A target-guided cross-camera continuous video tracking and monitoring method

Abstract

The present invention discloses a target-guided cross-camera continuous video tracking and monitoring method, comprising the following steps: obtaining target information, obtaining camera resource capability information of the target area and sensitivity information of the monitoring area; obtaining user priority information and monitoring task importance information; calculating, based on the target information and the camera resource capability information, alternative cameras capable of monitoring the target; calculating the monitoring effect of each alternative camera and the priority number of each alternative camera; and selecting and switching cameras for target tracking based on the monitoring effect and priority number of the alternative camera. The present invention adopts dynamic resource scheduling, incorporates user priority, the importance of the task performed by the camera, the sensitivity of the monitoring area, and the capability of the camera resource into resource scheduling, performs resource scheduling based on the monitoring effect of the camera, dynamically switches cameras, and economically realizes multi-target cross-camera continuous video tracking and monitoring.

Description

Technical Field

[0001] The present invention relates to a video tracking method, and in particular to a target-guided cross-camera continuous video tracking and monitoring method. Background Art

[0002] In various security surveillance applications, ship traffic management in ports and waterways, and air traffic management at airports, continuous tracking of targets is required to maintain constant focus. Because the managed area is larger than the camera's effective monitoring range, multiple cameras are typically used to cover the entire managed area.

[0003] Existing technologies typically implement continuous video tracking and monitoring across multiple cameras manually, or in "exclusive" mode. Exclusive mode eliminates resource competition during monitoring, making resource scheduling simple and reliable. However, this approach has the following drawbacks: 1) Simultaneous tracking of multiple targets requires the deployment of more redundant cameras, which is less cost-effective; 2) Exclusive mode resource scheduling cannot dynamically select the camera that best monitors the target. Therefore, to ensure that the monitoring effect meets the established requirements, higher-performance cameras must be selected, requiring greater optical zoom capabilities and higher image clarity, which inevitably increases system construction costs.

[0004] Therefore, it is necessary to provide a target-guided method that can automatically switch cameras for continuous video tracking monitoring, ensuring monitoring effects while reducing costs. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a target-guided cross-camera continuous video tracking and monitoring method, which dynamically switches cameras for tracking and monitoring according to the monitoring effect of the cameras.

[0006] The technical solution adopted by the present invention to solve the above-mentioned technical problems is to provide a cross-camera continuous video tracking and monitoring method based on target guidance, comprising the following steps: S1: obtaining target information, obtaining camera resource capability information of the target area and sensitivity information of the monitoring area; obtaining user priority information and monitoring task importance information; S2: according to the target information, through the resource capability information of the camera, calculating the alternative cameras with the ability to monitor the target; S3: calculating the monitoring effect of each alternative camera, and calculating the priority number of each alternative camera; S4: if the camera currently used is the camera with the best monitoring effect, the target is tracked through the current camera, otherwise the camera is switched according to the monitoring effect and priority number of the alternative camera to track the target; S5: repeating steps S1-S4 to continuously track and monitor the target through target guidance.

[0007] Furthermore, the camera resource capability information includes: the name of the camera; the functional configuration of the camera, including: video monitoring only in human-computer interaction mode; video monitoring only for automatic target linkage tracking; sharing mode; the video stream access parameters of the camera, including: resource address, user name and password; channel configuration of the camera with different clarity, including image resolution and frame rate of each channel; the installation location and installation parameters of the camera, including: longitude and latitude of the installation location, installation height, PTZ azimuth deviation, PTZ pitch deviation and PTZ roll deviation; camera PTZ working parameters, including: horizontal azimuth rotation limit range, pitch rotation limit range, azimuth rotation speed and pitch rotation speed; the optical zoom of the camera Focus capability, including: minimum focal length, maximum focal length and optical zoom ratio; camera optical sensor parameters, including: sensor size width, sensor size height, sensor pixel width and sensor pixel height; camera monitoring coverage, including: optimal monitoring range and maximum monitoring range; camera occlusion area list, indicating which monitoring blind spots exist within a given monitoring coverage range; the definition of monitoring blind spots is described using a fan-shaped area, defined using the following parameters: starting distance, ending distance, starting azimuth, ending azimuth, starting pitch angle and ending pitch angle; camera usage status, when controlled by the user, the usage status takes the value of 1, when not controlled, it is in idle state, and the usage status takes the value of 0.

