Video target tracking method and system of intelligent holder

Abstract

The invention relates to the technical field of image recognition, and provides a video target tracking method and system of an intelligent holder. The method comprises the following steps: acquiring video stream data acquired by a first holder camera at a first current position, and generating a target monitoring area of a tracking target in a preset time period based on the video stream data; judging whether the target monitoring area is included in a first global monitoring area of the first holder camera or not; and if the target monitoring area is not included in the first global monitoring area, determining a target second pan-tilt camera based on the first global monitoring area and the target monitoring area, and starting the target second pan-tilt camera in the preset time period. According to the method, the problem of monitoring blind areas can be effectively avoided.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a video target tracking method and system for an intelligent pan-tilt platform. Background Art

[0002] As one of the core devices of modern surveillance systems, PTZ cameras have powerful remote control functions and can flexibly adjust camera positions and track targets within a wide monitoring range.

[0003] However, existing intelligent PTZ video surveillance systems still face some technical challenges in the target tracking process. On the one hand, the monitoring range of PTZ cameras is usually limited. Traditional security monitoring technology mostly relies on the operation of a single PTZ camera, which easily leads to the creation of monitoring blind spots and cannot achieve comprehensive monitoring of the target. On the other hand, when multiple PTZ cameras work together, they are usually turned on at the same time. This monitoring method is prone to energy waste. Summary of the Invention

[0004] This application provides a video target tracking method and system for an intelligent pan-tilt camera to solve the problems raised by the above background technology.

[0005] In a first aspect, the present application provides a video target tracking method for an intelligent pan-tilt head, comprising: Obtaining video stream data collected by a first pan-tilt camera at a first current position, and generating a target monitoring area for a tracking target within a preset time period based on the video stream data; Determining whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera; If it is included in the first global monitoring area, determining whether the target monitoring area is included in the first partial monitoring area corresponding to the first PTZ camera at the first current position; If it is included in the first partial monitoring area, controlling the first pan-tilt camera to maintain the first current position within the preset time period; If it is not included in the first partial monitoring area, adjusting the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area; If it is not included in the first global monitoring area, a target second pan-tilt camera is determined based on the first global monitoring area and the target monitoring area, and the target second pan-tilt camera is started within the preset time period.

[0006] In a possible implementation, generating a target monitoring area of a tracking target within a preset time period based on the video stream data includes: Respectively obtaining a position of the tracking target corresponding to each video frame in the video stream data; For two adjacent image frames in the video stream data, calculating a distance between positions corresponding to the two adjacent image frames and a first moving direction of a tracking target from a position corresponding to a previous image frame to a position corresponding to a next image frame in the two adjacent image frames; determining the maximum distance among the distances as the target distance; Sequentially obtaining time differences between capture moments corresponding to adjacent video frames in the video stream data, and determining the minimum time difference among the time differences as a target time difference; Determining a ratio of the preset time period to the target time difference as a target value; For each of the first moving directions, determining the product of the target distance and the target value as the moving distance corresponding to the tracked target in the first moving direction, and determining the target node corresponding to the moving distance in the first moving direction with the first current position of the tracked target as the starting node; The starting node is sequentially connected to each target node in a clockwise direction to obtain a target area, and a circumscribed circle of the target area is determined as the target monitoring area.

[0007] In a possible implementation, respectively obtaining the position of the tracking target corresponding to each video frame in the video stream data includes: For any video frame in the video stream data, determining whether the video frame is the first video frame in the video stream data; If it is the first video frame, perform global detection on the video frame using a preset target detection algorithm to obtain a first confidence score for each object in the video frame, and determine the position corresponding to the object with the highest first confidence score as the position of the tracking target; If it is not the first video frame, respectively obtaining similarities between each object in the video frame and the tracked target detected in the first video frame, and determining whether a maximum similarity among the similarities is greater than a preset similarity; If the similarity is greater than a preset similarity, determining the position of the object corresponding to the maximum similarity as the position of the tracking target; If the similarity is not greater than a preset similarity, the video frame is cropped to obtain an initial cropped image; the center position of the initial cropped image is the position corresponding to the tracking target in the video frame before the video frame, and the size of the initial cropped image is twice the circumscribed rectangle of the image corresponding to the tracking target in the first video frame; The size of the initial cropped image is adjusted to be consistent with the size of the video frame to obtain a target cropped image, and a preset target detection algorithm is used to perform global detection on the target cropped image to obtain a second confidence level of each object in the target cropped image, and the position corresponding to the object with the highest second confidence level is determined as the position of the tracking target.

