Method for correcting device position anomaly and storage medium

Abstract

The application discloses a device position exception correction method and a storage medium. The correction method is applied to a device management end, the device management end is used for managing front-end devices, and the method comprises the following steps: acquiring reference information corresponding to each front-end device in a preset area; determining a high-confidence abnormal device and a target normal device associated with the high-confidence abnormal device from each front-end device according to the reference information; and performing position correction processing on the corresponding high-confidence abnormal device according to the target normal device, so as to obtain target position information of the high-confidence abnormal device. The above scheme can efficiently and accurately correct the position of the device with the position exception.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device positioning, in particular to a device position abnormality correction method and a storage medium. BACKGROUND

[0002] With the rapid development of communication technology, multiple front-end devices are often deployed in a designated area to collect relevant information, and intelligent analysis is performed on the relevant information to achieve related tasks. Therefore, the positioning accuracy of the front-end device is one of the important factors for the smooth implementation of the task.

[0003] At present, the method for positioning the front-end device generally needs relevant staff to conduct on-site survey and registration. The manual positioning method often has position record deviation or error, and even if the manual positioning is accurate, there may be device positioning drift problems during the operation of the front-end device, which will cause great trouble to the subsequent information analysis process.

[0004] For the front-end device with position abnormality, how to efficiently and accurately correct the position is a problem to be solved. SUMMARY

[0005] The present application at least provides a device position abnormality correction method, device, apparatus and computer readable storage medium.

[0006] The first aspect of the present application provides a device position abnormality correction method, which is applied to a device management end for managing front-end devices, and includes:

[0007] Obtaining reference information corresponding to each front-end device in a preset area, the reference information including device information of the front-end device and collection information obtained by the front-end device collecting a target object, or the reference information including device information of the front-end device, collection information corresponding to the front-end device and region information of a region of interest in the preset area; determining a high-confidence abnormal device and a target normal device associated with the high-confidence abnormal device from each front-end device according to the reference information; and performing position correction processing on the corresponding high-confidence abnormal device according to the target normal device, to obtain target position information of the high-confidence abnormal device.

[0008] In an embodiment, the reference information comprises acquisition information acquired by the front-end device on the target object, the acquisition information comprises a moving speed of the target object, and the position correction processing on the corresponding high-confidence abnormal device according to the target normal device comprises: acquiring abnormal position information of a currently analyzed high-confidence abnormal device and normal position information of a target normal device associated with the currently analyzed high-confidence abnormal device; determining a reference moving time of the target object moving from the currently analyzed high-confidence abnormal device to the target normal device at the moving speed according to the moving speed, the normal position information, and the abnormal position information; determining a moving time interval of the target object according to the reference moving time and a preset time difference threshold; determining a target point reached by the target object in the moving time interval according to the moving speed and the normal position information; and determining the target position information according to a quantity of the target points and position information of the target points.

[0009] In an embodiment, the determination of the target position information according to the quantity of the target points and the position information of the target points comprises: in response to the quantity of the target points being one, determining the position information of the target point as the target position information; and in response to the quantity of the target points being multiple, determining position information of a center point of the multiple target points as the target position information.

[0010] In an embodiment, the reference information comprises acquisition information acquired by the front-end device on the target object, the acquisition information comprises trajectory information of the target object, and the determination of the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device from the front-end devices according to the reference information comprises: performing scoring processing on the corresponding front-end device according to the reference information and a preset scoring strategy to obtain a target score; determining a front-end device with a target score less than a normal score threshold as a high-confidence normal device, the normal score threshold being less than an abnormal score threshold; determining a front-end device with a target score greater than the abnormal score threshold as the high-confidence abnormal device; and determining the high-confidence normal device associated with the high-confidence abnormal device according to the trajectory information to obtain the target normal device.

[0011] In an embodiment, the preset scoring strategy comprises multiple scoring strategies, and the scoring processing on the corresponding front-end device according to the reference information and the preset scoring strategy to obtain the target score comprises: performing scoring processing on a currently analyzed front-end device according to the reference information of the currently analyzed front-end device and multiple preset scoring strategies respectively to obtain initial scores corresponding to the preset scoring strategies; and performing summation processing on the initial scores to obtain the target score of the currently analyzed front-end device.

[0012] In an embodiment, the reference information further comprises device information of the front-end device and region information of a region of interest in the preset region, the device information comprising device position information and device name, and the region information comprising region position information and region name, and the scoring processing of the corresponding front-end device according to the reference information and a preset scoring strategy to obtain a target score comprises: searching for a target region of interest in each region of interest to which the current analyzed front-end device is pre-bound; in response to the current analyzed front-end device being in the target region of interest, determining a name similarity of the current analyzed front-end device according to the device name of the current analyzed front-end device and the region name of the corresponding target region of interest; and performing scoring processing according to the name similarity and a similarity threshold to obtain the target score of each front-end device.

[0013] In an embodiment, the collected information further comprises a collection time, and the scoring processing of the corresponding front-end device according to the reference information and a preset scoring strategy to obtain a target score comprises: performing sorting processing on each front-end device according to the collection time to obtain sorted front-end devices; obtaining a moving speed of the target object between two front-end devices in sequence; determining two front-end devices corresponding to a moving speed greater than a preset speed threshold as candidate abnormal devices and recording an abnormal judgment number; determining an associated front-end device corresponding to the current analyzed candidate abnormal device according to the trajectory information; and performing scoring according to the abnormal judgment number of the current analyzed candidate abnormal device and the abnormal judgment number of the associated front-end device to obtain the target score of the current analyzed candidate abnormal device.

[0014] In an embodiment, the scoring processing of the corresponding front-end device according to the reference information and a preset scoring strategy to obtain a target score comprises: determining a neighboring device of the current analyzed front-end device according to the trajectory information; and performing scoring according to distance information between the current analyzed front-end device and the neighboring device to obtain the target score of the current analyzed front-end device.

[0015] In an embodiment, the collected information further comprises a moving speed, and the scoring processing of the corresponding front-end device according to the reference information and a preset scoring strategy to obtain a target score comprises: determining a transfer probability between each front-end device according to the trajectory information; determining a candidate front-end device in which the transfer probability is greater than a transfer probability threshold; determining a candidate abnormal device associated with each candidate front-end device according to a preset abnormal data analysis method and a moving speed of the target object between each candidate front-end device; and performing scoring according to a number of candidate abnormal devices associated with the current analyzed candidate front-end device to obtain the target score of the candidate front-end device.

[0016] The second aspect of the present application provides a device position anomaly correction device, which is applied to a device management end for managing front-end devices, and includes: an information acquisition module, configured to acquire reference information corresponding to each front-end device in a preset area; a device determination module, configured to determine a high-confidence anomaly device and a target normal device associated with the high-confidence anomaly device from each front-end device according to the reference information; and a position correction module, configured to perform position correction processing on the corresponding high-confidence anomaly device according to the target normal device, to obtain target position information of the high-confidence anomaly device.

[0017] The third aspect of the present application provides an electronic device including a memory and a processor, the processor being configured to execute program instructions stored in the memory to implement the device position anomaly correction method described above.

[0018] The fourth aspect of the present application provides a computer-readable storage medium having program instructions stored thereon, the program instructions being executed by a processor to implement the device position anomaly correction method described above.

