An intelligent well lid positioning tracking method, device and system

Abstract

The application relates to the technical field of intelligent well lid data monitoring, and provides an intelligent well lid positioning and tracking method, device and system, which comprises the following steps: acquiring well lid alarm information in a preset time window, determining alarm correlation based on the well lid alarm information; in response to the alarm correlation being well lid alarm correlation, determining a predicted driving path based on the well lid alarm information; determining a monitoring node and a monitoring time window based on the predicted driving path; acquiring video data collected by the monitoring node in the monitoring time window, and performing well lid positioning and tracking based on the video data. The application can position and track the well lid when the positioning signal is shielded by predicting a track and determining a monitoring range based on multi-point alarm correlation, so that the tracking efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of intelligent manhole cover data monitoring technology, specifically to an intelligent manhole cover positioning and tracking method, device and system. Background Technology

[0002] Currently, anti-theft tracking of smart manhole covers generally relies on built-in GPS or BeiDou satellite positioning modules. When a smart manhole cover experiences abnormal displacement, the system receives positioning signals and reports its geographical location in real time, allowing managers to promptly recover the stolen facility. However, criminals may carry portable GPS signal jammers for theft. These devices can emit co-channel interference signals, causing the smart manhole cover's built-in positioning module to lose its satellite signal reception capability for a short period, resulting in the system being unable to obtain any valid location data and the tracking link being interrupted. Therefore, how to effectively track stolen manhole covers without relying on GPS signals is a pressing technical challenge that needs to be solved. Summary of the Invention

[0003] This application provides a method, device, and system for locating and tracking intelligent manhole covers, which can locate and track manhole covers even when GPS signals are unavailable.

[0004] This application provides a smart manhole cover positioning and tracking method, including:

[0005] Obtain manhole cover alarm information within a preset time window, the manhole cover alarm information including manhole cover identification, alarm time and location coordinates;

[0006] Based on the manhole cover alarm information, the alarm correlation is determined;

[0007] In response to alarm correlation, a manhole cover alarm is associated, and a predicted driving path is determined based on the manhole cover alarm information.

[0008] Based on the predicted driving path, the monitoring nodes and monitoring time windows are determined;

[0009] The monitoring node acquires video data collected during the monitoring time window and performs manhole cover location tracking based on the video data.

[0010] Optionally, based on the manhole cover alarm information, the alarm correlation is determined, including:

[0011] Determine the straight-line distance and actual road distance between two smart manhole covers with adjacent alarm times based on location coordinates;

[0012] The primary correlation is determined based on the number of manhole cover alarm messages;

[0013] The second correlation is determined based on all straight-line distances and all actual road distances;

[0014] Alarm correlation is determined based on the first and second correlations.

[0015] Optionally, the first correlation includes quantity correlation or no quantity correlation, and the second correlation includes distance correlation or no distance correlation. The determination of alarm correlation based on the first and second correlations includes:

[0016] Given that the first correlation is a quantity correlation and the second correlation is a distance correlation, the alarm correlation is determined to be a manhole cover alarm correlation;

[0017] If the first correlation is quantity-independent and / or the second correlation is distance-independent, the alarm correlation is determined to be manhole cover alarm-independent.

[0018] Optionally, the response to the alarm correlation is a manhole cover alarm correlation, and the determination of the predicted driving path based on the manhole cover alarm information includes:

[0019] Starting from the location coordinates of the target manhole cover, multiple candidate paths are obtained by extending a preset distance in the road network; the target manhole cover is the last manhole cover to trigger an alarm.

[0020] Based on each candidate path, determine the path distance, number of cameras, and number of manhole covers for each candidate path;

[0021] The total cost of each candidate path is determined based on the path distance, number of cameras, and number of manhole covers.

[0022] The predicted driving path is determined based on the sum of the costs of each candidate path.

[0023] Optionally, based on the predicted driving path, the monitoring nodes and monitoring time windows are determined, including:

[0024] Get driving speed;

[0025] Identify monitoring nodes based on cameras along the predicted driving path;

[0026] Based on the driving speed, the distance between the target manhole cover and the monitoring node, and the alarm time corresponding to the target manhole cover, the monitoring time window for each monitoring node is determined.

[0027] Optionally, the driving speed is obtained, including:

[0028] The driving speed is determined based on the alarm time and location coordinates in each manhole cover alarm message.

[0029] Optionally, the monitoring nodes are determined based on cameras along the predicted driving path, including:

[0030] Based on the coverage area of ​​the cameras, duplicate cameras are deduplicated to obtain the monitoring nodes.