[0008] In a possible implementation, adjusting the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area includes: Obtaining first coordinates of a center position of the first partial monitoring area, and obtaining second coordinates of a center position of the monitoring area; determining a displacement direction between the first coordinate and the second coordinate as a second movement direction; The first pan-tilt camera is controlled to move along the second moving direction until the monitoring area of the first pan-tilt camera is the target monitoring area.

[0009] In a possible implementation, determining a target second PTZ camera based on the first global monitoring area and the target monitoring area, and starting the target second PTZ camera within the preset time period, includes: Dividing the target monitoring area into a first target monitoring sub-area and a second target monitoring sub-area based on the first global monitoring area; wherein the first target monitoring sub-area is included in the first global monitoring area, and the second target monitoring sub-area is not included in the first global monitoring area; Obtain the second global monitoring area corresponding to each second PTZ camera; For each of the second PTZ cameras, calculating a ratio between an area of a local area of the second target monitoring sub-region located in the second global monitoring area of the second PTZ camera and a total area of the second target monitoring sub-region; Determine the second pan-tilt camera corresponding to the maximum ratio among the ratios as the target second pan-tilt camera; Obtaining a third coordinate of the center position of the second partial monitoring area corresponding to the second current position of the target second pan-tilt camera, and obtaining a fourth coordinate of the center position of the local area; determining a displacement direction between the third coordinate and the fourth coordinate as a third movement direction; The target second pan-tilt camera is controlled to move along the third moving direction until the monitoring area of the target second pan-tilt camera completely includes the local area.

[0010] In a possible implementation, the method further includes: After the preset time period, the first pan-tilt camera is turned off.

[0011] In a second aspect, the present application provides a video target tracking system for an intelligent pan-tilt platform, comprising: a generating module, configured to obtain video stream data collected by the first pan-tilt camera at the first current position, and generate a target monitoring area for tracking the target within a preset time period based on the video stream data; A first judgment module is used to judge whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera; a second judgment module, configured to judge, when the target monitoring area is included in the first global monitoring area, whether the target monitoring area is included in a first partial monitoring area corresponding to the first PTZ camera at the first current position; a control module, configured to control the first pan-tilt camera to maintain the first current position within the preset time period when the target monitoring area is included in the first partial monitoring area; an adjustment module, configured to adjust the position of the first PTZ camera within the preset time period based on the first partial monitoring area and the target monitoring area when the target monitoring area is not included in the first partial monitoring area; A determination module is used to determine a target second pan-tilt camera based on the first global monitoring area and the target monitoring area when the target monitoring area is not included in the first global monitoring area, and to start the target second pan-tilt camera within the preset time period.

