Monitoring method and system based on linkage of multiple cameras

Abstract

The invention relates to a monitoring method and system based on linkage of multiple cameras, and relates to networking of the multiple cameras, when any camera monitors a target and does not receive coordinate information shared by other cameras, the camera serves as a main device to execute target tracking, distribute a unique ID and monitor coordinates and tracks of the target in real time, and information is shared through a star flash protocol. After the other cameras receive the coordinate information shared by the master device, the other cameras serve as slave devices to execute preparation tasks, monitor suspected targets and match the information shared by the master device, if yes, it is confirmed that the targets enter the overlapped monitoring range, a tracking handover request is sent to the master device, and the master device stops tracking after receiving the request; after the slave device receives the confirmation information, the slave device is changed into the master device to continue to track the task. According to the invention, comprehensive monitoring and efficient tracking of the target can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring systems, and in particular to a monitoring method and system based on the linkage of multiple cameras. Background Art

[0002] A security surveillance system is a complex system of integrated hardware and software designed to monitor and record activities in a specific area or place to prevent and detect illegal behavior, accidents or other security threats.

[0003] The security monitoring system is designed to achieve comprehensive monitoring and efficient tracking. To achieve coverage without blind spots, there are currently two main methods: one is to increase the number of cameras, and the other is to use cross-camera multi-target tracking technology. Among them, increasing the number of cameras is simple to operate, but may cause information overload and increased operating costs; cross-camera multi-target tracking technology achieves automated monitoring through advanced algorithms such as deep learning. Although it improves monitoring efficiency, it has high requirements on computing resources and operation and maintenance costs.

[0004] In addition, the current security monitoring system mainly uses WI-FI wireless communication. The advantages are easy installation and maintenance, convenient expansion and integration, and the disadvantages are high latency, susceptibility to signal interference, stability issues, limited transmission distance and security issues. Summary of the invention

[0005] The purpose of the present invention is to provide a monitoring method and system based on the linkage of multiple cameras to achieve comprehensive monitoring and efficient tracking of targets.

[0006] To achieve the above object, the present invention provides a monitoring method based on the linkage of multiple cameras, wherein the multiple cameras are networked based on the Star Flash protocol, and the method comprises: Multiple cameras for target monitoring; When any camera detects a target and does not receive the coordinate information shared by other cameras, it determines itself as the current master device to perform the target tracking task; when the current master device performs the target tracking task, it assigns a unique ID to the target, monitors the coordinate information of the target in real time, generates the running trajectory of the target according to the coordinate information of the target, and shares the ID and coordinate information of the target with other cameras through the Star Flash communication protocol; When any camera receives the coordinate information of the target shared by the current master device, it determines itself as the current slave device to perform the preparatory task; when the current slave device performs the preparatory task, it monitors the target, and if a suspected target is detected, it matches the coordinate information of the suspected target with the coordinate information of the target shared by the current master device. If they match, it determines that the target enters the overlapping monitoring range between the current master device and the current slave device, monitors the target in real time, and sends a target tracking handover request to the current master device; When the current master device is executing the target tracking task, if it receives the target tracking handover request, it stops executing the target tracking task, changes to the current slave device executing the preparatory task, and sends tracking handover confirmation information to the current slave device that sent the target tracking handover request; When the current slave device is executing the preparation task, if the tracking handover confirmation information is received, the current master device is changed to execute the target tracking task.

[0007] Preferably, matching the coordinate information of the suspected target with the coordinate information of the target shared by the current master device includes: The coordinate information of the suspected target and the coordinate information of the target shared by the current main device are converted into the same reference coordinate system. Based on the same reference coordinate system, it is determined whether the coordinate information of the suspected target is consistent with the coordinate information of the target shared by the current main device. If so, they match; otherwise, they do not match.

[0008] Preferably, the method further comprises: When the current master device performs a target tracking task, it adjusts its own camera angle in real time according to the coordinate information of the target; the multiple cameras are all pan-tilt cameras.

[0009] Preferably, the method further comprises: When the current slave device performs a preparatory task, it predicts the trajectory of the target based on the coordinate information of the target shared by the current master device, and adjusts its own camera angle based on the trajectory prediction result to achieve seamless tracking and handover between different cameras; the multiple cameras are all pan-tilt cameras.