[0031] Optionally, acquiring video data collected by the monitoring node during the monitoring time window, and performing manhole cover location tracking based on the video data, includes:

[0032] In response to the detection of the same target object in the video data of two adjacent monitoring nodes, the speed of the target object is determined.

[0033] If the speed of the vehicle to be detected is less than or equal to the driving speed, and the alarm information of the manhole cover to be tracked is received when the vehicle to be detected is located between two adjacent monitoring nodes, then the vehicle to be detected is identified as the target object, and the manhole cover is located and tracked based on the target object. The manhole cover to be tracked is the manhole cover located between two adjacent monitoring nodes.

[0034] To achieve the above and other related objectives, this application provides an intelligent manhole cover positioning and tracking device, comprising:

[0035] The data acquisition module is used to acquire manhole cover alarm information within a preset time window. The manhole cover alarm information includes manhole cover identification, alarm time, and location coordinates.

[0036] The correlation determination module is used to determine the alarm correlation based on the manhole cover alarm information;

[0037] The path prediction module is used to determine the predicted driving path based on the manhole cover alarm information in response to the alarm correlation being associated with the manhole cover alarm.

[0038] The monitoring module is used to determine monitoring nodes and monitoring time windows based on the predicted driving path;

[0039] The positioning module is used to acquire video data collected by the monitoring node during the monitoring time window, and to perform manhole cover positioning and tracking based on the video data.

[0040] To achieve the above and other related objectives, this application also provides an intelligent manhole cover positioning and tracking system, comprising:

[0041] Storage media and processor;

[0042] The storage medium is used to store instructions;

[0043] The processor is configured to operate according to the instructions to execute one or more of the aforementioned smart manhole cover positioning and tracking methods.

[0044] As described above, the intelligent manhole cover positioning and tracking method, device, and system provided in this application have the following beneficial effects:

[0045] This application provides an intelligent manhole cover location and tracking method. This method correlates discrete manhole cover alarm information to determine if multiple alarm messages are related. Alarm correlation indicates that the target is a repeat offender with an intention to continue stealing. Therefore, when alarm correlation is positive, a predicted travel path can be determined. Monitoring nodes along the predicted travel path can collect video data within a monitoring time window and track the manhole cover based on the video data. By determining the predicted travel path and having monitoring nodes along it collect video data within the monitoring time window, manhole cover location and tracking can be performed even in scenarios where GPS signals are lost, enabling continuous tracking and location of stolen manhole covers, thus improving the reliability and efficiency of manhole cover location and tracking. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0047] Figure 1 This is a flowchart illustrating an exemplary embodiment of the smart manhole cover positioning and tracking method of this application;

[0048] Figure 2 This is a structural block diagram of an intelligent manhole cover positioning and tracking device, as illustrated in an exemplary embodiment of this application. Detailed Implementation

[0049] To better understand the above-mentioned objectives, features, and advantages of this application, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, where there is no conflict, the embodiments and features described in the embodiments of this application can be combined with each other.

[0050] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of this application is not limited to the specific embodiments disclosed below.

[0051] Please see Figure 1 The above is a flowchart illustrating an exemplary embodiment of a smart manhole cover positioning and tracking method according to this application. (See reference...) Figure 1 It can be seen that the intelligent manhole cover positioning and tracking method may include steps S110 to S150.

[0052] Step S110: Obtain manhole cover alarm information within a preset time window.

[0053] The manhole cover alarm information includes the manhole cover identification, alarm time, and location coordinates.

[0054] In one embodiment of this application, the execution device can continuously monitor the alarm event stream of smart manhole covers within a preset range. When any smart manhole cover is illegally moved and triggers an alarm, the alarm information corresponding to that smart manhole cover is recorded. The manhole cover identifier is used to distinguish different smart manhole covers; the alarm time is recorded and uploaded by the smart manhole cover's built-in clock chip; the location coordinates may include latitude and longitude data acquired by the smart manhole cover's built-in positioning module at the moment of the alarm, and may also include fixed coordinates of the smart manhole cover's installation location pre-stored in the execution device. The preset time window can be a duration range, determined based on the duration of continuous theft of multiple smart manhole covers in historical information, for example, it can be set to 15 minutes.

[0055] It should be noted that the execution device can be a terminal, a server, or a server cluster.

[0056] Step S120: Determine the alarm correlation based on the manhole cover alarm information.