[0012] The present application provides a video target tracking method and system for an intelligent pan-tilt camera. The method includes: obtaining video stream data collected by a first pan-tilt camera at a first current position, and generating a target monitoring area of the tracking target within a preset time period based on the video stream data; judging whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera; if included in the first global monitoring area, judging whether the target monitoring area is included in the first partial monitoring area corresponding to the first pan-tilt camera at the first current position; if included in the first partial monitoring area, controlling the first pan-tilt camera to maintain the first current position within the preset time period; if not included in the first partial monitoring area, adjusting the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area; if not included in the first global monitoring area, determining a target second pan-tilt camera based on the first global monitoring area and the target monitoring area, and starting the target second pan-tilt camera within the preset time period. This method, on the one hand, can intelligently determine whether to adjust the pan-tilt camera or switch the pan-tilt camera by dynamically predicting the relationship between the target monitoring area and the monitoring range of the pan-tilt camera within a preset time period, thereby effectively avoiding the problem of blind spots in traditional single camera monitoring and improving the continuity and stability of target tracking. On the other hand, a multi-pan-tilt camera collaboration mechanism is adopted. When the target monitoring area is not included in the first global monitoring area, the target second pan-tilt camera is determined based on the first global monitoring area and the target monitoring area, and the target second pan-tilt camera is started within the preset time period, thereby avoiding the energy waste caused by simultaneously starting multiple pan-tilt cameras in the existing technology and significantly reducing system power consumption. On the other hand, through the target monitoring area prediction within the preset time period and the optimized control of the pan-tilt camera, unnecessary pan-tilt rotation is reduced, which helps to extend the service life of the equipment and realize the organic combination of global monitoring and local precise tracking, while ensuring the security monitoring effect and optimizing resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0014] Figure 1 A schematic diagram of a flow chart of a video target tracking method for an intelligent pan-tilt platform provided in an embodiment of the present application; Figure 2 A schematic block diagram of the structure of the video target tracking system of the intelligent pan-tilt platform provided in an embodiment of the present application; Figure 3A schematic block diagram of the structure of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0016] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0017] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0018] It should be further understood that the term "and / or" used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0019] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0020] See also Figure 1 , Figure 1 A flow chart of a video target tracking method of an intelligent PTZ provided in an embodiment of the present application is shown as follows: Figure 1 As shown, the video target tracking method of the intelligent pan-tilt platform provided in the embodiment of the present application includes steps S1 to S6.

[0021] Step S1: Obtain video stream data collected by a first pan-tilt camera at a first current position, and generate a target monitoring area for tracking a target within a preset time period based on the video stream data.

[0022] Step S2: Determine whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera.

[0023] Step S3: If the target monitoring area is included in the first global monitoring area, determine whether the target monitoring area is included in the first partial monitoring area corresponding to the first pan-tilt camera at the first current position.

[0024] Step S4: If the first portion of the monitoring area is included, controlling the first PTZ camera to maintain the first current position within the preset time period; Step S5: If the target area is not included in the first partial monitoring area, adjusting the position of the first PTZ camera within the preset time period based on the first partial monitoring area and the target monitoring area; Step S6: If it is not included in the first global monitoring area, determine the target second pan-tilt camera based on the first global monitoring area and the target monitoring area, and start the target second pan-tilt camera within the preset time period.

[0025] It should be noted that the solution provided in this embodiment is used for security monitoring, and the execution subject of this embodiment can be a server, or a video target tracking system or terminal device of an intelligent pan-tilt platform.

[0026] This embodiment specifically includes: As described in the above step S1, the video stream data collected by the first pan-tilt camera at the first current position is obtained, and a target monitoring area of the tracking target within a preset time period is generated based on the video stream data. Specifically, step S1 includes: when the first pan-tilt camera is located at the first current position, obtaining the video frames captured by the first pan-tilt camera in real time to obtain the video stream data; respectively obtaining the position corresponding to each video frame of the tracking target in the video stream data; for two adjacent image frames in the video stream data, calculating the distance between the positions corresponding to the two adjacent image frames and the first moving direction of the tracking target moving from the position corresponding to the previous image frame in the two adjacent image frames to the position corresponding to the next image frame; determining the maximum distance among the distances as the target distance; respectively obtaining the time difference between the capture moments corresponding to adjacent video frames in the video stream data in sequence, and determining the minimum time difference among the time differences as the target time difference; determining the ratio of the preset time period to the target time difference as the target value; for each first moving direction, determining the product of the target distance and the target value as the moving distance corresponding to the tracking target in the first moving direction, and determining the target node corresponding to the moving distance in the first moving direction with the first current position of the tracking target as the starting node; The starting node is sequentially connected to each target node in a clockwise direction to obtain a target area, and a circumscribed circle of the target area is determined as the target monitoring area.

[0027] As described in step S2 above, it is determined whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera.

[0028] As described in step S3 above, when the target monitoring area is included in the first global monitoring area, it is determined whether the target monitoring area is included in the first partial monitoring area corresponding to the first pan-tilt camera at the first current position.