[0019] The above scheme is applied to a device management end for managing front-end devices, and performs multi-dimensional analysis by acquiring reference information corresponding to each front-end device in a preset area, so as to determine a high-confidence anomaly device and a target normal device associated with the high-confidence anomaly device from each front-end device according to the reference information. The high-confidence anomaly device is a front-end device that is more likely to have a position anomaly, and the target normal device is a front-end device that is associated with the high-confidence anomaly device and has a normal position. Therefore, the corresponding high-confidence anomaly device can be subjected to position correction processing according to the target normal device, to obtain target position information of the high-confidence anomaly device, thereby realizing position correction of the front-end device with a position anomaly. Compared with the traditional position correction method in which a standard address needs to be set in advance in the front-end device, and then the position is matched according to the standard address and the latitude and longitude information, in most cases, the front-end device may not be set with a standard address and the latitude and longitude matching effect is poor, which is difficult to effectively apply. The method of the present application improves the accuracy and efficiency of position anomaly correction.

[0020] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, but not limiting the present application. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the specification, serve to explain the technical solutions of the present application.

[0022] Figure 1 is a flowchart of an exemplary embodiment of the device position anomaly correction method of the present application;

[0023] Figure 2 This is an exemplary position correction scenario diagram in the device position anomaly correction method of this application;

[0024] Figure 3 This is an exemplary interactive interface diagram of the device position anomaly correction method of this application;

[0025] Figure 4 This is a block diagram illustrating a device for correcting abnormal device position, as shown in an exemplary embodiment of this application.

[0026] Figure 5 This is a schematic diagram of the structure of an embodiment of the electronic device of this application;

[0027] Figure 6 This is a schematic diagram of the structure of an embodiment of the computer-readable storage medium of this application. Detailed Implementation

[0028] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0029] 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.

[0030] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "many" in this document means two or more. Moreover, the term "at least one" in this document means any combination of at least two of any one or more of a plurality of objects. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0031] To facilitate understanding, one applicable scenario of this application is illustrated below. With the rapid development of communication technology, multiple front-end devices are commonly deployed in designated areas across various application scenarios to collect relevant information and perform intelligent analysis to achieve related tasks. Examples include smart cities and industrial production. These front-end devices can collect information on various types of target objects within the application scenario; for instance, intelligent traffic management can be achieved by collecting vehicle trajectory information and combining it with a spatiotemporal model, which will not be elaborated upon here. Therefore, the positioning accuracy of the front-end devices is one of the crucial factors for successfully completing the task.

[0032] Currently, the method for positioning the front-end device generally needs relevant staff to conduct field survey and manual registration. The manual positioning method often has position record deviation or error. Even if the manual positioning is accurate, there may be device positioning drift problem in the running process of the front-end device, which will cause great trouble to the subsequent information analysis process. Therefore, for the front-end device with position anomaly, how to efficiently and accurately correct the position is a problem to be solved.

[0033] Please refer to Figure 1 , Figure 1 is a flowchart of an exemplary embodiment of the device position anomaly correction method of the present application. The method is applied to a device management end, which is used to manage front-end devices. Specifically, it can include the following steps:

[0034] In step S110, the reference information corresponding to each front-end device in a preset area is obtained.

[0035] The reference information can include the device information of the front-end device and the collection information obtained by the front-end device collecting the target object, or the reference information includes the device information of the front-end device, the collection information corresponding to the front-end device, and the region information of the region of interest in the preset area.

[0036] Among them, the preset area refers to the area where the front-end device is deployed. The preset area can have one or more, each preset area can have one or more front-end devices, each preset area can also include one or more regions of interest, and the target object can have one or more (one or more kinds), different front-end devices can collect the same target object or different target objects, which is not limited here.

[0037] The reference information corresponding to each front-end device can include but is not limited to the device information of the front-end device and the collection information obtained by the front-end device collecting the target object, or it can also include the region information of the region of interest in the preset area where the front-end device is located on this basis. Among them, the device information can include but is not limited to the device identification (device id), the device position (such as the latitude and longitude and / or other coordinate system coordinate information, etc.), the device name and other information. The collection information can include but is not limited to the object identification (object id) of the collected target object, the collection time, the device id of the collected target object, and other information. The region information can include but is not limited to the region id of the region of interest, the region position (such as the region range, the region latitude and longitude, etc.), the region name and other information.

[0038] It should be noted that the types and functions of the various front-end devices of the present application can be the same or different, and are not limited herein. When the various front-end devices are analyzed and corrected for position abnormalities in the present application, if there are multiple types of front-end devices, the analysis and correction can be performed separately according to the types (types and / or functions) of the front-end devices, which will not be repeated herein. For example, the various front-end devices can include image acquisition devices, some of which can acquire motor vehicles as target objects, and some of which can acquire non-motor vehicles as target objects. In addition, the front-end devices can also include RFID sensing devices, which can acquire targets with RFID as target objects. Or some are image acquisition devices, and some are radar acquisition devices, etc.

[0039] In step S120, a high-confidence abnormal device and a target normal device associated with the high-confidence abnormal device are determined from the various front-end devices according to the reference information.

[0040] The high-confidence abnormal device refers to a front-end device that is more likely (confidence) to have a position abnormality. The high-confidence abnormal device can be associated with a front-end device with a position abnormality, or with a front-end device with a normal position. The target normal device can be a front-end device with a normal position associated with the high-confidence abnormal device, or a front-end device with a normal position that is more likely (confidence) to have a normal position among the front-end devices with normal positions associated with the high-confidence abnormal device. The present application does not limit the number of high-confidence abnormal devices and target normal devices. For example, one high-confidence abnormal device can be associated with zero, one or more target normal devices. The association between the high-confidence abnormal device and the associated target normal device can be pre-set or obtained by analyzing the reference information, which will not be repeated herein.

[0041] In combination with the foregoing steps, after obtaining the reference information corresponding to the front-end devices, the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device can be determined from the various front-end devices by analyzing the reference information. Since the target normal device is associated with the corresponding high-confidence abnormal device, the abnormal position information of the high-confidence abnormal device can be corrected according to the normal position information of the target normal device.

[0042] For example, the present application can provide one or more implementable analysis methods for analyzing the reference information; or alternatively, the type of reference information obtained can be selected from a plurality of analysis methods, which will not be repeated herein.

[0043] In step S130, the corresponding high-confidence abnormal device is corrected for position according to the target normal device, and target position information of the high-confidence abnormal device is obtained.

[0044] According to the foregoing steps, after the high-confidence abnormal device and the associated target normal device are determined, the abnormal position information of the corresponding high-confidence abnormal device can be corrected according to the normal position information of the target normal device to obtain the target position information (i.e., the corrected position information) of the high-confidence abnormal device.

[0045] Optionally, in addition to correcting the position of the corresponding high-confidence abnormal device according to the target normal device, a position correction interface for human-computer interaction can also be provided, and the user can input a related control instruction based on the position correction interface. In response to receiving the control instruction, the position of the high-confidence abnormal device is corrected according to the control instruction (for example, the control instruction contains position information for the position correction processing). In addition, the position correction interface can also support displaying the reference information of each front-end device, and then the abnormal device can be filtered through the map information and a manual correction function can be provided, and the target position information determined by the foregoing embodiments can also be compared and displayed with the actual manually corrected position information, and the like, which will not be described here.

[0046] As can be seen, the reference information corresponding to each front-end device in the preset area is obtained for multi-dimensional analysis, so that the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device can be determined from each front-end device according to the reference information. The high-confidence abnormal device is a front-end device that is more likely to have a position anomaly, and the target normal device is a front-end device that is associated with the high-confidence abnormal device and has a normal position. Therefore, the corresponding high-confidence abnormal device can be corrected according to the target normal device to obtain the target position information of the high-confidence abnormal device, and the position of the front-end device with a position anomaly is corrected. Compared with the traditional position correction method in which a standard address needs to be set in the front-end device, and then the standard address is matched with the latitude and longitude information to correct the position, in most cases, the front-end device may not set the standard address and the latitude and longitude matching effect is poor, which is difficult to effectively apply. The method of the present application improves the accuracy and efficiency of position anomaly correction.