[0057] In one embodiment of this application, after obtaining multiple manhole cover alarm messages within a preset time window, it can be determined whether there is an inherent spatiotemporal relationship between the manhole cover alarm messages. In reality, single-point alarms may be caused by sensor false alarms, construction disturbances, or accidental damage. Only when multiple manhole cover alarm messages are continuous in time and space can they point to continuous theft of the same target object.

[0058] For example, the system can determine whether alarm information from each manhole cover can form a theft path by calculating the time difference and distance relationship between adjacent alarm points. If the correlation determination is not valid, the system will only record the alarm without triggering the tracking process, thus avoiding wasting computing power on invalid events and interference from false alarms.

[0059] Step S130: In response to the alarm correlation being a manhole cover alarm correlation, a predicted driving path is determined based on the manhole cover alarm information.

[0060] In one embodiment of this application, when the alarm correlation of multiple manhole cover alarm messages is confirmed as manhole cover alarm association, the execution device can predict the possible movement trajectory of the target object based on the manhole cover alarm information, i.e., predict the driving path. Since the positioning signal of the smart manhole cover may have been lost or blocked at this time, and the real-time location cannot be directly obtained, it is necessary to generate several candidate paths based on the location coordinates of the last manhole cover alarm message and the urban road network topology. The process of determining the predicted driving path can be based on behavioral inference, such as whether the target object may continue to look for the next target or deliberately avoid areas with dense surveillance. Predicting the driving path can transform the manhole cover alarm information into a specific traceable path, which can define the scope for subsequent video investigation.

[0061] Step S140: Based on the predicted driving path, determine the monitoring nodes and monitoring time windows.

[0062] In one embodiment of this application, the monitoring node can be a camera device capable of capturing images of the target object along the predicted driving path; the monitoring time window can be the time period during which the target object is expected to pass through the monitoring node. The determination of the monitoring time window can be based on the alarm time of the last received manhole cover alarm, the path distance, and the estimated driving speed. For example, if the predicted driving path length is 1000 meters, the estimated speed is 10 meters per second, and the last alarm time is T0, then the center time of the monitoring time window corresponding to the camera 500 meters ahead is approximately T0+50 seconds, with a certain extension range set in conjunction with speed fluctuations and clock errors. This can reduce the massive amount of video data that originally required full-area, all-time retrieval to key segments within a smaller range, thereby reducing the amount of data for video analysis and the cost of manual review.

[0063] Step S150: Obtain video data collected by the monitoring node during the monitoring time window, and perform manhole cover location tracking based on the video data.

[0064] In one embodiment of this application, the execution device can retrieve corresponding video segments from a video networking platform or storage server based on determined monitoring nodes and monitoring time windows. The video data is then processed to identify the target object. Manhole cover location tracking based on video data can employ methods such as vehicle feature recognition, cross-camera tracking, and abnormal behavior detection to extract the target object matching the characteristics of the crime from the video footage.

[0065] In one embodiment of this application, the process of determining alarm correlation based on manhole cover alarm information in step S120 may include: determining the straight-line distance and actual road distance between two smart manhole covers with adjacent alarm times based on location coordinates; determining a first correlation based on the number of manhole cover alarm messages; determining a second correlation based on all straight-line distances and all actual road distances; and determining alarm correlation based on the first and second correlations.

[0066] The straight-line distance refers to the shortest geometric distance between two smart manhole covers on the Earth's surface, which can be calculated using the Haversine formula. The actual road distance refers to the shortest travel path length between the two smart manhole covers within the urban road network topology, which can be calculated based on road network data using the A* algorithm or other path search algorithms.

[0067] The first correlation includes two states: quantity correlation or quantity non-correlation. The execution device counts the total number k of manhole cover alarm messages within a preset time window and compares it with a preset quantity threshold Nth. When k is greater than or equal to Nth, the first correlation is determined to be quantity correlation; when k is less than Nth, it is determined to be quantity non-correlation.

[0068] For example, the quantity threshold Nth can be 3 to avoid including occasional false alarms or isolated damage events of a single manhole cover in the tracking process, thereby reducing the false trigger rate of the system.

[0069] The second correlation includes two states: distance correlation and distance non-correlation. For adjacent alarm point pairs sorted by time within a preset time window, the system checks whether the straight-line distance between each pair of adjacent points is less than or equal to a preset straight-line distance threshold Dmax, and whether the actual road distance is less than or equal to a preset road distance threshold Dmax_road. Only when all adjacent point pairs simultaneously meet both conditions is the second correlation determined to be distance correlation; if any pair of adjacent points does not meet either condition, it is determined to be distance non-correlation.