[0010] Preferably, the current slave device predicts the trajectory of the target according to the coordinate information of the target shared by the current master device, including: The current slave devices convert the coordinate information of the target shared by the current master device into their own coordinate systems to obtain the coordinate information of the target based on their own coordinate systems, and predict the trajectory of the target based on the coordinate information of the target based on their own coordinate systems.

[0011] As the same inventive concept, the present invention also provides a monitoring system based on linkage of multiple cameras, comprising multiple cameras, wherein the multiple cameras are networked based on the Star Flash protocol; The camera is used to determine itself as the current master device to perform the target tracking task when it detects the target and does not receive the coordinate information shared by other cameras; The current master device is used to assign a unique ID to the target, monitor the coordinate information of the target in real time, generate the running track of the target according to the coordinate information of the target, and share the ID and coordinate information of the target with other multiple cameras through the Star Flash communication protocol when performing the target tracking task; The camera is used to determine itself as the current slave device to perform the preparatory task when receiving the coordinate information of the target shared by the current master device; The current slave device is used to monitor the target when executing the preparatory task. If a suspected target is detected, the coordinate information of the suspected target is matched with the coordinate information of the target shared by the current master device. If they match, it is determined that the target enters the overlapping monitoring range between the current master device and the current slave device, the target is monitored in real time, and a target tracking handover request is sent to the current master device. The current master device is used to stop executing the target tracking task if it receives the target tracking handover request, change to the current slave device to execute the preparatory task, and send tracking handover confirmation information to the current slave device that sent the target tracking handover request; When the current slave device is executing the preparation task, if it receives the tracking handover confirmation information, it will be changed to the current master device to execute the target tracking task.

[0012] Preferably, the current slave device is used to convert the coordinate information of the suspected target and the coordinate information of the target shared by the current master device into the same reference coordinate system, and based on the same reference coordinate system, determine whether the coordinate information of the suspected target is consistent with the coordinate information of the target shared by the current master device. If so, they match, otherwise, they do not match.

[0013] Preferably, the current master device is used to adjust its own camera angle according to the coordinate information of the target in real time; and the multiple cameras are all pan-tilt cameras.

[0014] Preferably, the current slave device is used to predict the trajectory of the target based on the coordinate information of the target shared by the current master device, and adjust its own camera angle according to the trajectory prediction result to achieve seamless tracking and handover between different cameras; the multiple cameras are all pan-tilt cameras.

[0015] Preferably, the current slave device is used to convert the coordinate information of the target shared by the current master device into its own coordinate system to obtain the coordinate information of the target based on its own coordinate system, and predict the trajectory of the target based on the coordinate information of the target based on its own coordinate system.

[0016] The present invention has the following beneficial effects: (1) The present invention shares the coordinate information of the target in real time through the Star Flash protocol. The slave device camera only needs to perform a simple matching calculation between the target coordinates monitored locally and the target coordinates shared by the master device camera to determine whether the target enters the overlapping monitoring range. This mechanism avoids the complex feature comparison and re-identification process in traditional cross-camera multi-target tracking. The present invention significantly reduces the system's computing resource requirements and operation and maintenance costs by simplifying coordinate calculations and reducing dependence on complex algorithms. At the same time, it makes the method more suitable for deployment in end-side cameras. Compared with traditional methods based on central servers, it reduces the data dependence on central server processing and return time, and can track targets more timely.

[0017] (2) When the camera performs coordinate matching, real-time performance is crucial, and the delay problem of traditional WI-FI communication will seriously affect the accuracy of coordinate matching, thereby leading to failure or error in target tracking. The present invention uses the Star Flash protocol for camera networking to achieve rapid sharing of coordinate information and ensure the real-time and accuracy of coordinate matching. Compared with traditional WI-FI communication, the low latency and high stability of the Star Flash protocol are one of the key factors for the present invention to solve the problem, which enables lightweight computational matching of cameras on the end side. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 The present invention is a flow chart of a monitoring method based on linkage of multiple cameras in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The detailed description of the accompanying drawings is intended as an illustration of the current embodiments of the present invention and is not intended to represent the only form in which the present invention can be implemented. It should be understood that the same or equivalent functions can be accomplished by different embodiments that are intended to be included in the spirit and scope of the present invention.