[0029] As described in step S4 above, when the target monitoring area is included in the first partial monitoring area, the first pan-tilt camera is controlled to maintain the first current position within the preset time period.

[0030] As described in step S5 above, when the target monitoring area is not included in the first partial monitoring area, the position of the first pan-tilt camera within the preset time period is adjusted based on the first partial monitoring area and the target monitoring area. Specifically, step S4 includes: obtaining a first coordinate of the center position of the first partial monitoring area and obtaining a second coordinate of the center position of the monitoring area; determining the displacement direction between the first coordinate and the second coordinate as a second movement direction; and controlling the first pan-tilt camera to move along the second movement direction until the monitoring area of the first pan-tilt camera is the target monitoring area.

[0031] As described in step S6 above, when the target monitoring area is not included in the first global monitoring area, a target second PTZ camera is determined based on the first global monitoring area and the target monitoring area, and the target second PTZ camera is started within the preset time period. Specifically, step S6 includes: dividing the target monitoring area into a first target monitoring sub-area and a second target monitoring sub-area based on the first global monitoring area; wherein the first target monitoring sub-area is included in the first global monitoring area, and the second target monitoring sub-area is not included in the first global monitoring area; obtaining the second global monitoring area corresponding to each second pan-tilt camera; for each second pan-tilt camera, calculating the ratio between the area of the local area of the second target monitoring sub-area located in the second global monitoring area of the second pan-tilt camera and the total area of the second target monitoring sub-area; determining that the second pan-tilt camera corresponding to the maximum ratio among the ratios is the target second pan-tilt camera; obtaining the third coordinate of the center position of the second partial monitoring area corresponding to the second current position of the target second pan-tilt camera, and obtaining the fourth coordinate of the center position of the local area; determining the displacement direction between the third coordinate and the fourth coordinate as the third moving direction; controlling the target second pan-tilt camera to move along the third moving direction until the monitoring area of the target second pan-tilt camera completely includes the local area.

[0032] The method provided in this embodiment, on the one hand, can intelligently determine whether to adjust the pan-tilt camera or switch the pan-tilt camera by dynamically predicting the relationship between the target monitoring area and the monitoring range of the pan-tilt camera within a preset time period, thereby effectively avoiding the problem of blind spots in traditional single camera monitoring and improving the continuity and stability of target tracking. On the other hand, a multi-pan-tilt camera collaboration mechanism is adopted. When the target monitoring area is not included in the first global monitoring area, the target second pan-tilt camera is determined based on the first global monitoring area and the target monitoring area, and the target second pan-tilt camera is started within the preset time period, thereby avoiding the energy waste caused by simultaneously starting multiple pan-tilt cameras in the existing technology and significantly reducing system power consumption. On the other hand, through the prediction of the target monitoring area within the preset time period and the optimized control of the pan-tilt camera, unnecessary pan-tilt rotation is reduced, which helps to extend the service life of the equipment and realize the organic combination of global monitoring and local precise tracking, while ensuring the security monitoring effect and optimizing resource utilization.

[0033] In some embodiments, generating a target monitoring area for a tracking target within a preset time period based on the video stream data includes the following steps: Respectively obtaining a position of the tracking target corresponding to each video frame in the video stream data; For two adjacent image frames in the video stream data, calculating a distance between positions corresponding to the two adjacent image frames and a first moving direction of a tracking target from a position corresponding to a previous image frame to a position corresponding to a next image frame in the two adjacent image frames; determining the maximum distance among the distances as the target distance; Sequentially obtaining time differences between capture moments corresponding to adjacent video frames in the video stream data, and determining the minimum time difference among the time differences as a target time difference; Determining a ratio of the preset time period to the target time difference as a target value; For each of the first moving directions, determining the product of the target distance and the target value as the moving distance corresponding to the tracked target in the first moving direction, and determining the target node corresponding to the moving distance in the first moving direction with the first current position of the tracked target as the starting node; The starting node is sequentially connected to each target node in a clockwise direction to obtain a target area, and a circumscribed circle of the target area is determined as the target monitoring area.