[0008] Furthermore, the step S2 includes: S21: based on the size of the target and the current location information of the target, confirming a list of all cameras with video linkage tracking capabilities in the area where the target is located; wherein the current location information of the target includes the latitude, longitude and altitude of the target's location; S22: traversing all cameras with video linkage tracking capabilities, and calculating the distance and direction from the camera to the target's location; S23: checking the monitoring coverage of the camera in the resource capability information of the camera, and confirming a list of cameras whose monitoring coverage includes the target based on the calculation result of step S22; S24: traversing all cameras whose monitoring coverage includes the target, and according to the camera's occlusion area list, excluding cameras whose targets are located in the occlusion area of the camera, to obtain a list of alternative cameras with the ability to monitor the target.

[0009] Furthermore, the step S3 includes: S31: sorting the candidate cameras in order of distance from the target from near to far; S32: traversing all candidate cameras according to the sorting in step S31, and calculating the optical zoom multiple and the pixel size of the target in the field of view when the camera obtains the best monitoring image effect according to the distance between the target and the camera and the size of the target, according to the area ratio set by the target in the field of view and the optical sensor parameters of the camera; S33: excluding the cameras whose optical zoom multiple exceeds the optical zoom capability of the camera according to the optical zoom capability of the candidate cameras, and sorting the remaining candidate cameras according to the pixel size of the target in the field of view to obtain the candidate cameras. Switch the camera list. The more pixels, the clearer it is, the better the monitoring effect, and the higher the ranking, and the cameras can be switched and sorted according to the monitoring effect; S34: Calculate the priority number of the switchable cameras. The priority number calculation formula is as follows: Priority number = User priority × Task importance × Area sensitivity × Camera usage status, where user priority is an integer greater than or equal to 1. The larger the value, the higher the user priority and the higher the user authority; Task importance is an integer greater than or equal to 1. The larger the value, the higher the task importance; Area sensitivity is an integer greater than or equal to 1. The larger the value, the higher the area sensitivity; The camera usage status is 0 or 1, which is 0 when idle and 1 when controlled by the user.

[0010] Furthermore, switching cameras according to the monitoring effects and priority numbers of the alternative cameras in step S4 includes: S41: checking the currently used camera; confirming that it is not the camera with the best monitoring effect, indicating that the camera needs to be switched; S42: deleting cameras in the list of switchable cameras with priority numbers higher than the priority number of the current camera; S43: checking the priority numbers of cameras ranked before the current camera in the list of switchable cameras to determine whether there is a camera with a priority number lower than the priority number of the current camera; if not, there is no camera that can be preempted for switching; if so, attempting to switch to cameras with priority numbers lower than the current camera in order of ranking; S44: if the switched camera is in an idle state, directly switching; S45: if the switched camera is in a user-controlled state, calculating whether, after the camera is switched, the task originally using the camera can select a valid camera without causing chain switching to occur three or more times; if so, it indicates that the impact of this camera preemptive switching behavior is small and the switch can be performed; if not, marking the camera currently being attempted to be switched; S46: excluding the marked cameras, repeating steps S43-S45 and attempting to switch to the next ranked camera until the switch is completed or there are no more cameras to switch.

[0011] Furthermore, the target tracking by the camera in step S4 specifically includes: S61: according to the position of the target and the installation position and installation parameters of the camera, the target is transformed into a spherical coordinate system with the installation position of the camera as the center of the sphere, and the azimuth, pitch angle and distance of the target relative to the camera are calculated; S62: according to the calculation result of step S61, combined with the optical zoom capability of the camera and the size of the target, the expected focal length when the area ratio of the target in the field of view is the set value is calculated, and the focal length of the camera is adjusted to the expected focal length. If the expected focal length is less than the minimum focal length, the minimum focal length is used; if the expected focal length is greater than the maximum focal length, the maximum focal length is used; S63: reading the actual position of the target, and calculating the PTZ difference with the azimuth, pitch angle calculated in step S61 and the focal length calculated in step S62, and sending the PTZ difference to the PID filter; S64: the PTZ adjustment amount output by the PID filter is packaged in a command according to the specified PTZ control protocol, and then sent to the camera to execute PTZ control, so as to realize real-time tracking of the target by the camera.