[0021] An embodiment of the present invention provides a monitoring method based on the linkage of multiple cameras. The multiple cameras are networked based on the SparkLine protocol. Through the networking, the cameras can share the target's location information, status updates and other key data in real time. The cameras are equipped with a wireless chip that supports the SparkLine protocol. The wireless communication method adopts the SparkLine protocol. The SparkLine access layer adopts SLE (SparkLine Low Energy). This mode requires an SLE gateway, a lightweight device and strong generalization capability. The coordinate processing and analysis are completed on the local side, and the coordinate information is shared. The bottom layer provides linkage of multiple cameras, and developers can define upper-level functions according to specific needs.

[0022] Specifically, the method of this embodiment includes the following steps: Step S10, multiple cameras perform target monitoring; Specifically, after all cameras are networked based on the Star Flash protocol, they start to synchronously enter the target monitoring state and continuously scan the monitoring area.

[0023] Step S20, when any camera detects a target and does not receive the coordinate information shared by other cameras, it determines itself as the current master device to perform the target tracking task; when the current master device performs the target tracking task, it assigns a unique ID to the target, monitors the coordinate information of the target in real time, generates the running trajectory of the target according to the coordinate information of the target, and shares the ID and coordinate information of the target with other cameras through the Star Flash communication protocol; Specifically, when a camera detects a target for the first time and does not receive coordinate information shared by other cameras, the camera is automatically upgraded to the current master device.

[0024] The main tasks of the master device include the following aspects: Assign unique ID: Assign a globally unique identifier (ID) to the detected object for subsequent cross-camera tracking.

[0025] Coordinate monitoring and trajectory generation: collect the target’s coordinate information (such as 2D / 3D coordinates based on image recognition) in real time and generate the target’s running trajectory.

[0026] Information sharing: The target’s ID, real-time coordinates and trajectory data are broadcast to other cameras in the network through the Star Flash protocol.

[0027] During this stage, a "first come, first served" master device election mechanism is used to ensure that the same target is tracked by only one master device, avoiding information redundancy caused by repeated tracking by multiple cameras.

[0028] Step S30, when any camera receives the coordinate information of the target shared by the current master device, it determines itself as the current slave device to perform the preparatory task; when the current slave device performs the preparatory task, it performs target monitoring, and if a suspected target is detected, it matches the coordinate information of the suspected target with the coordinate information of the target shared by the current master device, and if they match, it determines that the target enters the overlapping monitoring range between the current master device and the current slave device, monitors the target in real time, and sends a target tracking handover request to the current master device; Specifically, after any camera receives the target coordinate information shared by the master device, it automatically becomes the current slave device.

[0029] The slave device tasks mainly include the following aspects: Continuous monitoring: Continuously scan within its own monitoring range to detect whether there are suspected targets that match the target characteristics of the main device.

[0030] Coordinate matching: When a suspected target is detected, the slave device compares the coordinate information of the target collected locally with the coordinate information of the target shared by the master device in real time. If the coordinate error is within the preset threshold, a match is determined. If the match is successful, it is determined that the target has entered the overlapping monitoring range of the master and slave devices.

[0031] Send handover request: Send a target tracking handover request to the master device to trigger the master-slave switching process.

[0032] It should be noted that the low latency of the Star Flash protocol can ensure that the coordinate information shared by the master device is highly synchronized with the coordinate timestamp detected locally by the slave device, avoiding matching failures due to communication delays. The slave device does not need to specifically detect and identify the features of the target, but only needs to determine whether there is a target at the coordinate position shared by the master device to determine whether the target is detected. Because two targets cannot appear in the same position space, only a simple coordinate comparison is required, without the need for complex deep learning models (such as target re-identification algorithms), reducing the computing pressure on the end side.

