature corresponding to the same perspective of the target object and candidate object, the degree of difference can be determined. This can be obtained by calculating at least one method such as the distance or difference between feature values. Then, the degree of difference of each perspective is comprehensively processed to obtain the object similarity between the target object and the candidate object. The comprehensive processing can include, but is not limited to, at least one of the following: normalization, weighted summation. This can more comprehensively and accurately measure the degree of similarity between the target object and the candidate object.

[0104] In one exemplary embodiment, the calculation of object similarity satisfies the following formula:

[0105]

[0106] The eigenvalue distance satisfies the following formula:

[0107]

[0108] The eigenvalue lengths satisfy the following formula:

[0109]

[0110] Where n represents the dimension of the feature value; k represents the feature value information under the same dimension; the standard feature value represents the current feature value information, equivalent to the object feature of the first mobile data of this disclosure; and the historical feature value represents the historical feature value information, equivalent to the object feature of the third mobile data of this disclosure. As can be seen from the above formula, the difference can be determined for the feature value of each dimension, and then summed for normalization to obtain the final object similarity.

[0111] The following will elaborate on how, while determining the trajectory within a region, it is also possible to issue anomaly alerts for target objects. The methods include:

[0112] Based on the first and third movement data, anomaly monitoring is performed on the target object;

[0113] An alarm will be triggered when an abnormal situation occurs.

[0114] In this disclosure, when the first and third movement data of the target object are determined, the movement status of the target object can be continuously monitored according to preset anomaly judgment rules. The preset anomaly judgment rules may include, but are not limited to, at least one of the following: body vector angle less than a set threshold, body waist shaking deviation exceeding a set threshold, and arm shaking deviation exceeding a set threshold. When the movement data meets the preset anomaly judgment rules, the target object can be considered abnormal, and an anomaly alarm can be triggered. For example, if the body vector angle is less than the set threshold, it is considered that the target object has fallen, and a fall warning is issued; if the body waist shaking deviation exceeds the set threshold, it is considered that the target object has experienced a recurrence of a lumbar disease, and a lumbar disease anomaly warning is issued; if the arm shaking deviation exceeds the set threshold, it is considered that the target object has an abnormal mental state, and a physical disease anomaly warning is issued.

[0115] The following details how image preprocessing can be performed when determining motion data from video data. Specific methods include:

[0116] Image preprocessing is performed on the first and second video data;

[0117] Preprocessing includes at least one of the following: image grayscale conversion, image enhancement, and image size normalization.

[0118] In this disclosure, image preprocessing can be performed on the first video data and / or the second video data. This preprocessing may include, but is not limited to, at least one of the following: image grayscale conversion, image enhancement, and image size normalization. Specifically, image grayscale conversion can be understood as converting a color image to a grayscale image, which provides clearer edge and texture information. Image enhancement can improve image quality, highlight the features of the target object, and improve image contrast and clarity. Image size normalization can uniformly adjust images of different resolutions and sizes to the same size. This facilitates consistent analysis and processing of different images in subsequent processing.

[0119] By performing the above preprocessing operations on the first and second video data, better image quality and a more unified data format can be provided for subsequent tasks such as target object recognition, motion data extraction, and trajectory determination, thereby improving the performance and reliability of the entire system.

[0120] In one exemplary embodiment, Figure 4 A flowchart of the complete trajectory determination method provided in the embodiments of this disclosure is shown below. Figure 4 As shown, the method includes:

[0121] Step 1: Initialize the camera configuration. The configuration includes the configuration information of the surveillance cameras (the computing unit can obtain the surveillance images from the cameras based on the configuration information) and the spatial relationship between the cameras (including overlapping areas between cameras). Cameras with overlapping monitoring areas are named adjacent cameras.

[0122] Step Two: Acquire surveillance images from the surveillance camera within the enclosed scene at a fixed acquisition frequency (which can be agreed upon as once per second), preprocess the surveillance images (equivalent to the image preprocessing disclosed herein), and cache the initialized images. The initialized images will subsequently be named "preprocessed images." (Image initialization includes converting the image to grayscale and then processing the grayscale image into a human body sharpening image).