[0012] Furthermore, the priority number of the camera can be dynamically adjusted by dynamically adjusting the priority of the user, dynamically adjusting the importance of the task and / or dynamically adjusting the sensitivity of the monitoring area.

[0013] Furthermore, the user priority information includes: user nature, indicating whether the user is a manual seat user or a system user; the user's initial priority, for manual seat users, an initial priority is pre-assigned; for system users, an initial priority is pre-assigned according to the category of the system user; when the initial priority is not set, the user's initial priority is defaulted to 1; the importance information of the task performed by the camera includes: task nature, indicating whether the task is a general monitoring task or a command and control task; the initial importance level of the task, for general monitoring tasks, the initial importance level of the task is pre-set; for command and control tasks, the highest level is automatically assigned to the initial importance level; when the initial importance level is not set, the default initial importance level of the task is 1.

[0014] Furthermore, the sensitivity information of the monitored area includes: the name or code of the area; the sensitivity level of the area, different sensitivity levels are set according to the division of the area; when the sensitivity level is not set, the default sensitivity level of the area is 1; the range definition of the area is defined in the form of a closed convex polygon, including a list of latitude, longitude and altitude of each vertex of the polygon, and the height of the airspace is composed of the higher triangular top surface of the adjacent pentahedron.

[0015] Furthermore, the target information in step S1 includes the current position information of the target and the size information of the target; the target is a radar target, an AIS target, an ADS-B target, a Beidou positioning RDSS target, an ASTERIX air traffic control radar target or a fusion target of the above targets, and the target information is obtained through a radar, an AIS system, an ADS-B system, a Beidou satellite radio positioning system or an air traffic control radar system.

[0016] Compared with the prior art, the present invention has the following beneficial effects: the target-guided cross-camera continuous video tracking and monitoring method provided by the present invention adopts a dynamic resource scheduling strategy, and incorporates factors such as user priority and dynamic adjustment of priority, the importance of tasks performed by the camera and dynamic adjustment of task importance, the sensitivity of the monitoring area and dynamic adjustment of sensitivity, the capacity of camera resources and dynamic adjustment of capacity into resource scheduling, performs resource scheduling according to the monitoring effect of the camera, dynamically switches cameras, and economically realizes multi-target cross-camera continuous video tracking and monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a target-guided cross-camera continuous video tracking and monitoring method in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below with reference to the accompanying drawings and examples.

[0019] Figure 1 This is a flow chart of a target-guided cross-camera continuous video tracking and monitoring method in an embodiment of the present invention.

[0020] See Figure 1 The target-guided cross-camera continuous video tracking and monitoring method according to an embodiment of the present invention includes the following steps:

[0021] Step S1: Obtain target information, obtain camera resource capability information of the target area and sensitivity information of the monitoring area; obtain user priority information and monitoring task importance information;

[0022] Camera resource capability information, which describes the camera's monitoring capabilities. Camera resource capability information includes:

[0023] The name of the camera;

[0024] The camera's functional configuration includes: video surveillance in human-computer interaction mode only; video surveillance for automatic target linkage tracking only; sharing mode;

[0025] The camera's video stream access parameters are used for connection access during switching, including: resource address, user name, and password. The resource address includes the IP address and port.

[0026] The camera's channel configurations for different resolutions, including the image resolution and frame rate of each channel; for example, the resolution and frame rate of the main stream channel, the resolution and frame rate of the sub-stream, and the resolution and frame rate of the tertiary stream;

[0027] The camera's installation location and parameters, including the latitude and longitude of the installation location, installation altitude, PTZ azimuth deviation, PTZ pitch deviation, and PTZ roll deviation. The PTZ azimuth deviation is referenced to true north, the PTZ pitch deviation is referenced to true north, and the PTZ roll deviation is referenced to true east.