[0033] Step S40, when the current master device is executing the target tracking task, if it receives the target tracking handover request, it stops executing the target tracking task, changes to the current slave device executing the preparatory task, and sends tracking handover confirmation information to the current slave device that sent the target tracking handover request; Specifically, when the master device receives the handover request from the slave device, it immediately performs the following operations: Stop tracking task: terminate the coordinate monitoring and trajectory update of the current target. You can also delete the generated trajectory to free up memory space, because the new master device will regenerate the trajectory. Role switching: downgrade itself to a slave device and enter the pre-task state; Send confirmation information: Send tracking handover confirmation information to the slave device that initiated the request through the Star Flash protocol.

[0034] Step S50, when the current slave device is performing the preparation task, if the tracking handover confirmation information is received, the current master device is changed to perform the target tracking task; Specifically, after receiving the confirmation message, the slave device that initiates the handover request immediately performs the following operations: Role switching: upgrade to the new master device and take over the target tracking task; Inherited data: Use the target ID assigned by the original master device and continue to generate trajectories based on the latest coordinates; Information broadcast: Share the updated coordinates to other cameras (slave devices) in the network.

[0035] It should be noted that the above process is triggered cyclically during the movement of the target. When the target enters the monitoring range of the next camera, the new slave device will repeat steps S20-S40 to achieve relay tracking across cameras. The method of this embodiment uses the fast communication and state synchronization of the Star Flash protocol, and the master-slave switching process is completed within milliseconds, ensuring that the trajectory continuity of the target is not interrupted when the camera field of view switches, thereby achieving comprehensive monitoring and efficient tracking.

[0036] The method of this embodiment has the following advantages: (1) When the camera is linked, the method of this embodiment only needs to perform simple coordinate calculations, without complex deep learning algorithms. This lightweight calculation greatly reduces the system's demand for computing resources. The Star Flash protocol shares the target's coordinate information in real time. The slave device only needs to perform a simple matching calculation between the locally monitored target coordinates and the master device's shared coordinates to determine whether the target enters the overlapping monitoring range. This mechanism avoids the complex feature comparison and re-identification process in traditional cross-camera multi-target tracking.

[0037] (2) Existing cross-camera multi-target tracking technologies usually rely on deep learning algorithms (such as convolutional neural networks, target re-identification models, etc.). Although these algorithms can improve tracking accuracy, they have extremely high requirements for computing resources and usually need to be deployed on the cloud or high-performance servers, resulting in a significant increase in system complexity and operation and maintenance costs. The method of this embodiment achieves efficient cross-camera target tracking through simple coordinate calculation and dynamic switching mechanism between master and slave devices without relying on complex deep learning algorithms. This enables the system to be deployed in the end-side camera, fully utilizing the local computing power of the camera and reducing dependence on the cloud or high-performance servers.

[0038] (3) Since the method of this embodiment has low requirements for computing resources, it can be easily deployed in the end-side camera. The end-side camera has a certain computing power (such as an embedded processor) that can meet simple coordinate calculation and real-time communication requirements. The traditional method requires uploading a large amount of video data to the cloud for processing, resulting in bandwidth occupation and data transmission delay. The method of this embodiment completes coordinate calculation and target tracking on the end side, and only needs to share a small amount of coordinate information, which significantly reduces the amount of data transmission, bandwidth requirements and communication costs.

[0039] (4) Since the method of this embodiment has low demand for computing resources, there is no need to deploy high-performance servers or cloud computing platforms, which reduces hardware procurement and maintenance costs. Traditional cross-camera multi-target tracking systems usually require complex architecture design (such as cloud computing, distributed storage, etc.), while the method of this embodiment simplifies the system architecture and reduces the difficulty and cost of operation and maintenance through end-side deployment and simple coordinate calculation.

[0040] (5) The computing tasks of the end-side camera are relatively light and the energy consumption is relatively low, which can extend the service life of the equipment and further reduce the operation and maintenance costs. Since the method of this embodiment has a low demand for computing resources, it can be easily expanded to more camera nodes without worrying about the bottleneck of computing resources. This flexibility enables the system to adapt to monitoring scenarios of different scales. The deployment method of the end-side camera is simple and fast, without the need for complex configuration and debugging, and can be put into use quickly, which improves the scalability and deployment efficiency of the system.