[0047] On the basis of the foregoing embodiments, the step of correcting the position of the corresponding high-confidence abnormal device according to the target normal device to obtain the target position information of the high-confidence abnormal device is described. The reference information includes the collection information of the target object collected by the front-end device, and the collection information includes the moving speed of the target object. Specifically, the method of the present embodiment includes the following steps:

[0048] obtain abnormal position information of the high-confidence abnormal device under current analysis and normal position information of the target normal device associated with the high-confidence abnormal device under current analysis; determine a reference moving time of the target object moving from the high-confidence abnormal device under current analysis to the target normal device at the moving speed according to the moving speed, the normal position information and the abnormal position information; determine a moving time interval of the target object according to the reference moving time and the preset time difference threshold; determine a target point reached by the target object within the moving time interval according to the moving speed and the normal position information; and determine target position information according to the number of target points and the position information of the target points.

[0049] With the foregoing embodiments, in the scenario where the reference information includes device information and collection information, if the high-confidence abnormal device and the target normal device associated therewith are determined, the abnormal position information of the high-confidence abnormal device under current analysis and the normal position information of the target normal device associated therewith can be obtained. The high-confidence abnormal device can be one or more. In the correction method of the present application, each high-confidence abnormal device can be analyzed and corrected one by one, or multiple high-confidence abnormal devices can be concurrently analyzed and corrected, which is not limited herein. For ease of illustration, the high-confidence abnormal device under current analysis and the target normal device associated therewith are mainly described in the examples of the present application.

[0050] It should be noted that the position information of the present application can be represented in various forms. For ease of understanding, the position information is mainly represented by longitude and latitude coordinates in the subsequent examples, but in actual application scenarios, the position information can also be represented by coordinates in other coordinate systems, which is not described herein.

[0051] For example, as shown in FIG. 1, Figure 2 Figure 2 is a schematic diagram of an example position correction in the device position abnormality correction method of the present application. The front-end device D is the high-confidence abnormal device under current analysis, the front-end devices associated therewith include A, B, C, E, F and G, and the target normal devices associated therewith are determined to include C, E, F and G from A, B, C, E, F and G. t1-t9 are the reference moving times consumed by the target object moving between devices at the moving speed.

[0052] ​According to the normal position information and the abnormal position information, distance information between the high-confidence abnormal device in the position abnormal state and the target normal device associated therewith can be determined. According to the moving speed of the target object, reference moving time that should be consumed by the target object to move from the abnormal position information to the normal position information (or from the normal position information to the abnormal position information) can be calculated, i.e., t5, t6, t7, and t9. Then, on the basis of the reference moving time, a preset time difference threshold T can be added to determine a moving time interval of the target object (for example, [t5, t5+T], [t6, t6+T], [t7, t7+T], and [t9, t9+T]). Based on the moving speed of the target object, it can be determined that the target object can reach the same target point d from each normal position information within the corresponding moving time interval, i.e., according to the position information of the target point, the target position information in which the high-confidence abnormal device should be located in the position normal state can be determined.

[0053] It should be further noted that the distance information between the high-confidence abnormal device in the position abnormal state and the target normal device associated therewith can be determined according to the straight-line distance between the abnormal position information and the normal position information, or can be determined according to the road network distance between the abnormal position information and the normal position information, which is not limited herein. The road network distance is the real distance that the target object needs to move according to the planned road in the scene. Because the moving road of the target object is not necessarily a straight line, the road network distance between two positions can be greater than the straight-line distance. Using the road network distance can more accurately represent the distance relationship between two positions, and can more accurately calculate the speed and moving time of the target object. The method for obtaining the road network distance can include but is not limited to being obtained according to preset map information, road network information, or being obtained by communicating with existing map software, navigation software, and the like, which is not described herein.

[0054] The above example process can also be equivalent to finding the target point d by analyzing the high-confidence normal devices (target normal devices C, E, F, and G) associated with the high-confidence abnormal device (device D). The target point d needs to satisfy the condition that the real time (T5, T6, T7, and T9) actually consumed by the target object to move from the point d (high-confidence abnormal device) to devices C, E, F, and G is less than the time difference threshold T (for example, the time difference between T5 and t5 needs to be less than T) of the reference moving time (t5, t6, t7, and t9) consumed by the target object to move between the two devices (for example, from the abnormal position information to the normal position information), and the position information of the target point d is the corrected position information (target position information) of the high-confidence abnormal device. Therefore, the same can be applied to the position correction of all high-confidence abnormal devices.

[0055] It should be noted that the method for obtaining the actual time consumed by the target object to move from the high-confidence abnormal device to devices C, E, F, and G respectively can be determined according to the collection time of the target object collected by the high-confidence abnormal device and the collection time of the target object collected by devices C, E, F, and G respectively. For example, if the target object moves from the high-confidence abnormal device to device C, the time when the target object is collected by device C (for example, 16:15 on x year y month z day) can be subtracted from the time when the target object is collected by the high-confidence abnormal device (for example, 16:00 on x year y month z day) to obtain the actual time T5 (15 minutes) consumed by the target object to move from the high-confidence abnormal device to device C. Alternatively, the trajectory information of the collected target object can also be used to assist in analysis, and the time consumed by the target object in each movement process can be calculated in sequence to improve the accuracy of time calculation.

[0056] Further, the target point determined by the foregoing example can be zero, one or more, and therefore the final target position information can be determined according to the number of target points and the position information of each target point. If there are zero target points, an abnormal prompt information can be selected and output, and the subsequent correction process can be paused or stopped; if there is one target point, the position information of the target point can be determined as the target position information; if there are multiple target points, the position information of a preferred target point can be selected from the multiple target points as the target position information, or the target position information can be calculated by integrating the position information of the multiple target points.

[0057] On the basis of the foregoing embodiment, the present embodiment describes the step of determining the target position information according to the number of target points and the position information of the target points. Specifically, the method of the present embodiment comprises the following steps:

[0058] In response to the number of target points being one, the position information of the target point is determined as the target position information; in response to the number of target points being multiple, the position information of the center point of the multiple target points is determined as the target position information.

[0059] In combination with the foregoing embodiment, after the target point is determined according to the foregoing example method, the target position information can be determined according to the number of target points and the position information of the target points.

[0060] For example, if the determined target point is one, the position information of the target point can be directly determined as the target position information. If the determined target point is multiple, the position information of the center point of the multiple target points is determined as the target position information. Subsequently, the method provided in the foregoing example can be repeated to mine and correct the high-confidence abnormal device in each front-end device until there is no high-confidence abnormal device in each front-end device or the high-confidence abnormal device cannot be corrected.

[0061] The method for determining the center point of the plurality of target points can be one or more, which is not limited here. For example, the geometric center of the plurality of target points can be determined as the center point, or the plurality of target points can be clustered by some implementable clustering algorithm, and the cluster center is determined as the center point, or the center point can be calculated by a weighted average method with a preset weight, which is not described here.

[0062] Therefore, in the single-center-point scenario, the target position information can be quickly determined; in the multi-center-point scenario, the interference of individual points can be effectively suppressed by center point calculation, the robustness of the positioning result is enhanced, and the actual position distribution of the device is more accurately reflected.