[0070] For example, the straight-line distance threshold Dmax can be 500 meters, and the road distance threshold Dmax_road can be 1000 meters. On the one hand, Dmax can filter out irrelevant alarms from a long distance, and on the other hand, Dmax_road can ensure that there is a reasonable driving path between adjacent alarm points, preventing the incorrect association of irrelevant events that are located in different areas but trigger a dense alarm in a short period of time due to sensor failure.

[0071] The first correlation includes quantity correlation or quantity non-correlation, and the second correlation includes distance correlation or distance non-correlation. The process of determining alarm correlation based on the first correlation and the second correlation may include: determining alarm correlation as manhole cover alarm correlation when the first correlation is quantity correlation and the second correlation is distance correlation; determining alarm correlation as manhole cover alarm non-correlation when the first correlation is quantity non-correlation and / or the second correlation is distance non-correlation.

[0072] The system only confirms that the manhole cover alarm information constitutes a valid sequence of crimes when the number of alarms reaches the typical threshold for gang theft, and all adjacent alarm points are both linearly adjacent and connected by roads. If the number of alarms is insufficient, or if the number is sufficient but the points lack reasonable spatial continuity, the system determines that they are unrelated and only reports the manhole cover alarm information for staff to verify.

[0073] In one embodiment of this application, step S130, which involves determining a predicted driving path based on manhole cover alarm information in response to alarm correlation, may include: taking the location coordinates of the target manhole cover as the starting point, extending a preset distance in the road network to obtain multiple candidate paths, where the target manhole cover is the last manhole cover to trigger an alarm; determining the path distance, number of cameras, and number of manhole covers for each candidate path; determining the total cost of each candidate path based on the path distance, number of cameras, and number of manhole covers; and determining the predicted driving path based on the total cost of each candidate path.

[0074] The target manhole cover can be the one with the latest alarm time among all manhole cover alarm messages, and its location coordinates represent the last known location of the target at the current moment. The preset distance can be set to 1500 meters, which can include both the target's escape route and the potential area where the target may continue to search for the next target.

[0075] For each candidate path, the execution device needs to extract feature data in three dimensions. Path distance represents the actual travel length along the road network from the target manhole cover location to the candidate path endpoint, calculated using path search algorithms such as A*, and represents the efficiency cost of movement. The number of cameras represents the total number of municipal monitoring devices along the candidate path, obtained by querying the camera GIS database. The number of manhole covers represents the total number of smart manhole covers along the candidate path that have not yet been stolen, calculated by querying the manhole cover GIS database and excluding those that have already triggered alarms.

[0076] The total cost of each candidate path is determined based on its path distance, number of cameras, and number of manhole covers. In this embodiment, the total cost is calculated using the following formula:

[0077] ;

[0078] in, This represents the total cost of the j-th candidate path. The lower the cost value, the higher the probability that the target object will choose this path. This represents the path distance of the j-th candidate path. This represents a preset distance, used for distance normalization. This represents the number of cameras on the j-th candidate path. This represents the maximum number of cameras among all candidate paths. If the maximum value is 0, it is set to 1 to avoid division by zero error. This is used to normalize the number of cameras. This represents the number of manhole covers on the j-th candidate path. This represents the maximum number of manhole covers among all candidate paths. If the maximum value is 0, it is set to 1, which is used to normalize the number of manhole covers. This represents the blind spot penalty. When the number of cameras on the candidate path is 0, a preset large value (e.g., 1) is taken. hour, Take 0. As the first weight, As the second weight, As the third weight, It is the fourth weight.

[0079] The value can be 0.35. The core objective of the target is to obtain manhole covers. The more manhole covers that are not stolen along the way, the greater the potential profit from continuing the crime. The value can be 0.25, indicating that the target object tends to avoid monitoring risks. The value can be 0.25, and its purpose is to balance the feasibility of tracking: although a path without cameras is advantageous for committing crimes, if the system cannot track it through video at all, then the path is useless for tracking; after introducing this item, unless a path without monitoring has an extremely high density of manhole covers, the system will appropriately reduce its priority and recommend a traceable path that has both manhole covers and a small number of cameras instead. The value can be 0.15.