[0041] Furthermore, the Star Flash protocol supports multi-channel communication, which means that different data streams (for example, video streams or coordinate information from different cameras) can be transmitted simultaneously on different communication channels without interfering with each other. Therefore, the method of this embodiment can track multiple targets at the same time, a master device can perform target tracking tasks for at least two targets at the same time, and at least two master devices can also perform target tracking tasks for different targets respectively, allowing a master device to track multiple targets at the same time, and multiple master devices can serve as each other's slave devices to achieve more efficient and reliable target tracking.

[0042] For example: The master device A1 is responsible for tracking target a, and at the same time, as the slave device B2, it is ready to track target b.

[0043] The master device B1 is responsible for tracking the target b, and at the same time, as the slave device A2, it is ready to track the target a.

[0044] Furthermore, the step S20, matching the coordinate information of the suspected target with the coordinate information of the target shared by the current master device, includes: The coordinate information of the suspected target and the coordinate information of the target shared by the current main device are converted into the same reference coordinate system. Based on the same reference coordinate system, it is determined whether the coordinate information of the suspected target is consistent with the coordinate information of the target shared by the current main device. If so, they match; otherwise, they do not match.

[0045] Specifically, the following is a specific example to illustrate the coordinate information matching process in step S20: Suppose we have a security monitoring system in a shopping mall, which consists of multiple cameras to monitor and track the behavior of customers. Now there are two cameras, camera A and camera B, and the goal is to track a specific customer, called target a.

[0046] Camera A (master device) is tracking target a and regularly updates and shares the coordinate information of target a. Currently, target a is in the field of view of camera A, and its coordinate information is (Xa, Ya, Za); target a moves and enters the field of view of camera B. Camera B (slave device) detects a suspected target b and obtains its coordinate information (Xb, Yb, Zb); in order to determine whether the suspected target b is actually target a, coordinate matching is required. The following is the matching process: a. Coordinate conversion: Camera B converts the coordinate information (Xb, Yb, Zb) of the suspected target b it detects into the world coordinate system to obtain (X'b, Y'b, Z'b), and converts the coordinates (Xa, Ya, Za) of the target a shared by camera A into the world coordinate system to obtain (X'a, Y'a, Z'a). This usually involves translation and rotation of the coordinates to compensate for the position and angle differences between the two cameras. For example, suppose the rotation matrix R and translation vector T of the coordinate system of camera A relative to the world coordinate system are as follows:

[0047]

[0048] Among them, each column of the rotation matrix represents the representation of a basis vector in the new coordinate system in the old coordinate system, {R11, R12, R13} is the component of the x-axis vector of the new coordinate system in the old coordinate system, {R21, R22, R23} is the component of the y-axis vector of the new coordinate system in the old coordinate system, and {R31, R32, R33} is the component of the z-axis vector of the new coordinate system in the old coordinate system; are the components of the translation vector T on the x, y, and z axes respectively.

[0049] Then we have:

[0050]

[0051]

[0052] After the conversion, camera B obtains the coordinates of target a in the world coordinate system (X'a, Y'a, Z'a); Similarly, we can get (X'b, Y'b, Z'b); b. Matching judgment: After conversion, compare (X'b, Y'b, Z'b) with (X'a, Y'a, Z'a); If (X'b, Y'b, Z'b) is very close to (X'a, Y'a, Z'a) or within a predetermined error range, it can be considered that the suspected target b is the same target as the target a, so they match; if (X'b, Y'b, Z'b) is significantly different from (X'a, Y'a, Z'a) and exceeds the error range, it is considered that the suspected target b is not the target a, so they do not match; Assuming that after conversion and comparison, (X'b, Y'b, Z'b) is found to be consistent with (X'a, Y'a, Z'a), then camera B confirms that the suspected target b is actually target a, and the match is successful. Next, camera B can take over the tracking task of target a, and camera A stops tracking target a. Through this process, the monitoring system can ensure that it can accurately track the same target even when the target moves into the field of view of different cameras.

[0053] Furthermore, the step S10 further includes: When the current master device performs a target tracking task, it adjusts its own camera angle in real time according to the coordinate information of the target; the multiple cameras are all pan-tilt cameras.