[0123] Step 3: By recognizing the preprocessed image, detect whether there is a human body (equivalent to the target object recognition of this disclosure);

[0124] Step 4: If the image does not contain a human body, return to Step 2. If the image contains human body information, the preprocessed image is used to obtain the regional location information of various parts of the human body. Human body feature values ​​(equivalent to the object features of this disclosure) are calculated based on the regional information between these parts, and can be named standard feature values. The location information of the human body in the image (equivalent to the movement vector of this disclosure) is also included. The human body feature values, location information, and time node information are then cached. (Due to different monitoring angles, different monitoring cameras may calculate different feature values ​​and location information for the same human body.)

[0125] The feature values ​​are multidimensional feature coefficients calculated using the top-left corner of the image plane as the origin, based on the feature information of various body parts and the correlation information between body parts. These feature coefficients include head region position information, body region position information, waist curvature, arm position information and curvature, leg position information and curvature, and a body vertical vector. The body vertical vector represents the deviation between the head region position information and the body region position information, indicating the degree of body tilt.

[0126] Step 5: Extract several historical feature values ​​from the previous two sampling periods of the camera, calculate the human body similarity (equivalent to the object similarity of this disclosure) by using the historical human body feature values ​​and standard feature values ​​(according to the actual human body similarity threshold set (default is 60%), when the similarity is greater than the threshold, it is judged to be the same person), and determine whether the historical images contain the same person.

[0127] Step 6: If the results calculated in the historical data and the latest data within the specified sampling period do not contain the same person, return to Step 2; if they contain the same person, calculate the movement coefficient (equivalent to the first movement data of this disclosure) using the location information in the historical data and the location information in the latest sampling data, and then cache the movement coefficient, camera information, and time node information and store them in the trajectory storage unit.

[0128] The movement coefficients are calculated by obtaining four movement vectors based on the temporal order of the movement of the four vertices of the rectangle containing the human figure in the image. The movement coefficients are then a set of the four movement vectors and standard eigenvalues.

[0129] It should be noted that steps seven and eight can be used alone or in combination with the steps mentioned above; there are no specific restrictions here.

[0130] Step 7: Obtain several historical feature values ​​from the previous two sampling periods for the same person in the previous step, and perform calculations, including calculating the degree deviation and tremor deviation of the waist bending, and caching the waist bending deviation; calculating the degree deviation and tremor deviation of the arm bending, and caching the arm bending deviation; calculating the degree deviation and tremor deviation of the leg bending, and caching the leg bending deviation; calculating and caching the vertical vector information of the body.

[0131] Step 8: Perform a comprehensive analysis based on the cached information from the previous step, and issue anomaly alerts based on the analysis results (equivalent to the anomaly alerts disclosed herein). This includes alerting the elderly person as having fallen if the body vector angle is less than a set threshold; alerting the elderly person as having fallen if the deviation of the waist tremor exceeds a set threshold; and alerting the elderly person as having a waist disease abnormality if the deviation of the arm tremor exceeds a set threshold.

[0132] Step Nine: After the above steps, obtain the historical motion coefficients of the two most recent sampling periods of adjacent cameras, calculate the motion coefficient similarity between the motion coefficients (equivalent to the target similarity of this disclosure), and determine whether they are the same trajectory (set the motion trajectory similarity threshold according to the actual situation (default is 60%); if the motion similarity is greater than the threshold, it is determined to be the same trajectory). If they are the same trajectory, cache the trajectory information and synchronize it to the trajectory storage unit.

[0133] Step 10: Return to Step 2 and repeat until Step 9.

[0134] This method allows us to extract the complete trajectory information of a target object in a closed scene.

[0135] This disclosure also provides a trajectory determination device. Figure 5 This is a structural block diagram of a trajectory determination device provided in an embodiment of the present disclosure, such as... Figure 5 As shown, the trajectory determining device 500 includes:

[0136] The processing unit 501 is used to perform image analysis on the first video data collected in the first region to obtain the first motion data of the target object; the first motion data includes: motion vector and object features.