[0063] On the basis of the above-mentioned embodiments, the embodiments of the present application explain the steps of determining the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device from each front-end device according to the reference information. The reference information includes the collection information of the target object collected by the front-end device, and the collection information includes the trajectory information of the target object. Specifically, the method of the present embodiment includes the following steps:

[0064] According to the reference information and a preset scoring strategy, the corresponding front-end device is scored to obtain a target score; the front-end device with a target score less than a normal score threshold is determined as a high-confidence normal device; the front-end device with a target score greater than an abnormal score threshold is determined as a high-confidence abnormal device, and the normal score threshold is less than the abnormal score threshold; the high-confidence normal device associated with the high-confidence abnormal device is determined according to the trajectory information to obtain the target normal device.

[0065] In combination with the foregoing embodiments, in the present application, each front-end device can be scored by a preset scoring strategy to obtain the target score of each front-end device. Then, according to the comparison result between the target score of each front-end device and the score threshold, the high-confidence abnormal device and the high-confidence normal device in each front-end device are determined. The reference information of each front-end device can be used to select a preset scoring strategy to score the corresponding front-end device. The preset scoring strategy can include one or more, which is not limited here.

[0066] In the scenario of one preset scoring strategy, the scoring result of the preset scoring strategy is the target score of the front-end device. In the scenario of multiple preset scoring strategies, the sum of the scoring results of the multiple preset scoring strategies is required as the target score of the front-end device.

[0067] It should be noted that the smaller the target score of the front-end device is, the higher the probability of normality is, and the larger the target score of the front-end device is, the higher the probability of abnormality is.

[0068] In the implementation of the embodiment, two integral thresholds can be set, including a normal integral threshold and an abnormal integral threshold.

[0069] The normal integral threshold is used to determine whether the front-end device is a high-confidence normal device. For example, the front-end device with a target integral less than the normal integral threshold can be determined as a high-confidence normal device, and the front-end device with a target integral greater than or equal to the normal integral threshold can be a low-confidence normal device, a low-confidence abnormal device, and / or a high-confidence abnormal device.

[0070] The abnormal integral threshold is used to determine whether the front-end device is a high-confidence abnormal device, and the abnormal integral threshold is greater than the normal integral threshold. For example, the front-end device with a target integral greater than the abnormal integral threshold can be determined as a high-confidence abnormal device, and the front-end device with a target integral less than or equal to the abnormal integral threshold can be a low-confidence abnormal device, a low-confidence normal device, and / or a high-confidence normal device.

[0071] By screening the front-end devices through two integral thresholds of different sizes, higher-confidence normal devices and abnormal devices can be determined, and low-confidence normal devices and abnormal devices are screened out. The abnormal position information of the high-confidence abnormal device is corrected by using the high-confidence normal device, so that the position correction result is more accurate and reliable, and the problems of false correction and low correction accuracy when using low-confidence normal devices and abnormal devices for correction are avoided.

[0072] Of course, another implementation manner of the present application can also be to divide each front-end device into a normal device and an abnormal device by using one integral threshold, and then correct the abnormal position information of the currently analyzed abnormal device by using the normal position information of the normal device associated with the currently analyzed abnormal device according to the correction method provided by the present application. Details are not described herein. In addition, in addition to using one integral threshold to perform binary classification processing when determining normal devices and abnormal devices, in other embodiments of the present application, if a threshold segmentation process is involved, a corresponding threshold can also be used for binary classification processing. Details are not described herein.

[0073] It should be further noted that the determination of the device association relationship in the present application can include but is not limited to determining the association relationship of each front-end device according to the obtained pre-set association relationship and / or the trajectory information of the target object (for example, each front-end device through which the target object passes during movement can be associated according to the trajectory information of the target object). That is, when determining the high-confidence normal device associated with the high-confidence abnormal device (or determining the normal device associated with the abnormal device), the determination can be based on the trajectory information.

[0074] On the basis of the above-mentioned embodiments, the step of scoring the corresponding front-end device according to the reference information and the preset scoring strategy to obtain the target score is described in the embodiments of the present application. The preset scoring strategy has multiple. Specifically, the method of the present embodiment comprises the following steps:

[0075] According to the reference information of the currently analyzed front-end device and multiple preset scoring strategies, the currently analyzed front-end device is scored to obtain the initial score corresponding to each preset scoring strategy. The initial scores are summed to obtain the target score of the currently analyzed front-end device.

[0076] In combination with the foregoing embodiments, the scoring strategy used in the present application for scoring the front-end device can have one or more.

[0077] If the preset scoring strategy includes multiple, the reference information of the currently analyzed front-end device needs to be scored according to multiple preset scoring strategies to obtain the scoring result (initial score) corresponding to each preset scoring strategy, and then the initial scores are summed (including cumulative summation or weighted summation) to obtain the target score of the currently analyzed front-end device. If the corresponding reference information is missing when a certain preset scoring strategy is executed, the preset scoring strategy is skipped, and / or the initial score corresponding to the preset scoring strategy is 0 or other preset value, which is not limited here.

[0078] Therefore, the accuracy of device latitude and longitude anomaly detection can be improved by multi-strategy scoring fusion, false judgments caused by single indicators can be avoided, and the possibility of device location anomaly can be more comprehensively reflected.

[0079] On the basis of the above-mentioned embodiments, the step of scoring the corresponding front-end device according to the reference information and the preset scoring strategy to obtain the target score is described in the embodiments of the present application. The reference information further includes device information of the front-end device and region information of a region of interest in a preset region, the device information includes device location information and device name, and the region information includes region location information and region name. Specifically, the method of the present embodiment comprises the following steps:

[0080] The target region of interest pre-bound with the currently analyzed front-end device in each region of interest is found. In response to the currently analyzed front-end device being in the target region of interest, the name similarity of the currently analyzed front-end device is determined according to the device name of the currently analyzed front-end device and the region name of the corresponding target region of interest. The target score of each front-end device is obtained by scoring processing according to the name similarity and the similarity threshold.

[0081] In combination with the foregoing embodiments, one of the preset scoring strategies of the present application can be to perform scoring processing according to the name similarity between the device name of the front-end device and the area name of the area in which the front-end device is located.

[0082] It should be noted that there can be one or more areas of interest (AOI) in a preset area. The front-end device and the AOI area in which the front-end device is located can be bound in advance, and the relevant device name of the front-end device (for example, "A Square" and "A Square front-end device") can also be selectively set according to the area name of the area in which the front-end device is located. Thus, in the case where the front-end device is normally positioned, the position information between the device with the similar or same name and the AOI is close.

[0083] Exemplarily, in the specific process of performing the present embodiment, the area of interest to which the currently analyzed front-end device is pre-bound can be found in each area of interest, to obtain a target area of interest. Then, mutual position judgment can be performed according to the position information of the currently analyzed front-end device and the position information of the target AOI. If the currently analyzed front-end device is in the target AOI, or the currently analyzed front-end device is outside the target AOI and the shortest distance between the currently analyzed front-end device and the target AOI is less than a preset distance threshold , then the name similarity between the device name of the currently analyzed front-end device and the area name of the corresponding target AOI is determined. The method of determining the name similarity can include but is not limited to calculating the similarity between the device name and the area name by using OLP text algorithm. For example, the device name and the area name can be segmented by using a segmentation algorithm to calculate the corresponding name vectors, and then the vector similarity is calculated as the name similarity between the device name and the area name.

[0084] Further, after obtaining the name similarity between the currently analyzed front-end device and the corresponding AOI, scoring processing can be performed according to the name similarity and a similarity threshold, to obtain the target score (or as the initial score) of the front-end device. For example, if the name similarity is greater than the similarity threshold , the score of the front-end device is . If the name similarity is less than the similarity threshold , the score of the front-end device is , wherein , . That is, in the case where the positions are close, whether the front-end device is positionally abnormal is determined by whether the name similarity is similar.