[0054] Specifically, suppose there is an intelligent security monitoring system for monitoring the security of a large exhibition hall. The system consists of multiple pan-tilt cameras that can rotate horizontally and vertically to cover a larger monitoring area.

[0055] Camera A (currently the master device) uses its built-in intelligent analysis algorithm to identify and start tracking a target, called target a. Target a may be a visitor to an exhibition. Camera A continuously receives target a's coordinate information, which is based on target a's position in the camera's field of view. The following are the steps for how camera A adjusts its camera angle based on target a's coordinate information: a. Target a moves to the right and is about to leave the field of view of camera A. The control system of camera A detects the coordinate change of target a.

[0056] b. PTZ control: The control system of camera A can command the PTZ to perform corresponding horizontal and / or vertical rotations to ensure that target a remains in the camera's field of view; c. Horizontal adjustment: The gimbal of camera A starts to rotate horizontally and adjusts the angle to the right to match the moving direction and speed of target a.

[0057] d. Vertical adjustment: If target a changes height (for example, going up or down stairs), the gimbal of camera A will also rotate vertically to keep target a in the field of view.

[0058] Through this real-time adjustment, camera A is able to ensure that target a is always within its monitoring range, even if target a moves within the exhibition hall.

[0059] Furthermore, the step S20 further includes: When the current slave device performs a preparatory task, it predicts the trajectory of the target based on the coordinate information of the target shared by the current master device, and adjusts its own camera angle based on the trajectory prediction result to achieve seamless tracking and handover between different cameras; the multiple cameras are all pan-tilt cameras.

[0060] Specifically, suppose there is a museum security monitoring system consisting of multiple pan-tilt cameras to ensure the safety of exhibits and visitors. Now, there are two cameras, camera A (master) and camera B (slave), which are working together to track a visitor, called target a.

[0061] Camera A is tracking target a and regularly updates and shares the coordinate information of target a. The current coordinate information of target a is (Xa, Ya, Za). Target a is moving towards the monitoring area of ​​camera B. In order to prepare for seamless handover, camera B starts to perform preparatory tasks as a slave device.

[0062] Camera B receives the coordinate information (Xa, Ya, Za) of target a from camera A, and the control system of camera B uses this coordinate information to predict the trajectory of target a. For example, if target a is moving at a certain speed and direction, camera B can predict the position of target a in the next few seconds. Based on the trajectory prediction results, camera B begins to adjust the angle of its pan / tilt so that it can immediately monitor target a when target a enters its field of view. When target a enters the field of view of camera B, camera B has already adjusted the camera angle and can immediately start tracking target a. At the same time, camera A receives the handover confirmation information, stops tracking target a, and instead performs the preparatory task as a slave device.

[0063] Through this method of predicting and adjusting the camera angle, the PTZ camera can achieve seamless tracking and handover between different monitoring areas, improving the overall performance and reliability of the monitoring system.

[0064] Furthermore, in step S20, the current slave device predicts the trajectory of the target according to the coordinate information of the target shared by the current master device, including: The current slave devices convert the coordinate information of the target shared by the current master device into their own coordinate systems to obtain the coordinate information of the target based on their own coordinate systems, and predict the trajectory of the target based on the coordinate information of the target based on their own coordinate systems.

[0065] Specifically, suppose there is a campus security monitoring system, which consists of multiple pan-tilt cameras for monitoring campus security. Now there are two cameras, camera A (master device) and camera B (slave device), which are working together to track a student, called target a.

[0066] Camera A is tracking target a and regularly updates and shares the coordinate information of target a. The coordinate information of target a in the field of view of camera A is (Xa, Ya, Za). Target a is moving towards the monitoring area of ​​camera B. In order to prepare for seamless handover, camera B starts to perform preparatory tasks as a slave device.

[0067] Camera B receives the coordinate information (Xa, Ya, Za) of target a from camera A. Since cameras A and B may be installed in different positions and directions, their coordinate systems are different. Therefore, camera B needs to convert the received coordinate information (Xa, Ya, Za) to the coordinate system of camera B. After the conversion, camera B obtains the coordinates of target a in the coordinate system of camera B (X'a, Y'a, Z'a).

