Monitoring method and system

Abstract

This application discloses a monitoring method and system. The monitoring method is applied to a monitoring platform and includes: when a target object is present in the monitoring video of a target area acquired by a first monitoring device, and the value information of the target object meets the value criteria, determining at least one second monitoring device, the monitoring range of each second monitoring device including the target area; and sending a first cooperation instruction to each second monitoring device, the first cooperation instruction instructing the second monitoring device to synchronously acquire the monitoring video of the target area. Through this method, on the one hand, by coordinating the monitoring of at least one second monitoring device, monitoring video of the target area can be acquired synchronously from multiple directions, improving the reliability of monitoring key target objects; on the other hand, the coordinating monitoring of at least one second monitoring device is only triggered when the value information of the target object meets the value criteria, enabling on-demand allocation of monitoring resources and reducing unnecessary coordinating monitoring.

Description

Technical Field

[0001] This application relates to the field of security technology, and in particular to a monitoring method and system. Background Technology

[0002] Surveillance equipment is the core front-end equipment in the field of security technology. It is widely deployed in various scenarios such as parks and buildings, roads and traffic, and industrial plants. It is an important infrastructure for realizing full-area security monitoring, on-site situational awareness, and post-event traceability.

[0003] In most mainstream monitoring systems, each monitoring device operates independently, meaning it typically completes video acquisition and recording storage within its own monitoring range independently. The monitoring range of a single monitoring device is limited, which can easily lead to blind spots and make it difficult to adapt to the monitoring needs of complex scenarios. Summary of the Invention

[0004] The main technical problem addressed by this application is to provide a monitoring method and system that can synchronously acquire monitoring videos of a target area from multiple perspectives, improve the reliability of monitoring key target objects, and allocate monitoring resources on demand, thereby reducing unnecessary collaborative monitoring.

[0005] To address the aforementioned technical problems, this application provides a monitoring method applied to a monitoring platform. The method includes: responding to a first monitoring device acquiring a monitoring video of a target area where a target object exists and the value information of the target object meets a value condition; determining at least one second monitoring device; wherein the monitoring range of each second monitoring device includes the target area; and sending a first cooperation instruction to each second monitoring device, the first cooperation instruction instructing the second monitoring device to synchronously acquire the monitoring video of the target area.

[0006] In one embodiment, the value information of the target object is determined using reference information, which includes at least one of the following: the type of the target object, the posture of the target object, the behavior of the target object, and the clarity of the monitoring footage of the target object. And / or, the value information of the target object is a value score, wherein the value condition includes: the value score of the target object is greater than or equal to a first score threshold.

[0007] In one embodiment, determining at least one second monitoring device includes: determining a group of cooperating devices corresponding to a target area; wherein the group of cooperating devices includes a first monitoring device and at least one third monitoring device, and the monitoring range of each third monitoring device includes the target area; and selecting at least one second monitoring device from the at least one third monitoring device.

[0008] In one embodiment, selecting at least one second monitoring device from at least one third monitoring device includes: for each third monitoring device, obtaining a score for the third monitoring device using at least one evaluation factor of the third monitoring device; wherein the at least one evaluation factor includes at least one of the reciprocal of the estimated rotation time of the third monitoring device, the reciprocal of the load data of the third monitoring device, and historical collaborative monitoring task execution data of the third monitoring device, the estimated rotation time of the third monitoring device refers to the time required for the third monitoring device to rotate from its current pose to its preset pose, and the monitoring screen of the third monitoring device includes the target area when the third monitoring device is in the preset pose; selecting at least one monitoring device whose score meets the requirements from at least one third monitoring device as at least one second monitoring device.

[0009] In one embodiment, the score of the third monitoring device is obtained by weighting each evaluation factor using the target weight of each evaluation factor. The step of obtaining the target weight of each evaluation factor includes: selecting a weighted set corresponding to the monitoring scenario of the collaborative device group from the weight mapping relationship; wherein, the weight mapping relationship includes several preset monitoring scenarios and preset weighted sets corresponding to each preset monitoring scenario, and the preset weighted sets include preset weights corresponding to each evaluation factor; and taking the weight of each evaluation factor in the selected weighted set as the target weight of each evaluation factor.

[0010] In one embodiment, the step of obtaining the preset weighted reassembly corresponding to each preset monitoring scenario includes: generating a plurality of candidate weighted reassemblies; wherein the candidate weighted reassemblies include candidate weights corresponding to each evaluation factor; for each candidate weighted reassembly, in at least one test collaborative monitoring task under the preset monitoring scenario, selecting the collaborative monitoring equipment for each test collaborative monitoring task using the candidate weighted reassemblies; for each test collaborative monitoring task, determining the task parameters of the test collaborative monitoring task when the collaborative monitoring equipment is used to execute the test collaborative monitoring task; and determining the performance score corresponding to the candidate weighted reassembly using the task parameters of each test collaborative monitoring task; wherein the task parameters of the test collaborative monitoring task include at least one of the task duration of the test collaborative monitoring task, the comprehensive value score of the test object in the test collaborative monitoring task, and the total rotational loss of all participating monitoring equipment in the test collaborative monitoring task; and selecting the weighted reassembly with the highest performance score from the plurality of candidate weighted reassemblies as the preset weighted reassembly corresponding to the preset monitoring scenario.

[0011] In one embodiment, the method further includes: in response to determining that a target object is about to leave the monitoring range of the first monitoring device and the second monitoring device, or determining that there is no available third monitoring device in the collaborative device group corresponding to the target area, predicting a new target area where the target object appears; selecting at least one fourth monitoring device from the new collaborative device group corresponding to the new target area; and sending a takeover instruction to the at least one fourth monitoring device, the takeover instruction being used to instruct each fourth monitoring device to acquire monitoring video of the new target area.

[0012] In one embodiment, a first cooperation instruction is sent to each of the second monitoring devices, including: determining a cooperation device group corresponding to the target area; for each second monitoring device, selecting a preset pose of the second monitoring device from the cooperation mapping relationship corresponding to the cooperation device group; and sending a first cooperation instruction carrying the preset pose of the second monitoring device to the second monitoring device; wherein the cooperation mapping relationship includes the preset poses corresponding to each monitoring device in the cooperation device group, and the monitoring screen of each monitoring device in the cooperation device group includes the target area when it is in the corresponding preset pose.

[0013] In one embodiment, the method further includes: in response to meeting calibration conditions, calibrating the collaborative mapping relationship corresponding to the collaborative device group; wherein the calibration conditions are any of the following: the time elapsed since the last calibration reaches a preset time elapsed, each monitoring device in the collaborative device group restarts, and when there is a fifth monitoring device in the collaborative device group that has received a collaborative instruction and the fifth monitoring device is in the corresponding preset pose, the monitoring screen does not include the target object.

[0014] In one embodiment, the method further includes: using the image with the highest value information from the surveillance videos acquired by the first monitoring device and each of the second monitoring devices during the collaboration period as a preview image for the collaboration period; and / or, in response to receiving a video playback command sent by the client regarding the first device, synchronously pushing the surveillance videos acquired by the first device and the second device during the collaboration period to the client; and in response to receiving a video pause playback command sent by the client regarding the first device, pausing the pushing of the surveillance videos acquired by the first device and the second device during the collaboration period to the client; wherein the first device and the second device are monitoring devices participating in the same collaborative monitoring task.