[0137] The acquisition unit 502 is used to acquire second movement data of the target object. The second movement data is obtained based on second video data. The second video data is acquired in a second region. The second region is adjacent to or overlaps with the first region. The acquisition interval between the second movement data and the first movement data is less than a first threshold.

[0138] The determining unit 503 is used to determine the movement trajectory of the target object based on the similarity between the first movement data and the second movement data.

[0139] In one exemplary embodiment, the movement vector is the displacement vector of the smallest rectangle where the target object is located within a collection interval; the object features include at least one of the following: head region position information, body region position information, waist bending degree, arm position information, arm bending degree, leg position information, leg bending degree, and body vertical vector information.

[0140] In one exemplary embodiment, determining the movement trajectory of a target object based on the similarity between first movement data and second movement data includes: obtaining movement direction similarity based on the first movement data and second movement data; obtaining object similarity between objects in the first movement data and second movement data; determining target similarity between the first movement data and second movement data based on movement direction similarity and object similarity; and when the target similarity reaches a preset first condition, fusing the movement trajectories of the first movement data and the second movement data to obtain the movement trajectory of the target object.

[0141] In one exemplary embodiment, obtaining the similarity of movement directions based on first movement data and second movement data includes: obtaining route information of the first movement data and route information of the second movement data; the route information is used to characterize the overall change of each movement vector; determining the vector distance between the route information of the first movement data and the route information of the second movement data; and determining the similarity of movement directions between the first movement data and the second movement data based on the vector distance.

[0142] In one exemplary embodiment, obtaining the route information of the first movement data includes: determining movement direction information based on the movement vector of the first movement data; the movement direction information is used to characterize the change of at least one movement vector; and the movement direction information is comprehensively processed to obtain the route information.

[0143] In one exemplary embodiment, determining the target similarity between the first moving data and the second moving data based on the similarity of the moving direction and the similarity of the object includes: weighting the similarity of the moving direction and the similarity of the object to obtain the similarity.

[0144] In one exemplary embodiment, image analysis is performed on the first video data collected in the first region to obtain the first movement data of the target object, including: identifying the target object in the first video data based on the shape features of the target object; and extracting the first movement data of the target object.

[0145] In one exemplary embodiment, the method further includes storing the first movement data.

[0146] In one exemplary embodiment, the method further includes: acquiring third motion data of each candidate object in the third video data based on the third video data collected in the first region; the acquisition interval between the third motion data and the first motion data is less than a second threshold; determining the object similarity between the target object and each candidate object based on the third motion data and the first motion data; and when the similarity of any object reaches a preset second condition, fusing the motion trajectory of the third motion data and the motion trajectory of the first motion data to obtain the trajectory of the target object within the region.

[0147] In one exemplary embodiment, determining the object similarity between the target object and each candidate object based on third movement data and first movement data includes: obtaining the degree of difference between the target object and each candidate object in each object feature; and performing comprehensive processing on the degree of difference to obtain the object similarity between the target object and each candidate object.

[0148] In one exemplary embodiment, the method further includes: performing anomaly monitoring on the target object based on the first movement data and the third movement data; and issuing an anomaly alarm when an anomaly occurs.

[0149] In one exemplary embodiment, the method further includes: performing image preprocessing on the first video data and the second video data; the preprocessing includes at least one of the following: image grayscale processing, image enhancement, and image size normalization processing.

[0150] Figure 6 This is a hardware block diagram of an electronic device provided according to an embodiment of the present disclosure. The electronic device 600 according to an embodiment of the present disclosure includes at least a processor; and a memory for storing computer-readable instructions. When the computer-readable instructions are loaded and executed by the processor, the processor performs a trajectory determination method according to any of the preceding embodiments of the present disclosure.