[0034] The method provided in this embodiment determines, for each first moving direction, the product of the target distance and the target value as the moving distance corresponding to the tracked target in the first moving direction, determines the maximum possible moving distance of the tracked target in each first moving direction, connects the starting node with each target node in sequence in a clockwise direction to obtain a target area, and determines the circumscribed circle of the target area as the target monitoring area, which helps to improve the reliability of the target monitoring area and effectively avoid the problem of tracking loss.

[0035] In some embodiments, respectively obtaining the position of the tracking target corresponding to each video frame in the video stream data includes the following steps: For any video frame in the video stream data, determining whether the video frame is the first video frame in the video stream data; If it is the first video frame, perform global detection on the video frame using a preset target detection algorithm to obtain a first confidence score for each object in the video frame, and determine the position corresponding to the object with the highest first confidence score as the position of the tracking target; If it is not the first video frame, respectively obtain the similarity between each object in the video frame and the tracking target detected in the first video frame, and determine whether the maximum similarity among the similarities is greater than a preset similarity; specifically, for each object in the video frame, input the image area corresponding to the object in the video frame into a preset image feature extraction model to obtain a first feature vector, and input the image area of the tracking target detected in the first video frame into a preset image feature extraction model to obtain a second feature vector, and determine the cosine value between the first feature vector and the second feature vector as the similarity.

[0036] If the similarity is greater than a preset similarity, determining the position of the object corresponding to the maximum similarity as the position of the tracking target; If the similarity is not greater than a preset similarity, the video frame is cropped to obtain an initial cropped image; the center position of the initial cropped image is the position corresponding to the tracking target in the video frame before the video frame, and the size of the initial cropped image is twice the circumscribed rectangle of the image corresponding to the tracking target in the first video frame; The size of the initial cropped image is adjusted to be consistent with the size of the video frame to obtain a target cropped image, and a preset target detection algorithm is used to perform global detection on the target cropped image to obtain a second confidence level of each object in the target cropped image, and the position corresponding to the object with the highest second confidence level is determined as the position of the tracking target.

[0037] The method provided in this embodiment, on the one hand, performs global detection on the first video frame through a preset target detection algorithm and selects the object with the highest confidence as the tracking target, thereby ensuring the accuracy of the initial target positioning and laying a reliable foundation for subsequent tracking; on the other hand, for non-first video frames, a similarity matching algorithm is used to determine the position of the tracking target, which helps to reduce system computing power and improve the efficiency of position determination; on the other hand, when the maximum similarity is not greater than the preset similarity, the video frame is cropped to obtain an initial cropped image, and the size of the initial cropped image is adjusted to be consistent with the size of the video frame to obtain a target cropped image, and the target cropped image is globally detected using a preset target detection algorithm to obtain a second confidence of each object in the target cropped image, and the position corresponding to the object with the highest second confidence is determined as the position of the tracking target, thereby achieving a balance between the accuracy and efficiency of position determination.

[0038] In some embodiments, adjusting the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area includes the following steps: Obtaining first coordinates of a center position of the first partial monitoring area, and obtaining second coordinates of a center position of the monitoring area; Determine the displacement direction between the first coordinate and the second coordinate as a second movement direction; it can be understood that the second movement direction refers to the direction from the first coordinate to the second coordinate; The first pan-tilt camera is controlled to move along the second moving direction until the monitoring area of the first pan-tilt camera is the target monitoring area.

[0039] The method provided in this embodiment, on the one hand, adjusts the position of the first pan-tilt camera within the preset time period when the tracking target has not completely left the first part of the monitoring area, which helps to prevent tracking loss; on the other hand, by calculating the center coordinate displacement of the target monitoring area and the current monitoring area, the movement direction of the pan-tilt that needs to be adjusted is accurately determined, so that the pan-tilt steering operation has a clear target directionality, avoiding the time delay caused by traditional trial-and-error adjustment; on the other hand, by setting the end point of the pan-tilt movement to the coverage range of the target monitoring area, the adjusted monitoring angle of view can completely cover the predicted activity area of the target, effectively preventing the problem of tracking failure caused by insufficient or excessive adjustment.