[0015] In one embodiment, the method further includes: in response to determining from the monitoring video of the first monitoring device or the monitoring video of the second monitoring device that a group event concerning a group target has occurred, determining the coverage area corresponding to the group event; determining at least one sixth monitoring device whose monitoring range includes the coverage area; and sending a second cooperation instruction to each of the sixth monitoring devices, the second cooperation instruction being used to instruct the sixth monitoring devices to synchronously acquire monitoring video concerning the coverage area.

[0016] To address the aforementioned technical problems, another technical solution adopted in this application is: providing a monitoring method, which is applied to a first monitoring device, and the method includes: in response to detecting the presence of a target object in a target area from a monitoring video, determining the value information of the target object; in response to the value information of the target object meeting the value condition, sending a cooperation request to a monitoring platform; wherein the cooperation request is used to instruct the monitoring platform to determine at least one second monitoring device and send a first cooperation instruction to each second monitoring device respectively, the monitoring range of each second monitoring device includes the target area, and the first cooperation instruction is used to instruct the second monitoring device to synchronously acquire monitoring video about the target area.

[0017] To address the aforementioned technical problems, another technical solution adopted in this application is: providing a monitoring method, which is applied to at least one second monitoring device, the monitoring range of each second monitoring device including the target area, and the method includes: in response to receiving a first cooperation instruction, acquiring monitoring video of the target area; wherein, the first cooperation instruction is sent by the monitoring platform when it determines that a target object exists in the monitoring video of the target area acquired by the first monitoring device and that the value information of the target object meets the value condition.

[0018] To solve the above-mentioned technical problems, another technical solution adopted in this application is to provide a monitoring system, which includes a monitoring platform and several monitoring devices, and is used to implement any of the above-mentioned monitoring methods.

[0019] In the above scheme, when the surveillance video of the target area acquired by the first monitoring device contains a target object and the value information of the target object meets the value criteria, at least one second monitoring device is identified as including the target area in the monitoring scope. A first cooperation instruction is then sent to each of the second monitoring devices to instruct them to simultaneously acquire surveillance video of the target area. This approach achieves two advantages: firstly, through the collaborative monitoring of at least one second monitoring device, surveillance video of the target area can be acquired synchronously from multiple perspectives, improving the reliability of monitoring key target objects; secondly, the collaborative monitoring of at least one second monitoring device is only triggered when the value information of the target object meets the value criteria, enabling on-demand allocation of monitoring resources and reducing unnecessary collaborative monitoring. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the framework of an embodiment of the monitoring system provided in this application; Figure 2 This is a flowchart illustrating an embodiment of the monitoring method provided in this application; Figure 3 This is a flowchart illustrating another embodiment of the monitoring method provided in this application; Figure 4This is a flowchart illustrating another embodiment of the monitoring method provided in this application; Figure 5 This is a flowchart illustrating another embodiment of the monitoring method provided in this application; Figure 6 This is a schematic diagram of the framework of an embodiment of the electronic device provided in this application; Figure 7 This is a schematic diagram of a framework of an embodiment of the computer-readable storage medium provided in this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0022] It should be noted that the term "and / or" in this application 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, or B existing alone. The terms "first" and "second" in this application are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. The term "multiple" in this application means at least two, such as two, three, etc. The term "several" in this application means at least two. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0023] Please see Figure 1 , Figure 1 This is a schematic diagram of the framework of an embodiment of the monitoring system provided in this application. Figure 1 As shown, the monitoring system includes a monitoring platform and several monitoring devices. The monitoring devices can be PTZ cameras. These devices connect to the monitoring platform via a network and can send the acquired video footage to the platform. The monitoring platform can issue control commands to the monitoring devices to control the PTZ rotation, thereby changing the device's position and allowing the acquisition of video footage from different angles.

[0024] In this embodiment, the monitoring devices of the monitoring system can be further divided into at least one collaborative device group. Each collaborative device group includes multiple monitoring devices, and the monitoring range of each monitoring device belonging to the same collaborative device group includes the same target area (the actual physical area), but the target areas corresponding to different collaborative device groups are different. For example, the monitoring devices include collaborative device group a and collaborative device group b, wherein collaborative device group a includes monitoring device 1, monitoring device 2, and monitoring device 3, and the monitoring ranges of monitoring device 1, monitoring device 2, and monitoring device 3 all include target area a; collaborative device group b includes monitoring device 4 and monitoring device 5, and the monitoring ranges of monitoring device 4 and monitoring device 5 both include target area b. By binding monitoring devices with overlapping monitoring views into a collaborative device group, monitoring devices belonging to the same collaborative device group can collaboratively monitor target objects appearing in the overlapping areas of their views.

[0025] Furthermore, a collaborative mapping relationship can be set for each collaborative device group. The collaborative mapping relationship of the collaborative device group includes the preset poses of each monitoring device in the collaborative device group, and when each monitoring device in the collaborative device group is in the corresponding preset pose, the monitoring screen includes the target area corresponding to the collaborative device group. Among them, the preset poses of the monitoring devices include the PTZ parameters of the monitoring devices (i.e., horizontal angle Pan, vertical angle Tilt, and zoom).

[0026] For example, collaborative device group a includes monitoring device 1, monitoring device 2, and monitoring device 3. The monitoring range of monitoring device 1, monitoring device 2, and monitoring device 3 all include the target area a. The preset poses of monitoring device 1, monitoring device 2, and monitoring device 3 are P11, P23, and P35, respectively. When the current pose of monitoring device 1 is P11, the current pose of monitoring device 2 is P23, and the current pose of monitoring device 3 is P35, the monitoring screens of monitoring device 1, monitoring device 2, and monitoring device 3 all include the target area a.

[0027] The collaboration mapping relationship corresponding to the above-mentioned collaborative device group can be stored in the monitoring platform, or the collaboration mapping relationship corresponding to the collaborative device group can be stored in each monitoring device of the collaborative device group.

[0028] By pre-setting corresponding collaborative mapping relationships for collaborative device groups, it becomes possible to quickly control each monitoring device in the collaborative device group to rapidly turn to the corresponding target area based on these collaborative mapping relationships.

[0029] Please see Figure 2 , Figure 2This is a flowchart illustrating an embodiment of the monitoring method provided in this application, which is executed by a monitoring platform. Figure 2 As shown, the method includes the following steps: S21: When the first monitoring device acquires a monitoring video of the target area containing a target object, and the value information of the target object meets the value condition, at least one second monitoring device is identified.

[0030] In this system, both the first and second monitoring devices are among several monitoring devices in the monitoring system. The monitoring range of the first monitoring device and the monitoring range of each of the second monitoring devices both include the target area, meaning there is an overlapping area of ​​view.

[0031] The target object is the object that needs to be focused on. For example, it could be a person, a vehicle, an animal, etc.

[0032] When the value information of a target object meets the value criteria, it indicates that the target object has a high value. At this point, at least one second monitoring device is further identified as a collaborating monitoring device with the first monitoring device to jointly monitor the target object within the target area.

[0033] When the value information of a target object does not meet the value criteria, it indicates that the target object has a low value. In this case, at least one second monitoring device will not be selected; only the first monitoring device will monitor the target object in the target area.

[0034] S22: Send the first cooperation instruction to each of the second monitoring devices respectively.

[0035] The first collaboration instruction is used to instruct the second monitoring devices to synchronously acquire surveillance video of the target area. Upon receiving the first collaboration instruction from the monitoring platform, each second monitoring device begins synchronously acquiring surveillance video of the target area.