[0085] Another example, if the current analysis of the front-end device is outside the target AOI and the shortest distance between the current analysis of the front-end device and the target AOI is greater than the preset distance threshold , the name similarity can be calculated according to the foregoing example method. If the name similarity is greater than the similarity threshold , the score of the device is , wherein , . That is, in the case of a large distance between positions, whether the front-end device is abnormal is determined by whether the name similarity is similar. If the name similarity is less than or equal to the similarity threshold , it may be a normal case of a large distance between positions. In this case, there is no high confidence to determine whether the position information of the device is normal or abnormal, so the scoring can not be based on this strategy, or it can be scored as 0 or other scores, etc., which are not limited here.

[0086] From the above, the present application effectively solves the positioning deviation problem caused by the lack of device name or insufficient standardization in the prior art through double verification of position information and name information. When the current position reflected by the device contradicts the area associated with the device, name similarity calculation can timely discover abnormalities, such as the case where the device name is "XX Mall" but the position information deviates from the AOI range of XX Mall. This multi-dimensional scoring mechanism improves the accuracy of abnormal device identification, and can also distinguish high-confidence normal devices and abnormal devices through the integral accumulation strategy, providing a reliable basis for subsequent position correction. Combined with dynamic verification of position information, it avoids the misjudgment that may be caused by simply relying on text similarity, especially in the case where the device name and the area name are similar or the same but the geographical positions of the two are quite different, which can more accurately identify abnormal devices.

[0087] On the basis of the above embodiments, the present embodiment describes the step of scoring the corresponding front-end device according to the reference information and the preset scoring strategy to obtain the target score. The collected information also includes the collection time. Specifically, the present embodiment method includes the following steps:

[0088] The collection time is used to sort the front-end devices to obtain the sorted front-end devices. The moving speed of the target object between two adjacent front-end devices in the sequence is obtained. The two front-end devices corresponding to the moving speed greater than the preset speed threshold are determined as candidate abnormal devices and the abnormal judgment times are recorded. The associated front-end device corresponding to the currently analyzed candidate abnormal device is determined according to the trajectory information. The target score of the currently analyzed candidate abnormal device is obtained by scoring according to the abnormal judgment times of the currently analyzed candidate abnormal device and the abnormal judgment times of the associated front-end device.

[0089] In combination with the foregoing embodiments, one of the preset scoring strategies of the application can be to score according to the movement of the target object between the front-end devices.

[0090] It can be understood that after the front-end device collects the collection information of the target object (for example, image collection of the target object), the collection time of each front-end device at the time of collection can be obtained, and the collection information of the same target object collected by multiple front-end devices can also be integrated to determine the movement trajectory (trajectory information) of the target object. The specific method can refer to related technologies in the art, and will not be repeated here. It should be noted that in the specific application scenarios of the application, the target object can include one or more, and the same front-end device and different front-end devices can collect the same or different target objects, which is not limited here. For ease of understanding and description, the subsequent examples of the application will mainly be described in the scenario where multiple front-end devices collect the same target object.

[0091] For example, all collection information of each target object in the preset area can be obtained. For a certain target object, the front-end devices that collect the target object can be sorted according to the collection time corresponding to the front-end devices that collect the target object, to obtain the sorted front-end devices. It can be understood that the order between the sorted front-end devices also represents the moving order (moving trajectory) of the target object to some extent.

[0092] Then the moving speed of the target object between the two front-end devices in sequence can be determined according to the collection information collected by the sorted front-end devices. The two front-end devices in sequence refer to the two front-end devices that continuously collect the target object. For example, after sorting the front-end devices that collect the target object, the sorted front-end devices are A (collection time is 12:00-12:10), B (collection time is 12:12-12:20), and C (collection time is 12:20-12:25). Therefore, A and B are two front-end devices in sequence, and B and C are two front-end devices in sequence.

[0093] Specifically, the moving speed of the target object between the two front-end devices in sequence can be determined according to the road network distance between the two front-end devices in sequence and the collection time difference between the two front-end devices in sequence for the target object. That is, the speed consumed by the same target object when passing through the two front-end devices deployed in sequence.

[0094] Then the moving speed of the target object can be compared with the preset speed threshold If the moving speed of the target object between the two front-end devices in sequence is greater than the preset speed threshold If these two front-end devices are identified as candidate abnormal devices, the number of abnormal judgments for each marked candidate abnormal device will be recorded. The preset speed threshold can be pre-defined or flexibly adjusted according to specific application scenarios. There can be one or more preset speed thresholds, which is not limited here. For example, different preset speed thresholds can be used for different types of target objects, different preset speed thresholds can be used for different road conditions (e.g., asphalt roads, dirt roads, cement roads), and different preset speed thresholds can be used for different traffic conditions (e.g., congestion, slow traffic, smooth traffic). Alternatively, multiple standards can be combined for analysis (e.g., combining information such as target object type, road conditions, and traffic conditions) and then using the corresponding preset speed thresholds for judgment, etc., which will not be elaborated here.

[0095] Furthermore, by analyzing the trajectory information of the target object, the associated front-end devices (i.e., associated front-end devices) of the currently analyzed candidate abnormal devices can be determined. It should be noted that the one or more devices associated with a candidate abnormal device can include devices with an initial abnormality score of zero and / or abnormal devices with an initial abnormality score of non-zero. Therefore, after obtaining the number of abnormality judgments for each candidate abnormal device, the number of abnormality judgments for the currently analyzed candidate abnormal device and the number of abnormality judgments for the associated front-end devices can be determined. Then, a score can be calculated based on the number of abnormality judgments for the currently analyzed candidate abnormal device and the number of abnormality judgments for the associated front-end devices to obtain the target score for the currently analyzed candidate abnormal device.

[0096] For example, the trajectory information of a target object can be analyzed. If the target object moves at a speed greater than a preset speed threshold three times... In this case, the candidate abnormal device pairs determined according to each abnormal movement speed are (A, B), (A, C), and (A, B), that is, the total number of candidate abnormal devices are (A, B, C). The number of abnormal judgments for each candidate abnormal device corresponds to... (That is, device A is identified as a candidate abnormal device 3 times, device B is identified as a candidate abnormal device 2 times, and device C is identified as a candidate abnormal device 1 time), and the number of other candidate abnormal devices associated with each candidate abnormal device is as follows: (That is, there are two candidate abnormal devices B and C among the front-end devices associated with device A, one candidate abnormal device A among the front-end devices associated with device B, and one candidate abnormal device A among the front-end devices associated with device C.)

[0097] Then, the abnormality is judged in combination with the relevant information obtained above. If the number of times of abnormality judgment corresponding to the candidate abnormal device currently analyzed and the number of times of abnormality judgment of the candidate abnormal device associated with the candidate abnormal device currently analyzed are both greater than a preset number of times of abnormality threshold, the score of the candidate abnormal device currently analyzed is If only one of the number of times of abnormality judgment corresponding to the candidate abnormal device currently analyzed and the number of times of abnormality judgment of the candidate abnormal device associated with the candidate abnormal device currently analyzed is greater than the number of times of abnormality threshold, the score of the candidate abnormal device currently analyzed is If both the number of times of abnormality judgment corresponding to the candidate abnormal device currently analyzed and the number of times of abnormality judgment of the candidate abnormal device associated with the candidate abnormal device currently analyzed are less than the number of times of abnormality threshold and greater than 0, the score of the candidate abnormal device currently analyzed is If the moving speed of the target object between two front-end devices adjacent in sequence is less than or equal to a preset speed threshold , the two front-end devices adjacent in sequence are not candidate abnormal devices (which can be marked as candidate normal devices). If the candidate normal device currently analyzed and each front-end device associated with the candidate normal device are all candidate normal devices, the score of the candidate normal device currently analyzed is , wherein .