[0040] In some embodiments, determining a target second PTZ camera based on the first global monitoring area and the target monitoring area, and starting the target second PTZ camera within the preset time period, includes the following steps: Dividing the target monitoring area into a first target monitoring sub-area and a second target monitoring sub-area based on the first global monitoring area; wherein the first target monitoring sub-area is included in the first global monitoring area, and the second target monitoring sub-area is not included in the first global monitoring area; Obtain the second global monitoring area corresponding to each second PTZ camera; For each of the second PTZ cameras, calculating a ratio between an area of a local area of the second target monitoring sub-region located in the second global monitoring area of the second PTZ camera and a total area of the second target monitoring sub-region; Determine the second pan-tilt camera corresponding to the maximum ratio among the ratios as the target second pan-tilt camera; Obtaining a third coordinate of the center position of the second partial monitoring area corresponding to the second current position of the target second pan-tilt camera, and obtaining a fourth coordinate of the center position of the local area; Determine the displacement direction between the third coordinate and the fourth coordinate as a third movement direction; it can be understood that the third movement direction refers to the direction from the third coordinate to the fourth coordinate; The target second pan-tilt camera is controlled to move along the third moving direction until the monitoring area of the target second pan-tilt camera completely includes the local area.

[0041] The method provided in this embodiment, on the one hand, adopts area ratio calculation and maximum ratio screening mechanism to realize the reasonable allocation of monitoring resources in the multi-pan-tilt camera system, which not only ensures tracking continuity but also avoids resource waste caused by redundant startup of multi-pan-tilt cameras. On the other hand, when the tracking target has not completely left the first global monitoring area, turning on the second pan-tilt camera of the target helps to prevent tracking loss.

[0042] In some embodiments, the method further comprises the following steps: After the preset time period, the first pan-tilt camera is turned off.

[0043] The method provided in this embodiment automatically turns off the first pan-tilt camera after the preset tracking time period, thereby effectively reducing system energy consumption while avoiding tracking loss.

[0044] See also Figure 2 , Figure 2 A schematic block diagram of the structure of the video target tracking system 100 of the intelligent pan-tilt platform provided in the embodiment of the present application is shown as follows: Figure 2 As shown, the video target tracking system 100 of the intelligent pan-tilt platform provided in the embodiment of the present application includes: The generating module 110 is configured to obtain video stream data collected by the first pan-tilt camera at the first current position, and generate a target monitoring area of the tracking target within a preset time period based on the video stream data.

[0045] The first determining module 120 is configured to determine whether the target monitoring area is included in the first global monitoring area of the first PTZ camera.

[0046] The second judgment module 130 is configured to judge whether the target monitoring area is included in the first partial monitoring area corresponding to the first pan-tilt camera at the first current position when the target monitoring area is included in the first global monitoring area.

[0047] The control module 140 is configured to control the first pan-tilt camera to maintain the first current position within the preset time period when the target monitoring area is included in the first partial monitoring area.

[0048] The adjustment module 150 is configured to adjust the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area when the target monitoring area is not included in the first partial monitoring area.

[0049] The determination module 160 is used to determine the target second pan-tilt camera based on the first global monitoring area and the target monitoring area when the target monitoring area is not included in the first global monitoring area, and to start the target second pan-tilt camera within the preset time period.

[0050] It should be noted that, those skilled in the art will clearly understand that for the sake of convenience and brevity in description, the specific working processes of the above-described system and each module can refer to the processes in the aforementioned embodiment of the video target tracking method of the intelligent pan-tilt head, and will not be repeated here.

[0051] The video target tracking system 100 of the intelligent PTZ provided in the above embodiment can be implemented in the form of a computer program. The computer program can be used in Figure 3 The system runs on the terminal device 200 shown.

[0052] See also Figure 3 , Figure 3 This is a schematic block diagram of the structure of a terminal device 200 provided in an embodiment of the present application. The terminal device 200 includes a processor 201 and a memory 202. The processor 201 and the memory 202 are connected via a system bus 203, wherein the memory 202 may include a non-volatile storage medium and an internal memory.