[0036] The surveillance video of the target area acquired by the first and second monitoring devices during the collaboration period can be stored in the corresponding local storage unit or sent to the monitoring platform for storage. The surveillance video of the target area acquired by the first and second monitoring devices during the collaboration period is recorded as a collaborative monitoring task. The collaborative monitoring task can be defined as the tracking process of the target object from the initiation of collaboration to the end of collaboration (such as the target object leaving the monitoring screen of the second monitoring device).

[0037] In this embodiment, when the surveillance video of the target area acquired by the first monitoring device contains a target object and the value information of the target object meets the value condition, at least one second monitoring device is determined to include the target area in the monitoring range. A first cooperation instruction is then sent to each of the second monitoring devices to instruct them to synchronously acquire surveillance video of the target area. This approach, on the one hand, enables synchronous acquisition of surveillance video of the target area from multiple perspectives through collaborative monitoring by at least one second monitoring device, improving the reliability of monitoring key target objects. On the other hand, collaborative monitoring by at least one second monitoring device is only triggered when the value information of the target object meets the value condition, allowing for on-demand allocation of monitoring resources and reducing unnecessary collaborative monitoring.

[0038] Please see Figure 3 , Figure 3 This is a flowchart illustrating another embodiment of the monitoring method provided in this application, which is executed by a monitoring platform. Figure 3 As shown, the method includes the following steps: S31: When the first monitoring device obtains a monitoring video of the target area containing a target object, and the value information of the target object meets the value condition, the corresponding cooperative device group of the target area is determined.

[0039] In one embodiment, the first monitoring device performs target detection on the acquired monitoring video of the target area. When a target object is detected in the video image, it further determines the value information of the target object and judges whether the value information of the target object meets the value condition. When it is determined that the value information of the target object meets the value condition, the first monitoring device sends a cooperation request to the monitoring platform. When the monitoring platform receives the cooperation request from the first monitoring device, it determines that a target object exists in the monitoring video of the target area acquired by the first monitoring device, and that the value information of the target object meets the value condition.

[0040] In another embodiment, the first monitoring device sends monitoring video of the target area to the monitoring platform. The monitoring platform performs target detection on the monitoring video of the target area acquired by the first monitoring device. When a target object is detected in the image of the monitoring video, the platform further determines the value information of the target object and judges whether the value information of the target object meets the value criteria.

[0041] In both of the above embodiments, when the first monitoring device or monitoring platform determines that the value information of the target object does not meet the value condition, that is, the value of the target object is low, the value information of the target object in the subsequent monitoring video will no longer be analyzed; when the first monitoring device or monitoring platform determines that the value information of the target object meets the value condition, that is, the value of the target object is high, the value information of the target object corresponding to each frame of the subsequent monitoring video can be continuously analyzed.

[0042] In some implementations, the value information of the target object may be the value level of the target object, the value score of the target object, etc.

[0043] In one scenario, the value information of the target object is its value level. A higher value level indicates a higher value, and a lower value level indicates a lower value. In this scenario, the value condition includes: the target object's value level is higher than a predetermined value level. The predetermined value level can be set according to actual needs. For example, a first, second, third, and fourth value level are pre-designed, with the target object's value level being one of these four levels. The predetermined value level can be the second value level; that is, if the target object's value level is higher than the second value level, the value condition is considered met; if the target object's value level is lower than or equal to the second value level, the value condition is considered not met.

[0044] In another scenario, the value information of the target object is its value score. A higher value score indicates a higher value, and a lower value score indicates a lower value. In this scenario, the value condition includes: the target object's value score is greater than or equal to a first score threshold. This first score threshold can be set according to actual needs, such as 0.7.

[0045] In this scenario, a second scoring threshold can be set, which is lower than the first scoring threshold, such as a second scoring threshold of 0.5. The first scoring threshold serves as the trigger threshold, and the second scoring threshold serves as the deactivation threshold. When the target object's value score is greater than or equal to the first scoring threshold, collaborative monitoring by the second monitoring device in the collaborative device group is triggered; when the target object's value score is less than or equal to the second scoring threshold, collaborative monitoring by the second monitoring device in the collaborative device group is terminated. By setting both the first and second scoring thresholds, frequent triggering or termination of collaborative monitoring by the second monitoring device in the collaborative device group can be avoided.

[0046] In one embodiment, the value information of the target object is determined using reference information. The reference information includes at least one of the following: the type of the target object, the pose of the target object, the behavior of the target object, and the clarity of the surveillance footage of the target object. Different types, poses, or behaviors of the target object correspond to different values. For example, the type of the target object can be one of a face, a body, a license plate, etc.; the pose of the target object can include its yaw angle, pitch angle, etc.; and the behavior of the target object can be loitering, lingering, running, etc. Different clarity of the acquired surveillance footage of the target object also corresponds to different values.

[0047] In this implementation, the acquired reference information can be input into the value assessment model, which then uses the reference information to obtain the value information of the target object. The value assessment model can be deployed in monitoring equipment or a monitoring platform; either the monitoring equipment can run the value assessment model to obtain the value information of the target object, or the monitoring platform can run the value assessment model to obtain the value information of the target object.

[0048] In one specific implementation, the acquired reference information includes the target object's type, pose, behavior, and image clarity, and the target object's value information includes its value score. The steps for acquiring the target object's value information through the value assessment model include: acquiring the target type weight corresponding to the target object's type, the pose score corresponding to the target object's pose, the clarity score corresponding to the clarity of the monitoring image of the target object, and the behavior gain factor corresponding to the target object's behavior; and determining the target object's value score using the target type weight, pose score, clarity score, and behavior gain factor.

[0049] In one example, the target type weight corresponding to the type of the target object is determined based on the mapping relationship between type and weight. This mapping relationship includes several different preset types and the type weight corresponding to each preset type. The target type weight corresponding to the type of the target object can be selected from this mapping relationship. For example, the preset types included in the type-weight mapping relationship include face, license plate, and human body. The type weight corresponding to face is 0.8; the type weight corresponding to license plate is 0.9; and the type weight corresponding to human body is 0.5.

[0050] In one example, a lightweight neural network, such as a CNN (Convolutional Neural Network), can be invoked to estimate the pose (yaw angle, pitch angle) of the target object, and then the pose score of the target object can be determined based on the pose of the target object.

[0051] In one example, the sharpness score corresponding to the clarity of the monitored image of the target object can be calculated using the Laplacian variance algorithm. The Laplacian variance algorithm is used to detect image edges and texture details. The sharper the image and the richer the details, the larger the Laplacian variance value, and the higher the corresponding sharpness score. Conversely, the blurrier the image, the smaller the Laplacian variance value, and the lower the corresponding sharpness score. For details on the Laplacian variance algorithm, please refer to known techniques; further explanation is not provided here.

[0052] In one example, the behavior gain factor corresponding to the target object is determined based on a behavior gain mapping relationship. The behavior gain mapping relationship may include several preset behaviors and the value range or behavior gain factor corresponding to each preset behavior. The target object's behavior is one of these preset behaviors. The value range of the behavior gain factor corresponding to the target object's behavior can be selected from the behavior gain mapping relationship, and any value within that range can be chosen as the behavior gain factor corresponding to the target object's behavior. Alternatively, the behavior gain factor corresponding to the target object's behavior can be directly selected from the behavior gain mapping relationship. For example, the behavior gain factor corresponding to normal walking is 1.0; the behavior gain factor range for loitering is 1.3–1.5; the behavior gain factor range for running or rapid movement is 1.5–1.8; and the behavior gain factor range for obviously abnormal behaviors such as climbing over or intruding is 1.8–2.0.