[0068] Camera B now predicts the trajectory of target a based on the converted coordinates (X'a, Y'a, Z'a), including analyzing the speed and direction of target a. Assuming that target a moves in the direction of camera B at a constant speed, camera B can predict that target a will reach a certain position (X'f, Y'f, Z'f) in its field of view after t seconds.

[0069] Camera B adjusts the angle of its gimbal according to the predicted target position (X'f, Y'f, Z'f) to ensure that target a can be detected immediately when it enters its field of view.

[0070] Through this method of coordinate transformation and trajectory prediction, camera B can accurately predict the position of target a and adjust the camera angle in advance, thereby achieving seamless tracking and handover with camera A.

[0071] As the same inventive concept, an embodiment of the present invention further provides a monitoring system based on linkage of multiple cameras, comprising multiple cameras, wherein the multiple cameras are networked based on the Star Flash protocol; The camera is used to determine itself as the current master device to perform the target tracking task when it detects the target and does not receive the coordinate information shared by other cameras; The current master device is used to assign a unique ID to the target, monitor the coordinate information of the target in real time, generate the running track of the target according to the coordinate information of the target, and share the ID and coordinate information of the target with other multiple cameras through the Star Flash communication protocol when performing the target tracking task; The camera is used to determine itself as the current slave device to perform the preparatory task when receiving the coordinate information of the target shared by the current master device; The current slave device is used to monitor the target when executing the preparatory task. If a suspected target is detected, the coordinate information of the suspected target is matched with the coordinate information of the target shared by the current master device. If they match, it is determined that the target enters the overlapping monitoring range between the current master device and the current slave device, the target is monitored in real time, and a target tracking handover request is sent to the current master device. The current master device is used to stop executing the target tracking task if it receives the target tracking handover request, change to the current slave device to execute the preparatory task, and send tracking handover confirmation information to the current slave device that sent the target tracking handover request; When the current slave device is executing the preparation task, if it receives the tracking handover confirmation information, it will be changed to the current master device to execute the target tracking task.

[0072] Furthermore, the current slave device is used to convert the coordinate information of the suspected target and the coordinate information of the target shared by the current master device into the same reference coordinate system, and based on the same reference coordinate system, determine whether the coordinate information of the suspected target is consistent with the coordinate information of the target shared by the current master device. If so, they match, otherwise, they do not match.

[0073] Furthermore, the current main device is used to adjust its own camera angle in real time according to the coordinate information of the target; and the multiple cameras are all pan-tilt cameras.

[0074] Furthermore, the current slave device is used to predict the trajectory of the target based on the coordinate information of the target shared by the current master device, and adjust its own camera angle according to the trajectory prediction result to achieve seamless tracking and handover between different cameras; the multiple cameras are all pan-tilt cameras.

[0075] Furthermore, the current slave device is used to convert the coordinate information of the target shared by the current master device into its own coordinate system to obtain the coordinate information of the target based on its own coordinate system, and predict the trajectory of the target based on the coordinate information of the target based on its own coordinate system.

[0076] The system of this embodiment corresponds to the method of the above embodiment, so the contents not described in detail in the system of this embodiment can be obtained by referring to the contents of the method of the above embodiment, and are therefore not repeated in this embodiment.

[0077] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A monitoring method based on linkage of multiple cameras, characterized in that: The multiple cameras are networked based on the Star Flash protocol, and the method includes: Multiple cameras for target monitoring; When any camera detects a target and does not receive the coordinate information shared by other cameras, it determines itself as the current master device to perform the target tracking task; when the current master device performs the target tracking task, it assigns a unique ID to the target, monitors the coordinate information of the target in real time, generates the running trajectory of the target according to the coordinate information of the target, and shares the ID and coordinate information of the target with other cameras through the Star Flash communication protocol; When any camera receives the coordinate information of the target shared by the current master device, it determines itself as the current slave device to perform the preparatory task; when the current slave device performs the preparatory task, it monitors the target, and if a suspected target is detected, it matches the coordinate information of the suspected target with the coordinate information of the target shared by the current master device. If they match, it determines that the target enters the overlapping monitoring range between the current master device and the current slave device, monitors the target in real time, and sends a target tracking handover request to the current master device; When the current master device is executing the target tracking task, if it receives the target tracking handover request, it stops executing the target tracking task, changes to the current slave device executing the preparatory task, and sends tracking handover confirmation information to the current slave device that sent the target tracking handover request; When the current slave device is executing the preparation task, if the tracking handover confirmation information is received, the current master device is changed to execute the target tracking task.