[0053] The non-volatile storage medium can store a computer program. The computer program includes program instructions, and when the program instructions are executed by the processor 201, the processor 201 can execute any of the above-mentioned video target tracking methods of the intelligent pan-tilt platform.

[0054] The processor 201 is used to provide computing and control capabilities to support the operation of the entire terminal device 200.

[0055] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor 201, the processor 201 can execute any of the above-mentioned video target tracking methods of the intelligent pan-tilt head.

[0056] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the terminal device 200 involved in the solution of the present application. The specific terminal device 200 may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0057] It should be understood that the processor 201 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0058] In some embodiments, the processor 201 is configured to execute a computer program stored in the memory to implement the following steps: Obtaining video stream data collected by a first pan-tilt camera at a first current position, and generating a target monitoring area for a tracking target within a preset time period based on the video stream data; Determining whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera; If it is included in the first global monitoring area, determining whether the target monitoring area is included in the first partial monitoring area corresponding to the first PTZ camera at the first current position; If it is included in the first partial monitoring area, controlling the first pan-tilt camera to maintain the first current position within the preset time period; If it is not included in the first partial monitoring area, adjusting the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area; If it is not included in the first global monitoring area, a target second pan-tilt camera is determined based on the first global monitoring area and the target monitoring area, and the target second pan-tilt camera is started within the preset time period.

[0059] It should be noted that, those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the terminal device 200 described above can refer to the corresponding process of the video target tracking method of the aforementioned intelligent pan-tilt head, and will not be repeated here.

[0060] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by one or more processors, the one or more processors implement the video target tracking method of the intelligent pan-tilt head provided in the embodiment of the present application.

[0061] The computer-readable storage medium may be an internal storage unit of the terminal device 200 in the aforementioned embodiment, such as a hard disk or memory of the terminal device 200. The computer-readable storage medium may also be an external storage device of the terminal device 200, such as a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc. equipped with the terminal device 200.

[0062] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A video target tracking method for an intelligent PTZ, characterized in that: include: Obtaining video stream data collected by a first pan-tilt camera at a first current position, and generating a target monitoring area for a tracking target within a preset time period based on the video stream data; Determining whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera; If it is included in the first global monitoring area, determining whether the target monitoring area is included in the first partial monitoring area corresponding to the first PTZ camera at the first current position; If it is included in the first partial monitoring area, controlling the first pan-tilt camera to maintain the first current position within the preset time period; If it is not included in the first partial monitoring area, adjusting the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area; If it is not included in the first global monitoring area, a target second pan-tilt camera is determined based on the first global monitoring area and the target monitoring area, and the target second pan-tilt camera is started within the preset time period.

2. The video target tracking method of the intelligent PTZ according to claim 1, characterized in that: The generating of a target monitoring area of a tracking target within a preset time period based on the video stream data includes: Respectively obtaining a position of the tracking target corresponding to each video frame in the video stream data; For two adjacent image frames in the video stream data, calculating a distance between positions corresponding to the two adjacent image frames and a first moving direction of a tracking target from a position corresponding to a previous image frame to a position corresponding to a next image frame in the two adjacent image frames; determining the maximum distance among the distances as the target distance; Sequentially obtaining time differences between capture moments corresponding to adjacent video frames in the video stream data, and determining the minimum time difference among the time differences as a target time difference; Determining a ratio of the preset time period to the target time difference as a target value; For each of the first moving directions, determining the product of the target distance and the target value as the moving distance corresponding to the tracked target in the first moving direction, and determining the target node corresponding to the moving distance in the first moving direction with the first current position of the tracked target as the starting node; The starting node is sequentially connected to each target node in a clockwise direction to obtain a target area, and a circumscribed circle of the target area is determined as the target monitoring area.