[0053] In one example, after determining the target type weight, pose score, clarity score, and behavior gain factor, the target object's value score is further determined using the following formula: S=W_type×(α×S_clarity+β×S_pose)×G_behavior Where S represents the value score of the target object; W_type represents the target type weight; α represents the weight corresponding to the clarity score, β represents the weight corresponding to the pose score, for example, α is 0.4 and β is 0.6; G_behavior represents the behavior gain factor.

[0054] The aforementioned value assessment model is not a fixed formula, but a dynamic model with online learning capabilities. Specifically, W_type, α, and β, as weight parameters of the value assessment model, can be optimized and adjusted through feedback data. The monitoring platform can receive feedback data from users, including positive and negative feedback samples. For example, when a high-value capture (such as a face with a value score > 0.8) is subsequently manually marked as a false alarm or an irrelevant person, the monitoring platform will mark it as a negative feedback sample and record the context information of this misjudgment. As another example, when a user marks a video segment as important, analyzes it for an extended period, or downloads it, the monitoring platform will mark it as a positive feedback sample and record the context information of that video segment. The aforementioned context information may include the monitoring scene (such as a parking lot entrance), time information (such as nighttime), the type of target object (such as a face), the model parameters or model parameter snapshots of the value assessment model used, and the value score of the target object corresponding to each frame of the video. The monitoring platform can periodically (such as weekly) or when the accumulated amount of feedback data reaches a certain threshold, using the feedback data to adjust the weight parameters of the value assessment model. Afterwards, the monitoring platform can use the optimized value assessment model to analyze the value information of the target object. Alternatively, the monitoring platform can distribute the weight parameters of the optimized value assessment model to each monitoring device, which can then use the optimized value assessment model to analyze the value information of the target object.

[0055] In one example, in a perimeter monitoring scenario, the initial setting of W_type[human body] is 0.5. However, during nighttime hours, users repeatedly marked videos of "human figures climbing over walls from behind" as important. After collecting these positive feedback samples, the monitoring platform automatically increased the target type weight W_type[human body] in the "perimeter-night" scenario to 0.75 using an online learning algorithm (such as stochastic gradient descent), and fine-tuned the weight α (clarity weight) corresponding to the clarity score. This made the optimized value assessment model more accurate in scoring the value of intruding human bodies in this scenario, thereby increasing the sensitivity of triggering collaborative monitoring. The details of the online learning algorithm can be found in known technologies and will not be elaborated upon here.

[0056] In another example, in a lobby monitoring scenario, cleaners with clear frontal faces were repeatedly identified as high-value targets, but were subsequently flagged as false positives by users. After collecting these negative feedback samples, the monitoring platform automatically introduced a scene suppression factor for face types in the lobby scenario to adjust the target type weight W_type (e.g., changing it from 0.8 to 0.3). This made the optimized value assessment model more accurate in this scenario, reducing the occurrence of misclassifying non-suspicious, ordinary people as high-value targets.

[0057] Furthermore, when it is determined that a target object exists in the target area and the value information of the target object meets the value condition, a collaborative device group corresponding to the target area is determined. In this embodiment, the collaborative device group corresponding to the target area includes a first monitoring device and at least one third monitoring device, and the monitoring range of each third monitoring device includes the target area.

[0058] S32: Select at least one second monitoring device from at least one third monitoring device included in the determined collaborative device group.

[0059] In step S32, at least one second monitoring device is selected as a cooperative monitoring device of the first monitoring device.

[0060] In one embodiment, at least one third monitoring device is arbitrarily selected from at least one third monitoring device included in the determined cooperative device group as at least one second monitoring device.

[0061] In another embodiment, selecting at least one second monitoring device from at least one third monitoring device included in the determined collaborative device group further includes the following steps: Step 1: For each third monitoring device, obtain a score for the third monitoring device using at least one evaluation factor of the third monitoring device.

[0062] Among them, at least one evaluation factor includes the reciprocal of the expected rotation time of the third monitoring device, the reciprocal of the load data of the third monitoring device, and at least one of the historical collaborative monitoring task execution data of the third monitoring device.

[0063] The estimated rotation time of the third monitoring device refers to the time required for the third monitoring device to rotate from its current pose to its preset pose. When the third monitoring device is in the corresponding preset pose, its monitoring screen includes the target area. The reciprocal of the estimated rotation time of the third monitoring device is directly proportional to its score. That is, the longer the estimated rotation time of the third monitoring device, the smaller its reciprocal, and the lower its score; conversely, the smaller the estimated rotation time, the larger its reciprocal, and the higher its score.

[0064] The load data of the third monitoring device can be its processor utilization, network utilization, or the number of tasks currently being processed by its processor. The reciprocal of the load data is directly proportional to the device's score. That is, the higher the load data, the lower the reciprocal, and the lower the score; conversely, the lower the load data, the higher the reciprocal, and the higher the score.

[0065] Historical collaborative monitoring task execution data for the third-party monitoring equipment can be either the number of successful executions of historical collaborative monitoring tasks in which the third-party monitoring equipment previously participated, or the quality score of those tasks. The quality score is assessed based on factors such as the clarity of the monitoring footage and the value of the target object in the video. Higher footage clarity results in a higher quality score. The historical collaborative monitoring task execution data is directly proportional to the third-party monitoring equipment's score. In other words, the more historical collaborative monitoring task execution data the third-party monitoring equipment has, the higher its score; conversely, the less historical collaborative monitoring task execution data the third-party monitoring equipment has, the lower its score.

[0066] In one specific implementation, each evaluation factor corresponds to a target weight, and the score of the third monitoring device is obtained by weighting each evaluation factor using the target weight of each evaluation factor.

[0067] For example, at least one evaluation factor includes the reciprocal of the estimated turnaround time of the third monitoring device, the reciprocal of the load data of the third monitoring device, and the historical collaborative monitoring task execution data of the third monitoring device. The score of the third monitoring device can be determined by the following formula: Score = K1 × (1 / Estimated Turnaround Time) + K2 × (1 / Load Data) + K3 × Historical Monitoring Task Execution Status. Wherein, K1, K2, and K3 are the target weights corresponding to the reciprocal of the estimated turnaround time, the reciprocal of the load data, and the historical collaborative monitoring task execution data, respectively. For example, K1, K2, and K3 are 0.5, 0.3, and 0.2, respectively.

[0068] In one specific implementation, the target weight of each evaluation factor is determined based on the monitoring scenario of the collaborative device group; that is, each evaluation factor has a corresponding exclusive target weight under each monitoring scenario. In this implementation, the step of obtaining the target weight of each evaluation factor further includes: selecting a weighted set corresponding to the monitoring scenario of the collaborative device group from the weight mapping relationship; and using the weights of each evaluation factor in the selected weighted set as the target weight of each evaluation factor. The weight mapping relationship includes several preset monitoring scenarios and preset weighted sets corresponding to each preset monitoring scenario, with each preset weighted set including the preset weights corresponding to each evaluation factor.

[0069] In this implementation, for each preset monitoring scenario, obtaining the preset weighted reassembly corresponding to the preset monitoring scenario further includes the following sub-steps: Sub-step one: Generate several candidate weight reorganizations.

[0070] Among them, the candidate weight reorganization includes the candidate weights corresponding to each evaluation factor.

[0071] In one example, several candidate weight reshuffles are generated randomly.