[0151] Figure 6 The illustrated electronic device 600 specifically includes a central processing unit (CPU) 601, a graphics processing unit (GPU) 602, and a memory 603. These units are interconnected via a bus 604. The CPU 601 and / or GPU 602 can function as the aforementioned processor, and the memory 603 can function as the aforementioned memory storing computer-readable instructions. Furthermore, the electronic device 600 may also include a communication unit 605, a storage unit 606, an output unit 607, an input unit 608, and an external device 609, all of which are also connected to the bus 604.

[0152] Figure 7 This is a schematic diagram of a computer-readable storage medium provided in an embodiment of this disclosure. (As shown...) Figure 7As shown, a computer-readable storage medium 700 according to an embodiment of the present disclosure stores computer-readable instructions 701 thereon. When the computer-readable instructions 701 are executed by a processor, the trajectory determination method according to any embodiment of the present disclosure described above with reference to the accompanying drawings is performed. The computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0153] This disclosure further provides a computer program product, including a computer program that, when executed by a processor, implements the trajectory determination method of any of the preceding embodiments of this disclosure.

[0154] The present disclosure provides a trajectory determination method, apparatus, electronic device, storage medium, and program product. The present disclosure obtains first motion data of a target object by performing image analysis on first video data collected in a first region; the first motion data is used to characterize the movement vector and object features of the target object; second motion data of the target object is obtained, the second motion data being derived from second video data collected in a second region, the second region being adjacent to or overlapping with the first region, and the acquisition interval between the second and first motion data being less than a first threshold; the movement trajectory of the target object is determined based on the similarity between the first and second motion data. Thus, by determining the motion data of the target object in multiple monitoring areas, when monitoring areas are adjacent or overlapping, all movement trajectories of the target object are determined based on the similarity of the motion data. In summary, the present disclosure determines the movement trajectory of a target object using motion data, which can be independent of facial feature information, but rather considers multiple aspects such as the movement vector and object features of the target object. In this way, even when the face of the target object is difficult to identify accurately, the analysis of the similarity of mobile data can accurately determine all the movement trajectories of the target object in the distance of the monitored area and in the adjacent / overlapping areas of the monitored area, thereby achieving more accurate trajectory determination of the target object.

[0155] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0156] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0157] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0158] Additionally, as used herein, the “or” used in a list of items beginning with “at least one” indicates a separate list, such that a list of, for example, “at least one of A, B, or C” means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word “exemplary” does not imply that the described example is preferred or better than other examples.

[0159] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0160] Various changes, substitutions, and modifications can be made to the technology herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0161] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0162] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A trajectory determination method, characterized by, The method includes: Image analysis is performed on the first video data collected in the first region to obtain the first motion data of the target object; the first motion data includes: motion vector and object features; Acquire second motion data of the target object. The second motion data is obtained based on second video data. The second video data is acquired in a second region. The second region is adjacent to or overlaps with the first region. The acquisition interval between the second motion data and the first motion data is less than a first threshold. Based on the similarity between the first movement data and the second movement data, the movement trajectory of the target object is determined; Wherein, the movement vector is the displacement vector of the vertex of the smallest rectangle where the target object is located within a collection interval; The object features include at least one of the following: head region location information, body region location information, waist curvature, arm location information, arm curvature, leg location information, leg curvature, and body vertical vector information. Determining the movement trajectory of the target object based on the similarity between the first movement data and the second movement data includes: Based on the first movement data and the second movement data, the similarity of movement directions is obtained; Obtain the object similarity between each object in the first mobile data and the second mobile data; Based on the movement direction similarity and the object similarity, the target similarity between the first movement data and the second movement data is determined; When the target similarity reaches a preset first condition, the movement trajectory of the first movement data and the movement trajectory of the second movement data are fused to obtain the movement trajectory of the target object; The step of obtaining the similarity of movement directions based on the first movement data and the second movement data includes: Obtain the route information of the first movement data and the route information of the second movement data; the route information is used to characterize the overall change of each movement vector. Determine the vector distance between the route information of the first mobile data and the route information of the second mobile data; Based on the vector distance, the similarity of the movement directions between the first moving data and the second moving data is determined.