[0080] It should be noted that the preset distance is the fixed maximum search boundary set by the system, the path distance is the actual length that the candidate path extends to under the road network constraints, and the predicted path is all possible routes generated within the constraints, but the actual length of each path is determined by the road network topology and is usually less than or equal to the preset distance.

[0081] In the formula for calculating total cost, the longer the path, the longer the time spent, and the higher the risk; the more cameras there are, the easier it is to be photographed, and the higher the risk; the more manhole covers on the path, the greater the probability that the target will choose that path. In the formula for calculating total cost, the higher the target's benefit, the lower the total cost should be. When there are zero cameras on a candidate path, the candidate path can be notified to operators for manual inspection, such as arranging inspectors at intersections to manually check vehicles.

[0082] The system can sort all candidate paths by total cost from smallest to largest, and select one or more paths with the lowest cost (selecting the top three) as the predicted driving path.

[0083] In one embodiment of this application, the process of determining the monitoring node and monitoring time window based on the predicted driving path in step S140 may include: obtaining the driving speed; determining the monitoring node based on the camera on the predicted driving path; and determining the monitoring time window of each monitoring node based on the driving speed, the distance between the target manhole cover and the monitoring node, and the alarm time corresponding to the target manhole cover.

[0084] In one embodiment of this application, the process of obtaining the driving speed may include: determining the driving speed based on the alarm time in each manhole cover alarm message. Driving speed can characterize the actual speed at which the target object moves on the road. Since there is inevitably a process of stopping, prying open, moving, and restarting the vehicle between two adjacent manhole covers, if the speed is calculated by directly dividing the distance between adjacent alarm points by the time difference, the resulting speed will be significantly lower than the actual vehicle speed because it includes a long stopping time. For example, if the distance between two manhole covers is 300 meters, the driving time is 20 seconds, but the stopping time is 100 seconds, the speed is only 2.5 meters per second, while the actual driving speed may reach 15 meters per second. If the time window is calculated based on this speed, it will cause a lag in the predicted vehicle arrival time, forcing the generated monitoring time window to be significantly extended to cover the uncertainty.

[0085] In one embodiment of this application, the process of determining monitoring nodes based on cameras along a predicted driving path may include: deduplicating cameras based on their coverage area to obtain monitoring nodes. Specifically, the predicted driving path may be densely populated with monitoring devices. If all devices are used as monitoring nodes, it will not only waste video retrieval resources but also increase the concurrent pressure on intelligent analysis. Coverage area refers to the physical radius of the area within which a single camera can effectively capture vehicle features, typically set to 100 meters in urban road scenarios. When the coverage areas of multiple cameras overlap spatially, meaning they can effectively capture the same road segment, the executing device can treat them as a group of redundant nodes and select the optimal device (or any device) as the final monitoring node. The selection strategy can comprehensively consider factors such as the angle between the camera's installation orientation and the path direction, image resolution, and night vision capability, prioritizing devices facing the oncoming traffic and with better image quality. For example, if an intersection has both a high-definition checkpoint camera and a low-resolution security PTZ camera, and both cover the central area of ​​the intersection, the system will only identify the high-definition checkpoint camera as a monitoring node.

[0086] The formula for calculating the monitoring time window may include:

[0087] ;

[0088] in, This represents the monitoring time window corresponding to the m-th monitoring node, including the start and end times. This indicates the alarm time corresponding to the target manhole cover. This represents the shortest road distance from the location coordinates of the target manhole cover to the m-th monitoring node along the predicted driving path, which can be obtained through a road network path search algorithm. Indicates travel speed. Indicates the time window tolerance. When the driving speed... When the speed is greater than 15 m / s, the time window tolerance It can be compressed to about 30 seconds; and when the driving speed... When the speed is less than 5 m / s, the time window tolerance It can be extended to 180 seconds. A narrow time window is generated in high-speed tracking scenarios, while the window is widened in low-speed or complex road conditions to avoid missed detections.

[0089] In one embodiment of this application, the process of obtaining the driving speed may include: determining the driving speed based on the alarm time and location coordinates in each manhole cover alarm information.

[0090] In one possible implementation, the initial speed can be calculated based on the alarm information of adjacent manhole covers in time, and the driving speed can be determined based on the initial speed.