[0072] In another example, each evaluation factor has a corresponding weight range and search step size. For each evaluation factor, several candidate weights can be obtained by searching within its weight range according to the search step size. Combining these candidate weights for each evaluation factor yields several candidate weight reorganizations. Furthermore, the candidate weights of each evaluation factor in each candidate weight reorganization can be normalized so that the sum of the candidate weights of each evaluation factor in each candidate weight reorganization is 1.

[0073] For example, the weights of K1, K2, and K3 all have a range of [0.1, 0.7] and a search step size of 0.1. Alternatively, two of the weights of K1, K2, and K3 can be obtained through a search, and the remaining weight can be calculated. For instance, the weights of K1 and K2 can be set to a range of [0.1, 0.7], and a search can be performed with a search step size of 0.1. K3 can then be calculated using the formula (K3 = 1 - K1 - K2).

[0074] Sub-step two: For each candidate weight reorganization, in at least one test collaborative monitoring task under the preset monitoring scenario, the collaborative monitoring device for each test collaborative monitoring task is selected by using the candidate weight reorganization. For each test collaborative monitoring task, the task parameters of the test collaborative monitoring task are determined when the collaborative monitoring device is used to execute the test collaborative monitoring task. And, using the task parameters of each test collaborative monitoring task, the performance score corresponding to the candidate weight reorganization is determined.

[0075] In the above sub-step two, for each candidate weight reorganization: at least one test collaborative monitoring task is performed under a preset monitoring scenario. In each test collaborative monitoring task under the preset monitoring scenario, the score of each monitoring device is calculated using candidate weight reorganization to select the collaborative monitoring device participating in the test collaborative monitoring task (the content of calculating the score to select the collaborative monitoring device can be referred to the previous introduction, which is briefly summarized here). Then, the collaborative monitoring device of the test collaborative monitoring task is used to collaboratively execute the test collaborative monitoring task and determine the task parameters of the test collaborative monitoring task. In this way, the task parameters of each test collaborative monitoring task can be obtained, and then the performance score corresponding to the candidate weight reorganization can be determined using the task parameters of each test collaborative monitoring task.

[0076] The task parameters for a collaborative monitoring test include at least one of the following: task duration, the overall value score of the test object, and the total rotational loss of all participating monitoring devices. The task duration is the time from initiation to termination of the collaborative monitoring task. The test object refers to the target object monitored in the collaborative monitoring task. The overall value score of the test object can be the highest value score, average value score, etc., obtained from the test object during the collaborative monitoring task. The total rotational loss of all participating monitoring devices refers to the sum of the gimbal rotational losses of all participating monitoring devices during the execution of the collaborative monitoring task.

[0077] In one example, the task parameters for each test collaboration monitoring task include the task duration, the comprehensive value score of the test object in the test collaboration monitoring task, and the total rotational loss of all participating monitoring devices in the test collaboration monitoring task. The steps for determining the performance score corresponding to candidate weight reorganization using the task parameters of each test collaboration monitoring task include: The average task duration of each test collaboration monitoring task is determined to obtain the average task duration; the average comprehensive value score of each test collaboration monitoring task is determined to obtain the average comprehensive value score; the average total rotation loss of each test collaboration monitoring task is determined to obtain the average rotation loss; and the performance score corresponding to candidate weight reorganization is obtained by combining the average task duration, average comprehensive value score, and average rotation loss. Specifically, the performance score corresponding to candidate weight reorganization can be calculated using the following formula: Performance Score = w1 × (1 / Average Task Duration) + w2 × Average Comprehensive Value Score – w3 × Average Rotation Loss. Where w1, w2, and w3 are pre-set weighting coefficients. For example, w1, w2, and w3 are 0.4, 0.3, and 0.3, respectively.

[0078] Sub-step three: Select the weight recombination with the highest performance score from several candidate weight recombinations, and use it as the preset weight recombination corresponding to the preset monitoring scenario.

[0079] Step 2: Select at least one monitoring device from at least one third monitoring device that meets the scoring requirements, and use it as at least one second monitoring device.

[0080] In one example, there is only one second monitoring device participating in the collaborative monitoring, and the monitoring device with the highest score is selected from at least one third monitoring device as the second monitoring device.

[0081] In another example, multiple monitoring devices ranked in the top preset positions in the score ranking are selected from at least one third monitoring device and designated as multiple second monitoring devices. The preset positions can be set according to actual needs. For example, two monitoring devices ranked in the top two in the score ranking are selected as two second monitoring devices.

[0082] S33: Send the first cooperation command to each of the second monitoring devices respectively.

[0083] In one embodiment, the monitoring platform sends a first cooperation instruction to each of the second monitoring devices, further including the following steps: Step 1: Identify the collaborative equipment group corresponding to the target area.

[0084] Step 2: For each second monitoring device, select the preset pose of the second monitoring device from the cooperative mapping relationship corresponding to the determined cooperative device group.

[0085] The collaborative mapping relationship includes the preset poses of each monitoring device in the collaborative device group, and the monitoring screen of each monitoring device in the collaborative device group includes the target area when it is in the corresponding preset pose.

[0086] Step 3: For each second monitoring device, send a first cooperation instruction carrying the preset pose of the second monitoring device.

[0087] After receiving the first cooperation instruction carrying the preset pose of the second monitoring device, the second monitoring device rotates from its current pose to the corresponding preset pose in order to synchronously acquire monitoring video of the target object.

[0088] In another embodiment, the first cooperation instruction sent by the monitoring platform to each second monitoring device does not carry the preset pose of the corresponding second monitoring device. For each second monitoring device, after receiving the first cooperation instruction, the second monitoring device retrieves the cooperation mapping relationship corresponding to the cooperation device group of the target area from its own storage unit, selects the preset pose of the second monitoring device from the cooperation mapping relationship, and rotates from the current pose to the corresponding preset pose.

[0089] In this embodiment, when the surveillance video of the target area acquired by the first monitoring device contains a target object and the value information of the target object meets the value condition, at least one second monitoring device whose monitoring range includes the target area can be determined from the cooperating device group corresponding to the target area. A first cooperation instruction is then sent to each of the second monitoring devices to instruct them to synchronously acquire surveillance video of the target area. This approach, on the one hand, enables synchronous acquisition of surveillance video of the target area from multiple perspectives through the collaborative monitoring of at least one second monitoring device, improving the reliability of monitoring key target objects; on the other hand, the collaborative monitoring of at least one second monitoring device is only triggered when the value information of the target object meets the value condition, allowing for on-demand allocation of monitoring resources and reducing unnecessary collaborative monitoring.

[0090] Optionally, in this embodiment, the collaborative mapping relationship of each collaborative device group can be calibrated to maintain the accuracy of the collaborative mapping relationship of the collaborative device group. Specifically, for each collaborative device group, when it is determined that the calibration conditions are met, the collaborative mapping relationship of the collaborative device group is calibrated. The calibration conditions include any of the following: the time elapsed since the last calibration reaches a preset time; each monitoring device in the collaborative device group restarts; or a fifth monitoring device in the collaborative device group receives a collaborative instruction and the monitoring screen of the fifth monitoring device does not include the target area.

[0091] In one scenario, the calibration condition is that the time elapsed since the last calibration reaches a preset duration. That is, the collaborative mapping relationship of the collaborative device group can be periodically calibrated at preset intervals. For example, the preset duration is 24 hours.

[0092] In another scenario, the calibration condition is that each monitoring device in the collaborative device group is restarted. That is, the collaborative mapping relationship of the collaborative device group can be calibrated after each monitoring device in the collaborative device group is restarted.