[0028] Camera pan / tilt operating parameters, including: horizontal azimuth rotation limit range, pitch rotation limit range, azimuth rotation speed, and pitch rotation speed;

[0029] The optical zoom capability of the camera, including: minimum focal length, maximum focal length, and optical zoom ratio; where the optical zoom ratio is equal to the maximum focal length divided by the minimum focal length;

[0030] The optical sensor parameters of the camera, including: sensor size width (mm), sensor size height (mm), sensor pixel width (pixels), sensor pixel height (pixels);

[0031] The camera's monitoring coverage, including: optimal monitoring range and maximum monitoring range;

[0032] The camera's blocked area list indicates the camera's blind spots within a given surveillance coverage area. Blind spots are described using sector-shaped areas and are defined using the following parameters: start distance, end distance, start azimuth, end azimuth, start pitch angle, and end pitch angle.

[0033] The camera's usage status: when it is controlled by the user, the usage status is 1; when it is not controlled, it is in idle state and the usage status is 0.

[0034] User priority information indicates the level of user authority. The principle of giving priority to users with higher authority is adopted. User priority information includes:

[0035] User nature, indicating whether the user is a human agent user or a system user;

[0036] The user's initial priority is pre-assigned to agent users and system users based on their category. If no initial priority is set, the default priority is 1. For example, system users such as violation alerts and evidence collection typically have higher priorities, while hazard alerts, such as collision alerts, typically have the highest priority.

[0037] The importance information of the camera's task indicates the importance of the task, and adopts the principle of priority of key tasks. The importance information of the camera's task includes:

[0038] The nature of the mission, indicating whether it is a general monitoring mission or a command and control mission;

[0039] The initial importance level of the task. For general monitoring tasks, the initial importance level of the task is pre-set; for command and control tasks, the highest level is automatically assigned to the initial importance level; when the initial importance level is not set, the default initial importance level of the task is 1.

[0040] The sensitivity information of the monitored area indicates the sensitivity of the monitored area. The principle of priority monitoring of key areas is adopted. The sensitivity information of the area includes:

[0041] the name or code of the area;

[0042] The sensitivity level of a region can be set to different levels based on its division. If no sensitivity level is set, the default sensitivity level for the region is 1. In airport air traffic control, a typical application scenario is to set a higher sensitivity level for the airspace where the runway is located to prevent dangerous situations such as airspace conflicts or runway incursions. In port and waterway vessel traffic management, a typical application scenario is to set a higher sensitivity level for the runway area of seaplanes to prevent other ships from invading and causing dangerous situations.

[0043] The scope of the area is defined in the form of a closed convex polygon, including a list of the latitude, longitude and altitude of each vertex of the polygon. The altitude of the airspace is composed of the higher triangular top surface of the adjacent pentahedron.

[0044] Target information includes the target's current location and size. The target can be a radar target, an AIS (Automatic Identification System) target, an ADS-B (Automatic Dependent Surveillance-Broadcast) target, a BeiDou RDSS (Radio Determination Satellite System) target, an ASTERIX air traffic control radar target, or a fusion of the above. Target information is acquired via radar, AIS, ADS-B, BeiDou RDSS, or air traffic control radar. If the target is capable of reporting its size (for example, AIS reports the ship's size using static messages), the reported size is used for calculation. If the target is not capable of reporting its size, or if no valid size information is received, the target size can be measured using an automatic measurement algorithm. If no subsequent automatic measurement algorithm exists, the target size in the target attribute information in the database can be used for calculation. If no corresponding database exists, a default size is used for calculation. During the linked tracking and monitoring process, the human agent user can also specify the target size for optimal tracking and monitoring results.