[0098] On the basis of the above embodiment, it needs to be explained by the embodiment of the application that the trajectory information of a certain target object is not necessarily continuous during the movement of the target object. For example, the target object can stay at a certain place for a long time, which will affect the analysis and judgment process in the examples of the application to some extent.

[0099] Therefore, the application can also provide a method for splitting the trajectory information of a target object to obtain a plurality of sub-trajectory information, and then performing the analysis and judgment of the above examples based on the sub-trajectory information.

[0100] For example, if the moving speed of the same target object between two front-end devices adjacent in sequence is much less than a normal speed threshold (the relationship of much less can be judged by setting a difference threshold, which is not described here), and the difference between the two collection times of the two front-end devices adjacent in sequence for the target object is greater than a time difference threshold , it can be determined that the target object stays between the two front-end devices, that is, the trajectory information determined according to the collection information of the target object collected by the two front-end devices does not have continuity, and the trajectory can be split.

[0101] Specifically, the trajectory information of the target object O determined according to the collection information of each front-end device is , wherein c represents a front-end device, and t represents the collection time of the front-end device for the target object. If and and , , then it can be determined that is continuous collection, is not continuous collection, and thus trajectory segmentation can be performed thereon, is continuous collection. Therefore, in this example, the trajectory of the target object O after trajectory segmentation is continuous, and the sub-trajectories are and .

[0102] Therefore, the trajectory information used in the examples of the present application can also refer to the sub-trajectory information after trajectory segmentation, which will not be described in detail.

[0103] On the basis of the above-mentioned embodiments, the step of obtaining the target score by scoring the corresponding front-end device according to the reference information and the preset scoring strategy is described in the embodiments of the present application. The collection information includes the trajectory information of the target object. Specifically, the method of the present embodiment includes the following steps:

[0104] The adjacent device of the front-end device currently analyzed is determined according to the trajectory information; and the target score of the front-end device currently analyzed is obtained by scoring according to the distance information between the front-end device currently analyzed and the adjacent device.

[0105] In combination with the foregoing embodiments, the adjacent device refers to the front-end device that can be reached by the target object in a short time from the front-end device currently analyzed.

[0106] Exemplarily, the associated device that can be reached by the target object from the front-end device currently analyzed can be determined according to the trajectory information, and then the adjacent device can be determined according to the comparison result of the real time consumed by the target object between the front-end device currently analyzed and the associated device (according to the difference between the collection times of the two devices) and the preset time threshold, i.e., if the real time consumed by the target object between the two devices is less than the preset time threshold, then the two devices are adjacent devices.

[0107] Further, the distance information (for example, the road network distance) between the adjacent devices is analyzed, and if the road network distance is greater than a road network distance threshold, then the set of adjacent devices is a candidate abnormal device. If the number of other different candidate abnormal devices associated with a certain candidate abnormal device is greater than a preset candidate device number threshold, then the score of the candidate abnormal device is . If it is less than or equal to the candidate device number threshold, then the score of the candidate abnormal device is . If the road network distance is less than or equal to the road network distance threshold, then the set of adjacent devices is not a candidate abnormal device, and the scores of the two adjacent devices are , wherein .

[0108] On the basis of the above-mentioned embodiments, the embodiments of the present application describe the step of scoring the corresponding front-end device according to the reference information and the preset scoring strategy to obtain the target score. The collected information further includes the moving speed. Specifically, the method of the embodiments includes the following steps:

[0109] According to the trajectory information, the transition probability between each front-end device is determined. The candidate front-end devices with transition probability greater than the transition probability threshold are determined. According to the preset abnormal data analysis method and the moving speed of the target object between each candidate front-end device, the candidate abnormal devices associated with each candidate front-end device are determined. The target score of the candidate front-end device is obtained by scoring according to the number of candidate abnormal devices associated with the candidate front-end device currently analyzed.

[0110] In combination with the foregoing embodiments, after obtaining the trajectory information (sub-trajectory information) of the target object, the transition probability of the target object between two front-end devices can be analyzed.

[0111] Exemplarily, there is a target object O, and one of the continuous trajectories after trajectory segmentation is The number of device transitions of the target object between devices is analyzed for the continuous trajectory of the target object O, for example, the next device of the target object O after passing through is , the number of times of moving from to the next device is , the number of times of moving from to the next device is

[0112] , and the number of device transitions of the target object between front-end devices is analyzed in turn, and the total number of device transitions is obtained by accumulating the number of device transitions of the target object between each front-end device. Then, the transition probability between each front-end device can be determined according to the number of device transitions of each front-end device and the total number of device transitions. For example, when , the probability of the next device of the target object being after passing through is . The transition probability between each front-end device can be obtained by generalizing to each front-end device passed by the target object.

[0113] It should be noted that if the target object moves between two devices multiple times in a period of time, the time sequence corresponding to the two devices can be generated, and then the median (or average, or other numerical value determined from the time sequence, etc.) of the time sequence can be taken as the moving time of the target object between the two devices. The moving time determined in this way is more representative, and the moving time in the foregoing example can be obtained in the same way, which will not be described here. For example, the trajectory information of the target object O in 10 days is , the moving time of the target object O from in the 10 days can be counted to obtain a time sequence. For another example, after , the next device is , and the M times of time are , the median of the time sequence is , and the time consumed after , the next device is . Similarly, the probability of each device to the previous device and the corresponding device transfer time can also be obtained. Thus, the device transfer speed of the target object, that is, the moving speed of the target object, can be calculated based on the distance information (such as road network distance) between the two devices. The moving speed in the foregoing example can be obtained in the same way, which will not be described here.

[0114] Further, the device association relationship unidirectional topological graph can also be constructed according to the association relationship between the front-end devices. After obtaining the transfer probabilities between the front-end devices, the device association relationships with transfer probabilities less than a transfer probability threshold value can be filtered, and candidate front-end devices with transfer probabilities greater than the transfer probability threshold value can be determined. The weights of the lines between the devices in the topological graph are the corresponding transfer probabilities.

[0115] Then, the candidate abnormal devices associated with each candidate front-end device can be determined according to a preset abnormal data analysis method and the moving speed of the target object between the candidate front-end devices. The abnormal data analysis method can include but is not limited to the box plot method, the isolation forest method, etc. The moving speed of the target object between the devices is analyzed by using the abnormal data analysis method to determine the abnormal transfer speed. The two devices corresponding to the abnormal transfer speed are determined as the candidate abnormal devices. If the number of other candidate abnormal devices associated with a certain candidate abnormal device is greater than a preset threshold value, the score of the candidate abnormal device is ; if it is less than or equal to the preset threshold value, the score of the candidate abnormal device is . For the front-end device that is not a candidate abnormal device, the score can be , where .

[0116] Based on the foregoing embodiments, this application can select one or more of the aforementioned preset scoring strategies for scoring processing. If multiple preset scoring strategies are used, the scores obtained by each front-end device after scoring processing by each preset scoring strategy need to be accumulated or weighted summed to obtain the final target score.

[0117] Furthermore, if the target score of the front-end device is less than the first threshold If the target integral of the device exceeds the normal integration threshold, the device is considered a high-confidence normal device. If the target integral of the device is greater than the second threshold... (Abnormal integration threshold), where If the condition is not met, the device is considered a high-confidence anomalous device. Subsequently, positional anomaly correction can be performed on the high-confidence normal device.