[0091] The system can calculate the actual distance on the road based on the location coordinates of two adjacent manhole covers in the manhole cover alarm information. And calculate the time difference based on the corresponding alarm timestamp. , The alarm time for the (i+1)th manhole cover alarm message. Let be the alarm time of the i-th manhole cover alarm message. The manhole cover alarm messages are ordered according to their alarm times. Using the formula... Calculate the initial velocity for each segment of the path. Wherein, This represents the initial velocity of the i-th path segment. This represents the actual shortest road distance between the (i+1)th manhole cover and the ith manhole cover. This indicates the time interval between two adjacent manhole cover alarm messages. In a manhole cover theft scenario, the initial speed includes not only the pure driving time of the vehicle between the two points, but also the time spent stopping, prying open the cover, moving the manhole cover, and restarting the vehicle.

[0092] The formula for calculating driving speed can be expressed as:

[0093] ;

[0094] in, This indicates the final determined driving speed. This represents the initial velocity of each path segment in the associated alarm sequence, where n is the number of initial velocities. This indicates the current speed limit for a road segment or urban road, such as an 80 km / h (approximately 22.2 m / s) speed limit on a main road. It is used to prevent abnormally high speeds caused by sensor clock drift or measurement errors over extremely short distances. The function represents selecting the maximum value from all initial velocities. The function compares the selected maximum value with the speed limit and takes the smaller value as the driving speed.

[0095] It should be noted that during the continuous theft of multiple manhole covers, although a stop is required at each smart manhole cover, the duration of the stop at each location often varies. For example, some manhole covers may be difficult to pry open due to severe rust, taking longer, while others may be easily obtained due to loose structure or the suspect's skilled technique. Among all adjacent manhole cover alarm information pairs, the one with the highest initial speed corresponds to a relatively short stopping time. Therefore, the initial speed in this segment is least affected by the stopping time and is closest to the actual speed of the vehicle on the road.

[0096] In another possible implementation, the initial speed can be directly determined as the driving speed. Since the initial speed may differ significantly from the actual speed, the monitoring time window can be enlarged to obtain sufficient video data.

[0097] In one embodiment of this application, the process of locating and tracking manhole covers based on video data in step S150 may include: in response to detecting the same object to be detected in the video data of two adjacent monitoring nodes, determining the speed of the object to be detected; if the speed of the object to be detected is less than or equal to the driving speed, and a manhole cover alarm message for the manhole cover to be tracked is received when the object to be detected is located between two adjacent monitoring nodes, then the object to be detected is identified as the target object, so as to perform manhole cover location and tracking based on the target object, wherein the manhole cover to be tracked is the manhole cover located between two adjacent monitoring nodes.

[0098] Vehicle Re-identification (VREI) technology can be used to detect whether the same target object appears in the video data of two adjacent monitoring nodes. If the same target object is detected in two adjacent monitoring nodes, it can be determined that the target object has passed on the predicted driving path. The speed of the target object can be determined based on the time the target object appears in the two adjacent monitoring nodes and the path distance between the two adjacent monitoring nodes in the road network data. If the target object's speed is less than the driving speed and a manhole cover alarm message is received when the target object is between two adjacent monitoring nodes, it indicates that a manhole cover theft event has occurred when the target object is between the two adjacent monitoring nodes. If the target object's speed is less than or equal to the driving speed, it indicates that the target object may have stopped. Therefore, when the target object's speed is less than the driving speed and a manhole cover alarm message is received when the target object is between two adjacent monitoring nodes, the target object can be identified as the target object, and manhole cover location and tracking can be performed based on the target object.

[0099] The system can automatically output information such as the appearance features, time of appearance, and location coordinates of the target object, which can be reviewed by the operator or used to trigger subsequent instructions.

[0100] In one possible implementation, when the predicted number of cameras along the driving path is zero and the number of manhole covers is greater than or equal to a preset threshold, the execution device triggers a blind spot control early warning mechanism. If a candidate path is deemed a low-risk, high-return path due to its extremely high manhole cover density, but it lacks any municipal monitoring equipment coverage along its route (i.e., the number of cameras is zero and the number of manhole covers is greater than or equal to a preset threshold), then this path cannot be targeted using conventional video tracking methods. In this case, the execution device can mark it as a high-value blind spot path and prompt the operator to investigate.

[0101] In another possible implementation, when the number of cameras on the predicted driving path is 1, the video data of the monitoring node within the monitoring time window can be analyzed to identify vehicles with obscured license plates, faces, or other abnormal behaviors, and these vehicles can be identified as potential targets for investigation and given risk warnings.