[0045] Step S2: Based on the target information and the camera's resource capability information, calculate the candidate cameras that have the capability to monitor the target; specifically, the process includes:

[0046] S21: Based on the size of the target and the current location information of the target, a list of all cameras with video linkage tracking capabilities in the area where the target is located is determined; wherein the current location information of the target includes the latitude, longitude and altitude of the target location;

[0047] S22: Traverse all cameras with video linkage tracking capabilities and calculate the distance and direction from the camera to the target location;

[0048] S23: Checking the camera's monitoring coverage in the camera's resource capability information, and confirming, based on the calculation result of step S22, a list of cameras whose monitoring coverage includes the target;

[0049] S24: Traverse all cameras whose monitoring coverage includes the target, and exclude cameras whose targets are located in the camera's blocking area list according to the camera's blocking area list, to obtain a list of candidate cameras capable of monitoring the target.

[0050] Step S3: Calculate the monitoring effect of each candidate camera and the priority number of each candidate camera; specifically, the steps include:

[0051] S31: sorting the candidate cameras in descending order of distance from the target;

[0052] S32: Traverse all candidate cameras according to the order in step S31, and calculate the optical zoom factor and pixel size of the target in the field of view when the camera obtains the best monitoring image effect according to the distance between the target and the camera, the size of the target, the set area ratio of the target in the field of view, and the optical sensor parameters of the camera;

[0053] S33: Based on the optical zoom capabilities of the candidate cameras, cameras with an optical zoom ratio exceeding the optical zoom capability of the camera are excluded, and the remaining candidate cameras are sorted according to the pixel size of the target in the field of view to obtain a list of switchable cameras. The cameras with more pixels and clearer images, and better monitoring effects, are ranked higher. The cameras can then be switched and sorted according to the monitoring effects.

[0054] The evaluation of surveillance effectiveness is based on image quality, the target's area percentage in the field of view, and the camera's ability to maintain or even increase the target's area percentage in the video image's field of view based on the target's motion. The camera's surveillance effectiveness is evaluated based on video image resolution, the target's area percentage in the field of view, and the camera's optical zoom capability. When the target's area percentage in the video image's field of view is constant, cameras with higher image clarity are more effective. Therefore, under the same conditions, cameras with high image quality (i.e., high video resolution) or higher-definition bitstreams are preferred. When the target's area percentage in the field of view is constant, cameras with the ability to further zoom in on the target are more effective. Therefore, under the same conditions, cameras with a higher zoom factor are preferred; under unequal conditions, cameras with a greater residual optical zoom capability are preferred.

[0055] S34: Calculate the priority number of the switchable camera. The priority number calculation formula is as follows:

[0056] Priority number = user priority × task importance × area sensitivity × camera usage status, where:

[0057] User priority is an integer greater than or equal to 1. The larger the value, the higher the user priority and the higher the user authority.

[0058] Task importance is an integer greater than or equal to 1. The larger the value, the higher the task importance.

[0059] Regional sensitivity is an integer greater than or equal to 1, and the larger the value, the higher the regional sensitivity;

[0060] The camera usage status is 0 or 1. It is 0 when idle and 1 when controlled by the user.

[0061] To ensure the smooth execution of various video-linked tracking and surveillance tasks and achieve optimal monitoring results, camera priorities are dynamically adjusted rather than statically assigned. In a video-linked tracking task, if a high-priority camera preempts a low-priority camera, the original low-priority camera immediately becomes high-priority, while the priority of the camera that was switched out becomes 0 (becoming an idle resource). To prevent camera switching from causing a decrease in overall monitoring effectiveness, the scheduler needs to dynamically adjust camera priorities to balance the use of system camera resources and prevent "priority reversal" from impacting monitoring performance. For example, if a high-priority camera preempts a low-priority camera, the task originally using that low-priority camera will be unable to select a suitable camera to continue video tracking and surveillance. Therefore, camera priorities can be dynamically adjusted by dynamically adjusting user priorities, task importance, and / or the sensitivity of the surveillance area.