[0093] In another scenario, the calibration condition is: there is a fifth monitoring device in the collaborative device group that has received a collaborative instruction, and the monitoring screen of the fifth monitoring device does not include the target area. The fifth monitoring device that receives the collaborative instruction is the monitoring device that needs to participate in collaborative monitoring. If the fifth monitoring device does not include the target object in its monitoring screen when it is in the corresponding preset pose, i.e., the target object cannot be seen, then the collaborative task is considered to have failed, and the collaborative mapping relationship of the collaborative device group needs to be adjusted.

[0094] In one embodiment, when any of the above calibration conditions are met, the cooperative mapping relationship of the cooperative device group can be calibrated. In this embodiment, the calibration process further includes the following sub-steps: Sub-step one: Control each monitoring device in the collaborative equipment group to be in its corresponding preset position.

[0095] Sub-step two: Identify the reference monitoring equipment and the monitoring equipment to be calibrated in the collaborative equipment group.

[0096] The reference monitoring device can be a monitoring device whose monitoring screen includes the target area when it is in a corresponding preset pose, or a monitoring device whose target object is located exactly in the center of the monitoring screen when it is in a corresponding preset pose. The monitoring device to be calibrated can be a monitoring device whose monitoring screen does not include the target object when it is in a corresponding preset pose, or whose target object is located in the edge area of ​​the monitoring screen.

[0097] Sub-step three: Obtain the reference image captured by the reference monitoring device and the calibration image captured by the monitoring device to be calibrated.

[0098] Sub-step four: Extract several first feature points from the reference image and several second feature points from the image to be calibrated.

[0099] For example, feature points in an image can be extracted using algorithms such as SIFT and ORB.

[0100] Sub-step five involves performing feature point matching on several first feature points and several second feature points to obtain at least one matched feature point pair.

[0101] A feature point pair includes a first feature point and a second feature point that are matched. It can be understood that the feature point matching process involves finding the correspondence between feature points in the reference image and feature points in the image to be calibrated.

[0102] The extraction of image feature points and the matching of feature points in sub-steps four and five can be found in known techniques and will not be explained further here.

[0103] Sub-step six: For each feature point pair, obtain the reference pixel coordinates of the first feature point in the reference image, obtain the actual pixel coordinates of the second feature point in the image to be calibrated, and determine the position offset using the reference pixel coordinates of the first feature point and the actual pixel coordinates of the second feature point in the feature point pair.

[0104] Sub-step seven: Determine the overall position offset by using the position offset of each feature point pair.

[0105] For example, the average positional offset of each feature point pair is obtained as the comprehensive positional offset.

[0106] Sub-step eight: If the overall position offset is less than the preset position offset, then the preset pose of the monitoring device to be calibrated will not be adjusted.

[0107] Sub-step nine: If the overall position offset is greater than or equal to the preset position offset, adjust the preset pose of the monitoring device to be calibrated to obtain a new preset pose corresponding to the device to be calibrated.

[0108] After executing sub-step nine, return to sub-step three and re-execute sub-step three and its subsequent steps until the latest acquired comprehensive position offset is less than the preset position offset.

[0109] Optionally, in this embodiment, the monitoring platform can also generate a summary video for the monitoring video acquired by the first monitoring device or the second monitoring device during the collaborative period. Specifically, a number of image frames whose value information of the target object meets the conditions are selected from the monitoring video acquired by the first or second monitoring device during the collaborative period, and the summary video is obtained using the selected image frames.

[0110] In some examples, image frames with a target object's value score greater than a third score threshold can be filtered. Alternatively, image frames with the target object having the highest value level can also be filtered. For example, the third score threshold could be 0.75, 0.8, etc.

[0111] In some examples, after filtering to obtain several image frames, these image frames can be arranged in chronological order or in order of value score to obtain a summary video.

[0112] By generating summary videos, users can quickly learn about relevant information about the target object during the collaboration period without having to browse all the surveillance videos within that period.

[0113] Optionally, in this embodiment, the image with the highest value information of the target object in the surveillance video acquired by the first monitoring device and each of the second monitoring devices during the collaborative period is used as the preview image for the collaborative period. For example, the image with the highest value score of the target object is used as the preview image for the collaborative period. When a user previews the recording during this collaborative period through the client, this preview image can be displayed, thereby providing the user with a preview image of higher value.

[0114] Optionally, in this embodiment, the client can associate and display the surveillance videos acquired by the first device and the second device during the collaborative period. The first device and the second device are monitoring devices participating in the same collaborative monitoring task. For example, the first device is the aforementioned first monitoring device, and the second device is the aforementioned second monitoring device. Alternatively, the first device is the aforementioned second monitoring device, and the second device is the aforementioned first monitoring device.

[0115] In one scenario, the monitoring platform can record the collaborative relationships between monitoring devices participating in the same collaborative monitoring task. For example, a first device requests a second device to collaborate on monitoring, indicating a collaborative relationship between the first and second devices. Furthermore, when the monitoring platform receives a video playback command from a client regarding the first device, it can synchronously push the monitoring videos acquired by the first and second devices during the collaborative period to the client based on the recorded collaborative relationships. At this time, the client can simultaneously play the monitoring videos acquired by the first and second devices during the collaborative period on its display interface. For example, the client's display interface can present a multi-screen playback effect with consistent playback progress. For instance, the top of the display interface shows the playback screen of the monitoring video acquired by the first device during the collaborative period, and the bottom of the display interface shows the playback screen of the monitoring video acquired by the second device during the collaborative period.

[0116] In another scenario, when the monitoring platform receives a command from the client to pause video playback from the first device, it pauses the push of monitoring videos acquired by the first and second devices during the collaborative period to the client. At this time, the client can simultaneously pause playback of the monitoring videos acquired by the first and second devices during the collaborative period on the display interface.

[0117] In the above method, when a user plays the monitoring video of the first device through the client, they can simultaneously watch the monitoring videos of other collaborative monitoring devices, which improves the user's playback experience.

[0118] In addition, the client can mark several high-value moments on the recording timeline of collaborative monitoring tasks. For example, a high-value moment can be a moment when the value score of the relevant target object is greater than the fourth score threshold (such as 0.9, 0.95, etc.), or a high-value moment can be a moment when the value level of the relevant target object is the highest value level. When a user clicks on one of the high-value moments, the monitoring image of that high-value moment can be displayed.

[0119] Optionally, in this embodiment, during the collaborative monitoring process of the first monitoring device and the second monitoring device, it may be further determined whether the target object is about to leave the monitoring range of the first monitoring device and the second monitoring device, or whether there is a usable third monitoring device in the collaborative device group corresponding to the target area.

[0120] For example, the location information of the target object at a future time can be predicted by the target tracking algorithm, and the target object can be determined accordingly whether it is about to leave the monitoring range of the first monitoring device and the second monitoring device.

[0121] For example, the availability of a third monitoring device can be determined based on its load data. For instance, the load data of a third monitoring device might be its processor utilization rate. When the processor utilization rate of a third monitoring device is greater than or equal to a utilization rate threshold, the third monitoring device is determined to be unavailable and cannot participate in collaborative monitoring; when the processor utilization rate of a third monitoring device is less than the utilization rate threshold, the third monitoring device is determined to be available and can participate in collaborative monitoring.