[0102] For example, consider the following scenario: Between 11:05 PM and 11:12 PM on a certain night, the system continuously receives alarm messages from three smart manhole covers. Because the target person used a GPS signal jammer, the GPS signal of all manhole covers immediately disappears after triggering the alarm, making it impossible to provide continuous real-time trajectory data. The system first acquires and records the alarm information, recording the alarm information for each of the three manhole covers, including the manhole cover identification, alarm time, and location coordinates at the moment of the alarm. Subsequently, the system initiates a spatiotemporal correlation determination process based on this alarm information. The three alarm points are within a preset time window of 900 seconds, and the straight-line distance between adjacent alarm points is less than 500 meters, the actual road distance is less than 1000 meters, and the number of alarms reaches a threshold of three. The system determines that the first correlation is a quantity correlation and the second correlation is a distance correlation, thus determining the alarm correlation as a manhole cover alarm correlation.

[0103] After confirming the association, the execution device can calculate the initial speed using the distance and time difference between adjacent alarm points. Since there is parking and manhole cover prying work at each point, the directly calculated initial speed is distorted by the parking time. The execution device can select the maximum value from multiple initial speeds as an approximation of the actual vehicle speed and adjust it according to road speed limits. In this embodiment, the maximum initial speed of one segment is 13.5 m / s. The execution device can determine this value as the driving speed. Starting from the location of the target alarm manhole cover, the execution device obtains three candidate paths for cost evaluation: Path A is a main urban road with 15 cameras along the way and only 2 unstolen manhole covers; Path B is a remote side road with 3 cameras and 8 unstolen manhole covers along the way; Path C is a dead-end road with no cameras and no manhole cover revenue. Path A was excluded due to high monitoring risk and low returns; Path C, although unmonitored, was downgraded due to lack of criminal value and triggering blind spot penalties; Path B, combining high manhole cover returns and low monitoring risk, calculated the lowest total cost and was selected as the predicted driving path. Based on the selected Path B, the system further determined the monitoring nodes and monitoring time windows. There are two effective cameras along Path B. The executing device spatially deduplicated the cameras based on their coverage area, retaining two devices as monitoring nodes. Using the previously determined driving speed of 13.5 m / s and the path distance, the system calculated the baseline time when the suspect vehicle was expected to arrive at each monitoring node, and adaptively narrowed the time window tolerance to 30 seconds based on the high-speed driving scenario, thus obtaining the monitoring time window.

[0104] After acquiring the video data collected by the monitoring nodes within the monitoring time window, the execution device performs manhole cover location tracking based on the video data. If three objects to be detected are detected in the video data of two monitoring nodes, and one of the objects to be detected has a vehicle speed of less than 13.5 m / s, and the execution device receives a manhole cover alarm information uploaded by the manhole cover located between the two monitoring nodes while the object to be detected is traveling between the two monitoring nodes, then the object to be detected with a vehicle speed of less than 13.5 m / s can be identified as the target object.

[0105] Please see Figure 2 This is a structural block diagram illustrating an exemplary embodiment of the intelligent manhole cover positioning and tracking device of this application. The intelligent manhole cover positioning and tracking device 200 may include:

[0106] The data acquisition module 210 is used to acquire manhole cover alarm information within a preset time window. The manhole cover alarm information includes manhole cover identification, alarm time, and location coordinates.

[0107] The correlation determination module 220 is used to determine the alarm correlation based on the manhole cover alarm information;

[0108] The path prediction module 230 is used to determine the predicted driving path based on the manhole cover alarm information in response to the alarm correlation being associated with a manhole cover alarm.

[0109] Monitoring module 240 is used to determine monitoring nodes and monitoring time windows based on the predicted driving path;

[0110] The positioning module 250 is used to acquire video data collected by the monitoring node during the monitoring time window, and to perform manhole cover positioning and tracking based on the video data.

[0111] It should be noted that the intelligent manhole cover positioning and tracking device and the intelligent manhole cover positioning and tracking method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the intelligent manhole cover positioning and tracking device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0112] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by one or more processors, enable the electronic device to implement the intelligent manhole cover positioning and tracking method provided in the above embodiments.

[0113] Another aspect of this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a computer's processor, causes the computer to perform the intelligent manhole cover positioning and tracking method provided in the various embodiments described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not be assembled into the electronic device.

[0114] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the intelligent manhole cover positioning and tracking method provided in the various embodiments described above.

[0115] Embodiments of this application also provide an intelligent manhole cover positioning and tracking system, comprising: a storage medium and a processor; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the intelligent manhole cover positioning and tracking method provided in the above embodiments.