[0062] To prevent priority inversion, dynamically adjust the user's priority and the camera's priority accordingly. By temporarily raising the user's priority, the camera assigned to that task is temporarily protected from being preempted by other tasks. Dynamically adjusting the importance of tasks and adjusting the camera's priority accordingly balances the conflict between "important" and "urgent" tasks. This allows cameras assigned to "urgent" but non-important tasks to preempt cameras assigned to "important" but non-urgent tasks. During the execution of a video-linked tracking task, as the monitored target gradually moves away from the monitored area, the task's importance decreases. Dynamically adjusting the sensitivity of the monitored area and adjusting the camera's priority can be used to respond to urgent or dangerous situations. For example, when multiple tasks are tracking multiple targets in a small area, potentially indicating a critical traffic situation, dynamically adjusting the sensitivity of that area can ensure effective monitoring and prevent the cameras assigned to these tasks from being preempted by other "non-urgent" tasks.

[0063] Step S4: If the currently used camera is the best camera for monitoring, the target is tracked through the current camera; otherwise, the camera is switched to track the target based on the monitoring effect and priority number of the candidate camera;

[0064] Among them, switching cameras according to the monitoring effect and priority of the alternative cameras includes:

[0065] S41: Check the currently used camera; if it is not the camera with the best monitoring effect, it means that the camera needs to be switched;

[0066] S42: Delete the camera whose priority number is higher than the current camera's priority number in the switchable camera list;

[0067] S43: Check the priority numbers of the cameras ranked before the current camera in the list of switchable cameras to determine whether there is a camera with a priority number lower than the priority number of the current camera. If not, there is no camera that can be preempted for switching. If so, try to switch to the cameras with a priority number lower than the current camera in order of ranking.

[0068] S44: If the camera to be switched is in idle state, switch directly;

[0069] S45: If the camera being switched is in a user-controlled state, calculate whether the task originally using the camera can select a valid camera after the camera is switched without causing chain switching to occur three or more times; if so, it means that the impact of this camera preemptive switching behavior is small and the switch can be performed; if not, mark the camera currently being attempted to be switched;

[0070] S46: Excluding the marked cameras, repeat steps S43-S45 and try to switch to the next sequenced camera until the switching is completed or there is no camera left to switch.

[0071] The camera performs target tracking specifically including:

[0072] S61: transforming the target into a spherical coordinate system with the camera installation position as the center according to the target position and the camera installation position and installation parameters, and calculating the azimuth, pitch angle, and distance of the target relative to the camera;

[0073] S62: Calculate the desired focal length when the area ratio of the target in the field of view is a set value based on the calculation result of step S61, combined with the optical zoom capability of the camera and the size of the target, and adjust the focal length of the camera to the desired focal length. If the desired focal length is less than the minimum focal length, the minimum focal length is used; if the desired focal length is greater than the maximum focal length, the maximum focal length is used.

[0074] S63: Read the actual position of the target, calculate the PTZ difference between the azimuth and pitch angles calculated in step S61 and the focal length calculated in step S62, and send the PTZ difference to the PID filter. PTZ is an abbreviation for Pan / Tilt / Zoom in security monitoring applications, representing full-scale (left / right / up / down) movement of the pan / tilt platform and lens zoom and focus control.

[0075] S64: The PTZ adjustment amount output by the PID filter is packaged into a command according to a specified PTZ control protocol, and then sent to the camera to perform PTZ control, thereby achieving real-time tracking of the target by the camera.

[0076] Step S5: Repeat steps S1-S4 to continuously track and monitor the target through target guidance.

[0077] The target-guided cross-camera continuous video tracking and monitoring method of the present invention is applied to the video linkage tracker of the AWxIFU intelligent fusion unit, and supports various types of target-guided PTZ cameras such as fusion targets, radar targets, AIS targets, ADS-B targets, Beidou RDSS targets, and ASTERIX air traffic control radar targets, to achieve continuous tracking and monitoring of multiple targets across cameras.

[0078] To sum up, the target-guided cross-camera continuous video tracking and monitoring method of the embodiment of the present invention adopts a dynamic resource scheduling strategy, and incorporates factors such as user priority and dynamic adjustment of priority, the importance of tasks performed by the camera and dynamic adjustment of task importance, the sensitivity of the monitoring area and dynamic adjustment of sensitivity, the capacity of camera resources and dynamic adjustment of capacity into resource scheduling. Resource scheduling is performed according to the monitoring effect of the camera, and the camera is dynamically switched to economically realize multi-target cross-camera continuous video tracking and monitoring.