3. The video target tracking method of the intelligent PTZ according to claim 2, characterized in that: The respectively obtaining the position of each video frame corresponding to the tracking target in the video stream data includes: For any video frame in the video stream data, determining whether the video frame is the first video frame in the video stream data; If it is the first video frame, perform global detection on the video frame using a preset target detection algorithm to obtain a first confidence score for each object in the video frame, and determine the position corresponding to the object with the highest first confidence score as the position of the tracking target; If it is not the first video frame, respectively obtaining similarities between each object in the video frame and the tracked target detected in the first video frame, and determining whether a maximum similarity among the similarities is greater than a preset similarity; If the similarity is greater than a preset similarity, determining the position of the object corresponding to the maximum similarity as the position of the tracking target; If the similarity is not greater than a preset similarity, the video frame is cropped to obtain an initial cropped image; the center position of the initial cropped image is the position corresponding to the tracking target in the video frame before the video frame, and the size of the initial cropped image is twice the circumscribed rectangle of the image corresponding to the tracking target in the first video frame; The size of the initial cropped image is adjusted to be consistent with the size of the video frame to obtain a target cropped image, and a preset target detection algorithm is used to perform global detection on the target cropped image to obtain a second confidence level of each object in the target cropped image, and the position corresponding to the object with the highest second confidence level is determined as the position of the tracking target.

4. The video target tracking method of the intelligent PTZ according to claim 1, characterized in that: The adjusting the position of the first pan-tilt camera within the preset time period based on the first partial monitoring area and the target monitoring area includes: Obtaining first coordinates of a center position of the first partial monitoring area, and obtaining second coordinates of a center position of the monitoring area; determining a displacement direction between the first coordinate and the second coordinate as a second movement direction; The first pan-tilt camera is controlled to move along the second moving direction until the monitoring area of the first pan-tilt camera is the target monitoring area.

5. The video target tracking method of the intelligent PTZ according to claim 1, characterized in that: The determining of the target second pan-tilt camera based on the first global monitoring area and the target monitoring area, and starting the target second pan-tilt camera within the preset time period, includes: Dividing the target monitoring area into a first target monitoring sub-area and a second target monitoring sub-area based on the first global monitoring area; wherein the first target monitoring sub-area is included in the first global monitoring area, and the second target monitoring sub-area is not included in the first global monitoring area; Obtain the second global monitoring area corresponding to each second PTZ camera; For each of the second PTZ cameras, calculating a ratio between an area of a local area of the second target monitoring sub-region located in the second global monitoring area of the second PTZ camera and a total area of the second target monitoring sub-region; Determine the second pan-tilt camera corresponding to the maximum ratio among the ratios as the target second pan-tilt camera; Obtaining a third coordinate of the center position of the second partial monitoring area corresponding to the second current position of the target second pan-tilt camera, and obtaining a fourth coordinate of the center position of the local area; determining a displacement direction between the third coordinate and the fourth coordinate as a third movement direction; The target second pan-tilt camera is controlled to move along the third moving direction until the monitoring area of the target second pan-tilt camera completely includes the local area.

6. The video target tracking method of the intelligent PTZ according to claim 1, characterized in that: The method further comprises: After the preset time period, the first pan-tilt camera is turned off.

7. A video target tracking system for an intelligent PTZ, characterized in that: include: a generating module, configured to obtain video stream data collected by the first pan-tilt camera at the first current position, and generate a target monitoring area for tracking the target within a preset time period based on the video stream data; A first judgment module is used to judge whether the target monitoring area is included in the first global monitoring area of the first pan-tilt camera; a second judgment module, configured to judge, when the target monitoring area is included in the first global monitoring area, whether the target monitoring area is included in a first partial monitoring area corresponding to the first PTZ camera at the first current position; a control module, configured to control the first pan-tilt camera to maintain the first current position within the preset time period when the target monitoring area is included in the first partial monitoring area; an adjustment module, configured to adjust the position of the first PTZ camera within the preset time period based on the first partial monitoring area and the target monitoring area when the target monitoring area is not included in the first partial monitoring area; A determination module is used to determine a target second pan-tilt camera based on the first global monitoring area and the target monitoring area when the target monitoring area is not included in the first global monitoring area, and to start the target second pan-tilt camera within the preset time period.

Images