[0118] It should also be noted that, based on the unidirectional topology graph of device association obtained in the aforementioned embodiments, when there is a bidirectional connection between two front-end devices, the average of the travel times corresponding to the two connections can be taken as the time required for the target object to pass through these two devices, thus forming a bidirectional topology graph of device association. For details, please refer to... Figure 2 As shown, further details will not be elaborated here.

[0119] exist Figure 2 In the correction process shown, when calculating the actual time taken from point d to devices C, E, F, and G, the moving speed can be represented by methods such as determining the median speed in the aforementioned embodiments, and the actual time can be obtained by using the quotient of the road network distance and the median speed. This will not be elaborated here.

[0120] Based on the above embodiments, this application may also provide a user interface for correcting the position anomaly of a front-end device.

[0121] For example, refer to Figure 3 As shown, Figure 3is an exemplary interactive interface diagram in the device position anomaly correction method of the present application. In the diagram, relevant information of the front-end device can be displayed in a map, the left side of the interface can display a corresponding device list and the type of target object collected by the front-end device, and filtering display of different types of collected target objects can be supported. At the same time, devices with position anomalies can also be directly filtered, and a prompt mark can be displayed at the position anomaly device, which can include but is not limited to a red exclamation mark, etc. Clicking on the corresponding device in the device list, the map displays the position of the device (which can be represented by blue), and displays several other devices with strong association with the device. Among them, high-confidence normal devices can be green devices, high-confidence abnormal devices can be red devices, and the AOI information in which the device is located can also be displayed. Clicking on the AOI area displays the AOI name. In addition, the position of the device after position correction according to the method of the present application (target position information) can also be inferred through the light blue device identifier.

[0122] Among them, the user interactive interface of the present application can also support the user to issue control instructions, such as selecting the front-end device by clicking the instruction and manually modifying the position information or other related information, etc. The user can also click on the position of the corrected device to update the high-confidence abnormal device. For example, the high-confidence abnormal device is associated with high-confidence normal devices that are far away from the device, so the device is likely to have a position anomaly, and the device may be near the associated high-confidence normal devices. After the device position is corrected, the corresponding AOI can be automatically bound, and clicking can also display the AOI name after binding. The interface platform can also automatically determine whether the position is normal based on the updated position information in the background, including but not limited to the method mentioned in the foregoing examples, based on the updated position information. If normal, the corrected longitude and latitude are displayed subsequently, and if abnormal, a prompt message can be output.

[0123] In summary, the device position anomaly correction method of the present application can correct the position based on the position information of the position normal devices associated with the position anomaly device. This method uses the time information between two devices to calculate the actual distance between the devices, which is accurate and effective. Among them, a variety of preset scoring strategies can be used to determine the position anomaly device and the position normal device. For example, the device position anomaly judgment method based on the text similarity between the device and the AOI, the spatio-temporal anomaly between the devices, the transfer speed anomaly, etc. These strategies combine a variety of spatio-temporal algorithms to mine longitude and latitude anomalies and normal devices in multiple dimensions, with high algorithm accuracy. In addition, a user interactive interface is also provided, which allows the user to intuitively feel the change of the device position, and the position that does not need to be corrected can not be corrected.

[0124] It should be further explained that the execution subject of the device position anomaly correction method can be a device position anomaly correction apparatus, for example, the device position anomaly correction method can be executed by a terminal device or a server or other processing device, wherein the terminal device can be a user equipment (UE), a computer, a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a PDA, a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementation manners, the device position anomaly correction method can be implemented by a processor invoking computer readable instructions stored in a memory.

[0125] Figure 4 is a block diagram of a device position anomaly correction apparatus according to an example embodiment of the present application. As shown in the figure, the example device position anomaly correction apparatus 400 includes an information acquisition module 410, a device determination module 420 and a position correction module 430. Specifically: Figure 4

[0126] The information acquisition module 410 is configured to acquire reference information corresponding to each front-end device in a preset area, wherein the reference information includes device information of the front-end device and acquisition information obtained by the front-end device from a target object, or the reference information includes device information of the front-end device, acquisition information corresponding to the front-end device and region information of a region of interest in the preset area.

[0127] The device determination module 420 is configured to determine a high-confidence abnormal device and a target normal device associated with the high-confidence abnormal device from each front-end device according to the reference information.

[0128] The position correction module 430 is configured to perform position correction processing on the corresponding high-confidence abnormal device according to the target normal device, to obtain target position information of the high-confidence abnormal device.

[0129] ​In the exemplary device position anomaly correction apparatus, the reference information corresponding to each front-end device in a preset area is obtained for multi-dimensional analysis, so that a high-confidence abnormal device and a target normal device associated with the high-confidence abnormal device can be determined from the front-end devices according to the reference information. The high-confidence abnormal device is a front-end device that is more likely to have a position anomaly, and the target normal device is a front-end device that is associated with the high-confidence abnormal device and has a normal position. Therefore, the corresponding high-confidence abnormal device can be corrected in position according to the target normal device, the target position information of the high-confidence abnormal device is obtained, and the front-end device with a position anomaly is corrected in position. Compared with the traditional position correction method, in which a standard address needs to be set in advance in the front-end device, and the position is matched according to the standard address and the latitude and longitude information, in most cases, the front-end device may not set the standard address and the latitude and longitude matching effect is poor, which is difficult to effectively apply. The method of the present application improves the accuracy and efficiency of position anomaly correction.

[0130] It should be noted that the apparatus provided in the above embodiments and the method provided in the above embodiments belong to the same concept, wherein the specific manner in which each module and unit performs operations has been described in detail in the method embodiments, which will not be repeated here. The apparatus provided in the above embodiments can allocate the above functions to different functional modules according to the needs, that is, the internal structure of the apparatus is divided into different functional modules to complete all or part of the functions described above, which is not limited here.

[0131] The functions of each module can be referred to the device position anomaly correction method embodiments, which will not be repeated here.

[0132] Please refer to Figure 5 , Figure 5 is a structural schematic diagram of an embodiment of an electronic device of the present application. The electronic device 100 includes a memory 101 and a processor 102, and the processor 102 is configured to execute program instructions stored in the memory 101 to implement the steps in any of the above device position anomaly correction method embodiments. In one specific implementation scenario, the electronic device 100 can include but is not limited to a microcomputer, a server, in addition, the electronic device 100 can also include a notebook computer, a tablet computer and other mobile devices, which are not limited here.

[0133] In particular, the processor 102 is configured to control itself and the memory 101 to implement the steps in any of the above-mentioned embodiments of the method for correcting device position anomaly. The processor 102 can also be referred to as a CPU (Central Processing Unit). The processor 102 can be an integrated circuit chip having a processing capability of signals. The processor 102 can also be a general-purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. In addition, the processor 102 can be implemented by an integrated circuit chip together.

[0134] In the example electronic device, the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device are determined from the front-end devices according to the reference information by performing multi-dimensional analysis on the reference information corresponding to the front-end devices in the preset area. The high-confidence abnormal device is a front-end device that is more likely to have a position anomaly, and the target normal device is a front-end device that is associated with the high-confidence abnormal device and has a normal position. Therefore, the position of the corresponding high-confidence abnormal device can be corrected according to the target normal device, and the target position information of the high-confidence abnormal device is obtained, so that the front-end device with a position anomaly is corrected. Compared with the traditional position correction method in which a standard address needs to be set in the front-end device and then the position is matched according to the standard address and the latitude and longitude information, in most cases, the front-end device may not set the standard address and the latitude and longitude matching effect is poor, which is difficult to effectively apply. The method of the present application improves the accuracy and efficiency of position anomaly correction.