[0079] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be based on the definition of the claims.

Claims

1. A target-guided cross-camera continuous video tracking and monitoring method, characterized in that: The steps include: S1: Obtain target information, including camera resource capability information in the target area and sensitivity information of the surveillance area; obtain user priority information and monitoring task importance information; S2: Based on the target information and the camera's resource capability information, calculate the candidate cameras that have the capability to monitor the target; S3: Calculate the monitoring effect of each candidate camera and calculate the priority number of each candidate camera; S4: If the currently used camera is the best camera for monitoring, the target is tracked through the current camera; otherwise, the camera is switched to track the target based on the monitoring effect and priority of the alternative camera; S5: Repeat steps S1-S4 to continuously track and monitor the target through target guidance; The step S3 comprises: S31: Sort the candidate cameras by their distance from the target, from near to far; S32: Traverse all candidate cameras according to the order in step S31, and calculate the optical zoom factor and pixel size of the target in the field of view when the camera obtains the best monitoring image effect based on the distance between the target and the camera, the size of the target, the area ratio of the target in the field of view, and the optical sensor parameters of the camera; S33: Based on the optical zoom capabilities of the candidate cameras, cameras with optical zoom ratios exceeding the optical zoom capability of the camera are excluded, and the remaining candidate cameras are sorted according to the pixel size of the target in the field of view to obtain a list of switchable cameras; S34: Calculate the priority number of switchable cameras: Priority number = user priority × task importance × area sensitivity × camera usage status, Among them, user priority is an integer greater than or equal to 1. The larger the value, the higher the user priority, the higher the user authority; task importance is an integer greater than or equal to 1. The larger the value, the higher the task importance; area sensitivity is an integer greater than or equal to 1. The larger the value, the higher the area sensitivity; the camera usage status is 0 or 1. It is 0 when it is idle and 1 when it is controlled by the user. The step S4 comprises: S41: Check the currently used camera; if it is not the camera with the best monitoring effect, it means that the camera needs to be switched; S42: Delete the camera whose priority number is higher than the current camera's priority number in the switchable camera list; S43: Check the priority numbers of the cameras that are ranked before the current camera in the list of switchable cameras to determine whether there is a camera with a priority number lower than the priority number of the current camera. If not, there is no camera that can be preempted for switching. If so, try to switch to the cameras with a priority number lower than the current camera in order of ranking. S44: If the camera to be switched is in idle state, switch directly; S45: If the camera being switched is in a user-controlled state, calculate whether the task originally using the camera can select a valid camera after the camera is switched without causing chain switching to occur three or more times; if so, switch is performed; if not, mark the camera currently being attempted to be switched; S46: Excluding the marked cameras, repeat steps S43-S45 and try to switch to the next sequenced camera until the switching is completed or there is no camera left to switch.

2. The target-guided cross-camera continuous video tracking and monitoring method according to claim 1, wherein: The camera resource capability information includes: The name of the camera; The camera's functional configuration includes: video surveillance in human-computer interaction mode only; video surveillance for automatic target linkage tracking only; sharing mode; The camera's video stream access parameters, including: resource address, user name, and password; Channel configurations of different camera resolutions, including image resolution and frame rate for each channel; The camera's installation location and installation parameters, including: the longitude and latitude of the installation location, installation altitude, PTZ azimuth deviation, PTZ pitch deviation, and PTZ roll deviation; Camera pan / tilt operating parameters, including: horizontal azimuth rotation limit range, pitch rotation limit range, azimuth rotation speed, and pitch rotation speed; The optical zoom capability of the camera, including: minimum focal length, maximum focal length and optical zoom ratio; The camera's optical sensor parameters, including sensor size width, sensor size height, sensor pixel width, and sensor pixel height; The camera's monitoring coverage, including: optimal monitoring range and maximum monitoring range; The camera's blocked area list indicates the camera's blind spots within a given surveillance coverage area. Blind spots are described using sector-shaped areas and are defined using the following parameters: start distance, end distance, start azimuth, end azimuth, start pitch angle, and end pitch angle. The camera's usage status: when it is controlled by the user, the usage status is 1; when it is not controlled, it is in idle state and the usage status is 0.