2. The monitoring method according to claim 1, characterized in that: The matching the coordinate information of the suspected target with the coordinate information of the target shared by the current master device includes: The coordinate information of the suspected target and the coordinate information of the target shared by the current main device are converted into the same reference coordinate system. Based on the same reference coordinate system, it is determined whether the coordinate information of the suspected target is consistent with the coordinate information of the target shared by the current main device. If so, they match; otherwise, they do not match.

3. The monitoring method according to claim 1, characterized in that: The method further comprises: When the current master device performs a target tracking task, it adjusts its own camera angle in real time according to the coordinate information of the target; the multiple cameras are all pan-tilt cameras.

4. The monitoring method according to claim 1, characterized in that: The method further comprises: When the current slave device performs a preparatory task, it predicts the trajectory of the target based on the coordinate information of the target shared by the current master device, and adjusts its own camera angle based on the trajectory prediction result to achieve seamless tracking and handover between different cameras; the multiple cameras are all pan-tilt cameras.

5. The monitoring method according to claim 4, characterized in that: The current slave device predicts the trajectory of the target according to the coordinate information of the target shared by the current master device, including: The current slave devices convert the coordinate information of the target shared by the current master device into their own coordinate systems to obtain the coordinate information of the target based on their own coordinate systems, and predict the trajectory of the target based on the coordinate information of the target based on their own coordinate systems.

6. A monitoring system based on multiple camera linkage, characterized in that: Comprising a plurality of cameras, wherein the plurality of cameras are networked based on the Star Flash protocol; The camera is used to determine itself as the current master device to perform the target tracking task when it detects the target and does not receive the coordinate information shared by other cameras; The current master device is used to assign a unique ID to the target, monitor the coordinate information of the target in real time, generate the running track of the target according to the coordinate information of the target, and share the ID and coordinate information of the target with other multiple cameras through the Star Flash communication protocol when performing the target tracking task; The camera is used to determine itself as the current slave device to perform the preparatory task when receiving the coordinate information of the target shared by the current master device; The current slave device is used to monitor the target when executing the preparatory task. If a suspected target is detected, the coordinate information of the suspected target is matched with the coordinate information of the target shared by the current master device. If they match, it is determined that the target enters the overlapping monitoring range between the current master device and the current slave device, the target is monitored in real time, and a target tracking handover request is sent to the current master device. The current master device is used to stop executing the target tracking task if it receives the target tracking handover request, change to the current slave device to execute the preparatory task, and send tracking handover confirmation information to the current slave device that sent the target tracking handover request; When the current slave device is executing the preparation task, if it receives the tracking handover confirmation information, it will be changed to the current master device to execute the target tracking task.

7. The monitoring system according to claim 6, characterized in that: The current slave device is used to convert the coordinate information of the suspected target and the coordinate information of the target shared by the current master device into the same reference coordinate system, and based on the same reference coordinate system, determine whether the coordinate information of the suspected target is consistent with the coordinate information of the target shared by the current master device. If so, they match, otherwise, they do not match.

8. The monitoring system according to claim 6, characterized in that: The current master device is used to adjust its own camera angle in real time according to the coordinate information of the target; and the multiple cameras are all pan-tilt cameras.

9. The monitoring system according to claim 6, characterized in that: The current slave device is used to predict the trajectory of the target according to the coordinate information of the target shared by the current master device, and adjust its own camera angle according to the trajectory prediction result to achieve seamless tracking and handover between different cameras; the multiple cameras are all pan-tilt cameras.

10. The monitoring system according to claim 9, characterized in that: The current slave device is used to convert the coordinate information of the target shared by the current master device into its own coordinate system to obtain the coordinate information of the target based on its own coordinate system, and predict the trajectory of the target based on the coordinate information of the target based on its own coordinate system.

Images