[0116] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0117] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for intelligent manhole cover positioning and tracking, characterized in that, include: Obtain manhole cover alarm information within a preset time window, the manhole cover alarm information including manhole cover identification, alarm time and location coordinates; Based on the manhole cover alarm information, the alarm correlation is determined; In response to alarm correlation, a manhole cover alarm is associated, and a predicted driving path is determined based on the manhole cover alarm information. Based on the predicted driving path, the monitoring nodes and monitoring time windows are determined; The monitoring node acquires video data collected during the monitoring time window and performs manhole cover location tracking based on the video data.

2. The intelligent manhole cover positioning and tracking method according to claim 1, characterized in that, The determination of alarm correlation based on the manhole cover alarm information includes: Determine the straight-line distance and actual road distance between two smart manhole covers with adjacent alarm times based on location coordinates; The primary correlation is determined based on the number of manhole cover alarm messages; The second correlation is determined based on all straight-line distances and all actual road distances; Alarm correlation is determined based on the first and second correlations.

3. The intelligent manhole cover positioning and tracking method according to claim 2, characterized in that, The first correlation includes quantity correlation or no quantity correlation; the second correlation includes distance correlation or no distance correlation. The determination of alarm correlation based on the first and second correlations includes: Given that the first correlation is a quantity correlation and the second correlation is a distance correlation, the alarm correlation is determined to be a manhole cover alarm correlation; If the first correlation is quantity-independent and / or the second correlation is distance-independent, the alarm correlation is determined to be manhole cover alarm-independent.

4. The intelligent manhole cover positioning and tracking method according to claim 1, characterized in that, The aforementioned response to alarm correlation refers to manhole cover alarm correlation. Based on the manhole cover alarm information, a predicted driving path is determined, including: Starting from the location coordinates of the target manhole cover, multiple candidate paths are obtained by extending a preset distance in the road network; the target manhole cover is the last manhole cover to trigger an alarm. Based on each candidate path, determine the path distance, number of cameras, and number of manhole covers for each candidate path; The total cost of each candidate path is determined based on the path distance, number of cameras, and number of manhole covers. The predicted driving path is determined based on the sum of the costs of each candidate path.

5. The intelligent manhole cover positioning and tracking method according to claim 1, characterized in that, The determination of monitoring nodes and monitoring time windows based on the predicted driving path includes: Get driving speed; Identify monitoring nodes based on cameras along the predicted driving path; Based on the driving speed, the distance between the target manhole cover and the monitoring node, and the alarm time corresponding to the target manhole cover, the monitoring time window for each monitoring node is determined.

6. The intelligent manhole cover positioning and tracking method according to claim 5, characterized in that, The method of obtaining driving speed includes: The driving speed is determined based on the alarm time and location coordinates in each manhole cover alarm message.

7. The intelligent manhole cover positioning and tracking method according to claim 5, characterized in that, The method of determining monitoring nodes based on cameras along the predicted driving path includes: Based on the coverage area of ​​the cameras, duplicate cameras are deduplicated to obtain the monitoring nodes.

8. The intelligent manhole cover positioning and tracking method according to claim 1, characterized in that, The process of acquiring video data collected by the monitoring node during the monitoring time window and performing manhole cover location tracking based on the video data includes: In response to the detection of the same target object in the video data of two adjacent monitoring nodes, the speed of the target object is determined. If the speed of the vehicle to be detected is less than or equal to the driving speed, and the alarm information of the manhole cover to be tracked is received when the vehicle to be detected is located between two adjacent monitoring nodes, then the vehicle to be detected is identified as the target object, and the manhole cover is located and tracked based on the target object. The manhole cover to be tracked is the manhole cover located between two adjacent monitoring nodes.

9. A smart manhole cover positioning and tracking device, characterized in that, include: The data acquisition module is used to acquire manhole cover alarm information within a preset time window. The manhole cover alarm information includes manhole cover identification, alarm time, and location coordinates. The correlation determination module is used to determine the alarm correlation based on the manhole cover alarm information; The path prediction module is used to determine the predicted driving path based on the manhole cover alarm information in response to the alarm correlation being associated with the manhole cover alarm. The monitoring module is used to determine monitoring nodes and monitoring time windows based on the predicted driving path; The positioning module is used to acquire video data collected by the monitoring node during the monitoring time window, and to perform manhole cover positioning and tracking based on the video data.

10. An intelligent manhole cover positioning and tracking system, characterized in that, include: Storage media and processor; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the method of any one of claims 1-8.

Images