3. The target-guided cross-camera continuous video tracking and monitoring method according to claim 2, wherein: The step S2 comprises: S21: Based on the size of the target and the current location information of the target, a list of all cameras with video linkage tracking capabilities in the area where the target is located is determined; wherein the current location information of the target includes the latitude, longitude and altitude of the target location; S22: Traverse all cameras with video linkage tracking capabilities and calculate the distance and direction from the camera to the target location; S23: Checking the camera's monitoring coverage in the camera's resource capability information, and confirming, based on the calculation result of step S22, a list of cameras whose monitoring coverage includes the target; S24: Traverse all cameras whose monitoring coverage includes the target, and exclude cameras whose targets are located in the camera's blocking area list according to the camera's blocking area list, to obtain a list of candidate cameras capable of monitoring the target.

4. The target-guided cross-camera continuous video tracking and monitoring method according to claim 1, wherein: The target tracking by the camera in step S4 specifically includes: S61: transforming the target into a spherical coordinate system with the camera installation position as the center according to the target position and the camera installation position and installation parameters, and calculating the azimuth, pitch angle, and distance of the target relative to the camera; S62: Calculate the desired focal length when the area ratio of the target in the field of view is a set value based on the calculation result of step S61, combined with the optical zoom capability of the camera and the size of the target, and adjust the focal length of the camera to the desired focal length. If the desired focal length is less than the minimum focal length, the minimum focal length is used; if the desired focal length is greater than the maximum focal length, the maximum focal length is used. S63: Read the actual position of the target, calculate the PTZ difference between the azimuth and pitch angles calculated in step S61 and the focal length calculated in step S62, and send the PTZ difference to the PID filter; S64: The PTZ adjustment amount output by the PID filter is packaged into a command according to a specified PTZ control protocol, and then sent to the camera to perform PTZ control, thereby achieving real-time tracking of the target by the camera.

5. The target-guided cross-camera continuous video tracking and monitoring method according to claim 1, wherein: The priority number of the camera can be dynamically adjusted by dynamically adjusting the priority of the user, dynamically adjusting the importance of the task and / or dynamically adjusting the sensitivity of the monitoring area.

6. The target-guided cross-camera continuous video tracking and monitoring method according to claim 1, wherein: The user priority information includes: user nature, indicating whether the user is a manual seat user or a system user; the user's initial priority, for manual seat users, an initial priority is pre-assigned; for system users, an initial priority is pre-assigned according to the category of the system user; when the initial priority is not set, the user's initial priority is defaulted to 1; the importance information of the camera's task execution includes: task nature, indicating whether the task is a general monitoring task or a command and control task; the initial importance level of the task, for general monitoring tasks, the initial importance level of the task is pre-set; for command and control tasks, the highest level is automatically assigned to the initial importance level; when the initial importance level is not set, the default initial importance level of the task is 1.

7. The target-guided cross-camera continuous video tracking and monitoring method according to claim 1, wherein: The sensitivity information of the monitored area includes: the name or code of the area; the sensitivity level of the area, different sensitivity levels are set according to the division of the area; when the sensitivity level is not set, the default sensitivity level of the area is 1; the range definition of the area, which is defined in the form of a closed convex polygon, including a list of the latitude, longitude and altitude of each vertex of the polygon.

8. The target-guided cross-camera continuous video tracking and monitoring method according to claim 1, wherein: The target information in step S1 includes the current position information of the target and the size information of the target; the target is a radar target, an AIS target, an ADS-B target, a Beidou positioning RDSS target, an ASTERIX air traffic control radar target, or a fusion target of the above targets, and the target information is obtained through a radar, an AIS system, an ADS-B system, a Beidou satellite radio positioning system, or an air traffic control radar system.

Images