2. The method of claim 1, wherein, The process of obtaining the route information of the first movement data includes: Based on the movement vector of the first movement data, movement direction information is determined; the movement direction information is used to characterize the change of at least one of the movement vectors. The movement direction information is processed to obtain the distance information.

3. The method of claim 1, wherein, Determining the target similarity between the first moving data and the second moving data based on the movement direction similarity and the object similarity includes: The similarity is obtained by weighting the similarity of the movement direction and the similarity of the object.

4. The method of claim 1, wherein, The step of performing image analysis on the first video data collected in the first region to obtain the first movement data of the target object includes: Based on the shape features of the target object, target object recognition is performed in the first video data; Extract the first movement data of the target object.

5. The method of claim 4, wherein, The method further includes storing the first movement data.

6. The method of claim 1, wherein, The method further includes: Based on the third video data collected in the first region, the third motion data of each candidate object in the third video data is obtained; the acquisition interval between the third motion data and the first motion data is less than a second threshold. Based on the third movement data and the first movement data, the object similarity between the target object and each candidate object is determined; When the similarity of any object reaches the preset second condition, the movement trajectory of the third movement data and the movement trajectory of the first movement data are fused to obtain the trajectory of the target object within the region.

7. The method of claim 6, wherein, The step of determining the object similarity between the target object and each candidate object based on the third movement data and the first movement data includes: Obtain the degree of difference between the target object and each candidate object in each object feature; The differences are comprehensively processed to obtain the object similarity between the target object and each candidate object.

8. The method according to claim 6, characterized in that, The method further includes: Based on the first movement data and the third movement data, the target object is monitored for anomalies. An alarm will be triggered when an abnormal situation occurs.

9. The method according to claim 1, characterized in that, The method further includes: Image preprocessing is performed on the first video data and the second video data; The preprocessing includes at least one of the following: image grayscale conversion, image enhancement, and image size normalization.

10. A trajectory determining device, characterized in that, The device includes: The processing unit is used to perform image analysis on the first video data collected in the first region to obtain the first motion data of the target object; the first motion data includes: motion vector and object features; The acquisition unit is used to acquire second movement data of the target object. The second movement data is obtained based on second video data. The second video data is acquired in a second region. The second region is adjacent to or overlaps with the first region. The acquisition interval between the second movement data and the first movement data is less than a first threshold. The determining unit is configured to determine the movement trajectory of the target object based on the similarity between the first movement data and the second movement data; Wherein, the movement vector is the displacement vector of the vertex of the smallest rectangle where the target object is located within a collection interval; The object features include at least one of the following: head region location information, body region location information, waist curvature, arm location information, arm curvature, leg location information, leg curvature, and body vertical vector information. Determining the movement trajectory of the target object based on the similarity between the first movement data and the second movement data includes: Based on the first movement data and the second movement data, the similarity of movement directions is obtained; Obtain the object similarity between each object in the first mobile data and the second mobile data; Based on the movement direction similarity and the object similarity, the target similarity between the first movement data and the second movement data is determined; When the target similarity reaches a preset first condition, the movement trajectory of the first movement data and the movement trajectory of the second movement data are fused to obtain the movement trajectory of the target object; The step of obtaining the similarity of movement directions based on the first movement data and the second movement data includes: Obtain the route information of the first movement data and the route information of the second movement data; the route information is used to characterize the overall change of each movement vector. Determine the vector distance between the route information of the first mobile data and the route information of the second mobile data; Based on the vector distance, the similarity of the movement directions between the first moving data and the second moving data is determined.

11. An electronic device, characterized in that, include: Memory, used to store computer-readable instructions; as well as A processor for executing the computer-readable instructions, causing the electronic device to perform the method as described in any one of claims 1-9.

12. A non-transitory computer-readable storage medium for storing computer-readable instructions, characterized in that, When the computer-readable instructions are executed by a processor, the processor performs the method as described in any one of claims 1-9.

13. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1-9.

Images