[0122] Furthermore, when it is determined that the target object is about to leave the monitoring range of the first and second monitoring devices, or when it is determined that there is no available third monitoring device in the cooperative device group to which the first monitoring device belongs, the following steps can be performed: Step 1: Predict the new target area where the target object will appear.

[0123] In step one, a target tracking algorithm can be used to predict the movement trajectory of the target object, and based on the movement trajectory of the target object, the new target area where the target object will appear next can be determined.

[0124] Step 2: Select at least one fourth monitoring device from the new collaborative device group corresponding to the new target area.

[0125] The process of selecting at least one fourth monitoring device from the new collaborative device group can be referred to the process of selecting at least one second monitoring device in step S32 above, and will not be repeated here.

[0126] Step 3: Send a takeover command to at least one fourth monitoring device. The takeover command is used to instruct each fourth monitoring device to acquire monitoring video of the new target area.

[0127] Through steps one to three above, when it is determined that the target object is about to leave the monitoring range of the first and second monitoring devices, or when it is determined that there is no available third monitoring device in the cooperative device group to which the first monitoring device belongs, the monitoring of the target object can be taken over by at least one fourth monitoring device, thereby achieving continuous tracking of high-value target objects.

[0128] Optionally, in this embodiment, group events can also be detected, and collaborative monitoring can be performed when a group event is detected. This process may include the following steps: Step 1: In response to determining from the monitoring video of the first monitoring device or the monitoring video of the second monitoring device that a group event related to a group target has occurred, determine the coverage area corresponding to the group event.

[0129] Group events involve group targets, i.e., multiple target objects. The coverage area corresponding to a group event must at least cover the group targets. For example, group events can be events such as area aggregation or linear tracking.

[0130] In one implementation, a group event is determined to have occurred when the triggering conditions for a group event are met. For example, the triggering conditions for a group event include: the number of target objects appearing is greater than a quantity threshold, the duration is greater than a duration threshold, and the spatial range involved is within a preset spatial range. Different triggering conditions can be designed for different group events. The quantity threshold, duration threshold, and preset spatial range can be set according to actual needs.

[0131] In one implementation, when a mass incident is determined to have occurred, an alarm message can be sent to the client so that the user can be promptly informed of the mass incident through the client.

[0132] In one implementation, the triggering conditions for each group event can be adaptively adjusted based on the historical trigger count and the effective alarm rate of the historical alarm information. For example, the monitoring platform records the historical trigger count of the group event and the effective alarm rate of the historical alarm information generated after the trigger (such as whether it is marked as important or a false alarm by the user). If the group event is frequently triggered (e.g., occurring 3 times in 1 day) and the effective alarm rate is low (e.g., below the first alarm rate threshold), the quantity threshold and duration threshold in the triggering conditions can be increased to reduce trigger sensitivity. If the historical trigger count of the group event is low and the effective alarm rate is high (e.g., above the second alarm rate threshold), the quantity threshold and duration threshold in the triggering conditions can be decreased to increase trigger sensitivity.

[0133] Step 2: Determine the monitoring scope to include at least one sixth monitoring device covering the coverage area.

[0134] In step two, at least one seventh monitoring device whose monitoring range includes the coverage area can be determined first. The at least one seventh monitoring device is all monitoring devices whose monitoring range includes the coverage area. Then, at least one sixth monitoring device is selected from the at least one seventh monitoring device.

[0135] In some implementations, at least one seventh monitoring device can be used as at least one sixth monitoring device, meaning that all monitoring devices covering the coverage area participate in collaborative monitoring to synchronously acquire monitoring video of the coverage area from different directions as much as possible, forming a comprehensive monitoring network. Alternatively, the process of selecting at least one second monitoring device in step S32 above can be used to select at least one sixth monitoring device from at least one seventh monitoring device, which will not be elaborated here.

[0136] Step 3: Send the second cooperation command to each of the sixth monitoring devices.

[0137] The second collaboration instruction is used to instruct the sixth monitoring device to synchronously acquire surveillance video of the coverage area. Upon receiving the second collaboration instruction from the monitoring platform, each sixth monitoring device begins synchronously acquiring surveillance video of the coverage area.

[0138] Optionally, in this embodiment, when a group event is detected, the value information of all target objects involved in the group event can be set to the highest value, such as setting the value level to the highest value level and the value score to the highest value score (e.g., 1), to ensure that the relevant video recordings of the group event can be processed with priority in the future.

[0139] Please see Figure 4 , Figure 4 This is a flowchart illustrating another embodiment of the monitoring method provided in this application, which is executed by a first monitoring device. Figure 4 As shown, the method includes the following steps: S41: When a target object is detected in the target area from the surveillance video, determine the value information of the target object.

[0140] S42: When the value information of the target object meets the value condition, a cooperation request is sent to the monitoring platform; wherein, the cooperation request is used to instruct the monitoring platform to identify at least one second monitoring device and send a first cooperation instruction to each second monitoring device respectively, the monitoring range of each second monitoring device includes the target area, and the first cooperation instruction is used to instruct the second monitoring device to synchronously acquire monitoring video of the target area.

[0141] For details of steps S41 and S42, please refer to the foregoing. Figure 2 or Figure 3 The illustrated embodiment is omitted in detail here.

[0142] Please see Figure 5 , Figure 5 This is a flowchart illustrating another embodiment of the monitoring method provided in this application, which is executed by at least one second monitoring device. Figure 5 As shown, for each second monitoring device, the method includes the following steps: S51: When the first cooperation instruction is received, the monitoring video of the target area is acquired; wherein, the first cooperation instruction is sent by the monitoring platform when it determines that the target object exists in the monitoring video of the target area acquired by the first monitoring device and the value information of the target object meets the value condition, and the monitoring range of each second monitoring device includes the target area.

[0143] For details of step S52, please refer to the foregoing. Figure 2 or Figure 3 The illustrated embodiment is omitted in detail here.

[0144] Please see Figure 6 , Figure 6 This is a schematic diagram of a framework of an embodiment of the electronic device provided in this application. In this embodiment, the electronic device 60 includes a memory 61 and a processor 62.

[0145] Processor 62 can also be referred to as a CPU (Central Processing Unit). Processor 62 may be an integrated circuit chip with signal processing capabilities. Processor 62 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 gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor, or processor 62 can be any conventional processor 62, etc.

[0146] The memory 61 in the electronic device 60 is used to store the program instructions required for the processor 62 to run.

[0147] The processor 62 is used to execute program instructions to implement the monitoring method in this application.

[0148] Please see Figure 7 , Figure 7 This is a schematic diagram of a framework of an embodiment of the computer-readable storage medium provided in this application. The computer-readable storage medium 70 of this application embodiment stores program instructions 71, which, when executed, implement the monitoring method provided in this application. The program instructions 71 can form a program file and be stored in the aforementioned computer-readable storage medium 70 in the form of a software product, so that a computer device (which may be a personal computer, server, or network device, etc.) can execute all or part of the steps of the methods of various embodiments of this application. The aforementioned computer-readable storage medium 70 includes various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, or terminal devices such as computers, servers, mobile phones, and tablets.

[0149] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0150] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to, and for the sake of brevity, they will not be repeated here.

[0151] In the several embodiments provided in this application, it should be understood that the disclosed methods, apparatuses, and systems can be implemented in other ways. For example, the apparatus implementations described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or units may be electrical, mechanical, or other forms.

[0152] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0153] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0154] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0155] It should be noted that if the technical solution of this application involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, the personal information processing rules are clearly informed through signs / information, and authorization is obtained through pop-up information or by asking the individual to upload their personal information. The personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

[0156] This application provides only the technology. The specific data required for the application shall be subject to the actual situation. However, users are advised that when using this technology, the collection and processing of data should comply with laws and regulations related to data and personal information protection.