[0135] Please refer to Figure 6 , Figure 6 is a structural schematic diagram of an embodiment of the computer readable storage medium of the present application. The computer readable storage medium 110 stores program instructions 111 capable of being executed by a processor, and the program instructions 111 are used to implement the steps in any of the above-mentioned embodiments of the method for correcting device position anomaly.

[0136] In the exemplary storage medium, by running the program instructions in the storage medium, the reference information corresponding to each front-end device in the preset area is acquired for multi-dimensional analysis, so that the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device can be determined from each front-end device according to the reference information. The high-confidence abnormal device is a front-end device that is more likely to have a position abnormality, and the target normal device is a front-end device that is associated with the high-confidence abnormal device and has a normal position. Therefore, the corresponding high-confidence abnormal device can be corrected in position according to the target normal device, the target position information of the high-confidence abnormal device is obtained, and the front-end device with a position abnormality is corrected in position. Compared with the traditional position correction method in which a standard address needs to be set in advance in the front-end device, and then the position is matched according to the standard address and the latitude and longitude information, in most cases, the front-end device may not set the standard address and the latitude and longitude matching effect is poor, and it is difficult to effectively apply. The method provided in the application improves the accuracy and efficiency of position abnormality correction.

[0137] In some embodiments, the apparatus provided by the embodiments of the present disclosure has functions or includes modules that can be used to perform the methods described in the above method embodiments, and the specific implementation can refer to the description of the above method embodiments. For brevity, it will not be repeated here.

[0138] The above description of each embodiment tends to emphasize the differences between each embodiment, and the same or similar parts can be mutually referred to. For brevity, it will not be repeated here.

[0139] In several embodiments provided in the present application, it should be understood that the disclosed method and device can be implemented by other means. For example, the above-described device implementation is only schematic, for example, the division of modules or units is only a logical function division, and actual implementation can have another division manner, for example, a unit or component can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0140] In addition, each of the function units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit. When the integrated unit is realized in the form of a software function 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 solutions of the present application, essentially or in the form of a contribution to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute all or part of the steps of the methods in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

Claims

1. A method of correcting an abnormality of a device position, characterized by, The method is applied to a device management end for managing front-end devices, and comprises: acquiring reference information corresponding to each front-end device in a preset area; the reference information comprises collection information obtained by the front-end device collecting a target object, and the collection information comprises trajectory information of the target object; determining a high-confidence abnormal device and a target normal device associated with the high-confidence abnormal device from the front-end devices according to the reference information; the step of determining the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device from the front-end devices according to the reference information comprises: performing score processing on the corresponding front-end device according to the reference information and a preset score strategy to obtain a target score; determining a high-confidence abnormal device and a high-confidence normal device in each front-end device according to a comparison result between the target score and a score threshold; and determining the high-confidence normal device associated with the high-confidence abnormal device according to the trajectory information to obtain the target normal device; performing position correction processing on the corresponding high-confidence abnormal device according to the target normal device to obtain target position information of the high-confidence abnormal device.

2. The method of claim 1, wherein, The reference information comprises collection information obtained by the front-end device collecting a target object, and the collection information comprises a moving speed of the target object, and the step of performing position correction processing on the corresponding high-confidence abnormal device according to the target normal device to obtain target position information of the high-confidence abnormal device comprises: acquiring abnormal position information of a currently analyzed high-confidence abnormal device and normal position information of a target normal device associated with the currently analyzed high-confidence abnormal device; determining a reference moving time of the target object moving from the currently analyzed high-confidence abnormal device to the target normal device at the moving speed according to the moving speed, the normal position information, and the abnormal position information; determining a moving time interval of the target object according to the reference moving time and a preset time difference threshold; determining a target point reached by the target object in the moving time interval according to the moving speed and the normal position information; determining the target position information according to a number of the target points and position information of the target points.

3. The method of claim 2, wherein, The step of determining the target position information according to the number of the target points and the position information of the target points comprises: in response to the number of the target points being one, determining the position information of the target point as the target position information; in response to the number of the target points being multiple, determining position information of a center point of the multiple target points as the target position information.

4. The method of claim 1, wherein, The step of determining the high-confidence abnormal device and the target normal device associated with the high-confidence abnormal device from the front-end devices according to the reference information comprises: performing score processing on the corresponding front-end device according to the reference information and a preset score strategy to obtain the target score; determining a front-end device with a target score less than a normal score threshold as a high-confidence normal device; and determine the front-end device with the target score greater than an abnormal score threshold as the high-confidence abnormal device, the normal score threshold being less than the abnormal score threshold; determine a high-confidence normal device associated with the high-confidence abnormal device according to the trajectory information, to obtain the target normal device.

5. The method of claim 4, wherein, The preset scoring strategy includes a plurality of scoring strategies, and the scoring processing of the corresponding front-end device according to the reference information and the preset scoring strategy includes: perform scoring processing on the currently analyzed front-end device according to the reference information of the currently analyzed front-end device and a plurality of preset scoring strategies, to obtain initial scores corresponding to the preset scoring strategies; perform sum processing on the initial scores, to obtain a target score of the currently analyzed front-end device.

6. The method of claim 4, wherein, The reference information further includes device information of the front-end device and region information of a region of interest in the preset region, the device information includes device position information and a device name, and the region information includes region position information and a region name, and the scoring processing of the corresponding front-end device according to the reference information and the preset scoring strategy includes: find a target region of interest in each region of interest to which the currently analyzed front-end device is pre-bound; in response to the currently analyzed front-end device being in the target region of interest, determine a name similarity of the currently analyzed front-end device according to a device name of the currently analyzed front-end device and a region name of the corresponding target region of interest; perform scoring processing according to the name similarity and a similarity threshold, to obtain the target score of each front-end device.

7. The method of claim 4, wherein, The collection information further includes a collection time, and the scoring processing of the corresponding front-end device according to the reference information and the preset scoring strategy includes: perform sorting processing on each front-end device according to the collection time, to obtain sorted front-end devices; obtain a moving speed of the target object between two front-end devices adjacent in sequence; determine two front-end devices corresponding to a moving speed greater than a preset speed threshold as candidate abnormal devices and record an abnormal judgment number; determine an associated front-end device corresponding to the currently analyzed candidate abnormal device according to the trajectory information; perform scoring according to the abnormal judgment number of the currently analyzed candidate abnormal device and the abnormal judgment number of the associated front-end device, to obtain a target score of the currently analyzed candidate abnormal device.

8. The method of claim 4, wherein, The scoring processing of the corresponding front-end device according to the reference information and the preset scoring strategy includes: determine a neighboring device of the currently analyzed front-end device according to the trajectory information; perform scoring according to distance information between the currently analyzed front-end device and the neighboring device, to obtain a target score of the currently analyzed front-end device.

9. The method of claim 4, wherein, The collection information further includes a moving speed, and the scoring processing of the corresponding front-end device according to the reference information and the preset scoring strategy includes: determine a transfer probability between each front-end device according to the trajectory information; determine a candidate front-end device in which the transfer probability is greater than a transfer probability threshold; According to the preset abnormal data analysis method and the moving speed of the target object between the candidate front-end devices, candidate abnormal devices associated with each candidate front-end device are determined; According to the number of candidate abnormal devices associated with the currently analyzed candidate front-end device, a target score of the candidate front-end device is obtained.

10. A computer-readable storage medium having stored thereon program instructions, wherein, The program instructions, when executed by the processor, implement the method of any one of claims 1 to 9.

Images