[0157] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A monitoring method characterized by, Applied to a monitoring platform, the method includes: In response to the presence of a target object in the surveillance video of the target area obtained by the first monitoring device, and the value information of the target object meeting the value condition, at least one second monitoring device is determined; wherein the monitoring range of each second monitoring device includes the target area; A first cooperation instruction is sent to each of the second monitoring devices, the first cooperation instruction being used to instruct the second monitoring devices to synchronously acquire monitoring video of the target area.

2. The method according to claim 1, characterized in that, The value information of the target object is determined using reference information, which includes at least one of the target object's type, the target object's posture, the target object's behavior, and the clarity of the monitoring footage of the target object. And / or, the value information of the target object is the value score of the target object, and the value condition includes: the value score of the target object is greater than or equal to a first score threshold.

3. The method of claim 1, wherein, The determination of at least one second monitoring device includes: Determine the collaborative device group corresponding to the target area; wherein, the collaborative device group includes the first monitoring device and at least one third monitoring device, and the monitoring range of each of the third monitoring devices includes the target area; The at least one second monitoring device is selected from the at least one third monitoring device.

4. The method of claim 3, wherein, Selecting the at least one second monitoring device from the at least one third monitoring device includes: For each of the aforementioned third monitoring devices, a score for the third monitoring device is obtained using at least one evaluation factor of the third monitoring device; wherein, the at least one evaluation factor includes at least one of the following: the reciprocal of the estimated rotation time of the third monitoring device, the reciprocal of the load data of the third monitoring device, and the historical collaborative monitoring task execution data of the third monitoring device; the estimated rotation time of the third monitoring device refers to the time required for the third monitoring device to rotate from its current pose to its preset pose; when the third monitoring device is in the preset pose, the monitoring screen of the third monitoring device includes the target area. At least one monitoring device whose score meets the requirements is selected from the at least one third monitoring device and is selected as the at least one second monitoring device.

5. The method according to claim 4, characterized in that, The score of the third monitoring device is obtained by weighting each evaluation factor using the target weight of each evaluation factor. The step of obtaining the target weight of each evaluation factor includes: Select a weighted set corresponding to the monitoring scenario of the collaborative device group from the weighted mapping relationship; wherein, the weighted mapping relationship includes a number of preset monitoring scenarios and preset weighted sets corresponding to each preset monitoring scenario, and the preset weighted sets include preset weights corresponding to each evaluation factor; The weights of each evaluation factor in the selected weighted reorganization are respectively used as the target weights of each evaluation factor.

6. The method of claim 5, wherein, For each of the preset monitoring scenarios, the step of obtaining the preset weight reassembly corresponding to the preset monitoring scenario includes: A number of candidate weight reorganizations are generated; wherein, the candidate weight reorganization includes candidate weights corresponding to each of the evaluation factors; For each candidate weight reorganization, in at least one test collaborative monitoring task under the preset monitoring scenario, the candidate weight reorganization is used to select the collaborative monitoring device for each test collaborative monitoring task. For each test collaborative monitoring task, the task parameters of the test collaborative monitoring task are determined when the collaborative monitoring device executes the test collaborative monitoring task. Furthermore, the performance score corresponding to the candidate weight reorganization is determined using the task parameters of each test collaborative monitoring task. The task parameters of the test collaborative monitoring task include at least one of the following: the task duration of the test collaborative monitoring task, the comprehensive value score of the test object in the test collaborative monitoring task, and the total rotational loss of all participating monitoring devices in the test collaborative monitoring task. The weighted recombination with the highest performance score is selected from the candidate weighted recombinations and used as the preset weighted recombination corresponding to the preset monitoring scenario.

7. The method of claim 3, wherein, The method further includes: In response to determining that the target object is about to leave the monitoring range of the first monitoring device and the second monitoring device, or determining that there is no available third monitoring device in the cooperative device group corresponding to the target area, a new target area for the target object is predicted to appear. Select at least one fourth monitoring device from the new collaborative device group corresponding to the new target area; A takeover command is sent to the at least one fourth monitoring device, the takeover command being used to instruct each of the fourth monitoring devices to acquire monitoring video of the new target area.

8. The method of claim 1, wherein, Sending the first cooperation instruction to each of the second monitoring devices includes: Identify the collaborative device group corresponding to the target area; For each of the second monitoring devices, a preset pose of the second monitoring device is selected from the cooperation mapping relationship corresponding to the cooperation device group, and a first cooperation instruction carrying the preset pose of the second monitoring device is sent to the second monitoring device. The collaborative mapping relationship includes the preset poses corresponding to each monitoring device in the collaborative device group, and the monitoring screen of each monitoring device in the collaborative device group includes the target area when it is in the corresponding preset pose.

9. The method of claim 8, wherein, The method further includes: In response to meeting the calibration conditions, the cooperative mapping relationship corresponding to the cooperative device group is calibrated. The calibration conditions are any of the following: the time elapsed since the last calibration reaches a preset time, each monitoring device in the collaborative device group restarts, or when there is a fifth monitoring device in the collaborative device group that has received a collaborative instruction and the fifth monitoring device is in the corresponding preset pose, the monitoring screen does not include the target object.

10. The method according to claim 1, characterized in that, The method further includes: The image with the highest value information in the monitoring video acquired by the first monitoring device and each of the second monitoring devices during the cooperation period is used as the preview image of the cooperation period. And / or, in response to receiving a video playback command from a client regarding the first device, synchronously push the monitoring videos acquired by the first device and the second device during the collaboration period to the client; in response to receiving a video pause playback command from the client regarding the first device, pause pushing the monitoring videos acquired by the first device and the second device during the collaboration period to the client; wherein, the first device and the second device are monitoring devices participating in the same collaborative monitoring task.

11. The method of claim 1, wherein, The method further includes: In response to determining from the monitoring video of the first monitoring device or the monitoring video of the second monitoring device that a group event related to a group target has occurred, the coverage area corresponding to the group event is determined; The monitoring range is defined to include at least one sixth monitoring device covering the coverage area; A second cooperation instruction is sent to each of the sixth monitoring devices, the second cooperation instruction being used to instruct the sixth monitoring device to synchronously acquire monitoring video of the coverage area.

12. A monitoring method characterized by, Applied to a first monitoring device, the method includes: In response to detecting the presence of a target object in a target area from surveillance video, the value information of the target object is determined; In response to the target object's value information meeting the value criteria, a cooperation request is sent to the monitoring platform; The collaboration request is used to instruct the monitoring platform to identify at least one second monitoring device and send a first collaboration instruction to each of the second monitoring devices. The monitoring range of each second monitoring device includes the target area, and the first collaboration instruction is used to instruct the second monitoring device to synchronously acquire monitoring video of the target area.

13. A monitoring method characterized by, The method, applied to at least one second monitoring device, wherein the monitoring range of each second monitoring device includes the target area, comprises: In response to receiving the first cooperation instruction, acquire surveillance video of the target area; The first collaboration instruction is sent by the monitoring platform when it determines that a target object exists in the monitoring video of the target area obtained by the first monitoring device, and the value information of the target object meets the value condition.

14. A monitoring system, characterized by The monitoring system includes a monitoring platform and several monitoring devices, and the monitoring system is used to implement the monitoring method according to any one of claims 1 to 13.

Images