Dynamic tracking monitoring system and method based on liquid lens

Abstract

The present invention provides a dynamic tracking monitoring system and method based on a liquid lens, wherein the system comprises: a target detection module, which is used to perform target detection on a real-time panoramic video based on a preset target recognition model, and determine whether a new tracking target appears in a monitoring area; a task scheduling module, which is used to assign the new tracking target to a target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in a target tracking device group after confirming that a new tracking target appears in the monitoring area; a target locking module, which is used to control the target lens to lock the new tracking target based on the position information of the new tracking target; and an intelligent tracking module, which is used to control the target lens to adjust the lens parameters according to the movement of the tracking target to obtain a dynamic tracking image. The present invention realizes the precise tracking of the tracking target in the monitoring area and the automatic separation of the tracking target image.

Description

Technical Field

[0001] The present invention relates to the field of intelligent tracking technology, and in particular to a liquid lens-based dynamic tracking monitoring system and method. Background Art

[0002] With the development of information technology, intelligent monitoring systems are widely used in production and life, and video monitoring is gradually spreading to every corner of our lives. Existing video monitoring technologies mostly acquire all images in the monitoring area indiscriminately to achieve full-area monitoring of the monitoring area. Therefore, it is impossible to achieve accurate tracking of multiple monitoring targets. When the monitoring management end views a certain monitoring target, it is necessary to manually select and enlarge the monitoring target to view the details of the monitoring target, and it is impossible to achieve automatic segmentation of the monitoring video of the monitoring target. Therefore, the present invention proposes a dynamic tracking monitoring system and method based on a liquid lens to at least solve the above-mentioned shortcomings. Summary of the invention

[0003] The present invention provides a dynamic tracking monitoring system and method based on a liquid lens, which realizes accurate tracking of tracking targets in a monitoring area and intelligent allocation of tracking tasks within a tracking device group, and adaptively adjusts the target lens parameters in a timely manner after completing the tracking task allocation, so that each tracking target in the target lens can be clearly photographed, and the tracking target image is automatically separated while accurately tracking the tracking target. When viewing the monitoring video corresponding to a single tracking target, the management personnel only needs to click or input the tracking target information to obtain the tracking image corresponding to the tracking target in the monitoring area, thereby reducing the monitoring query task volume of the management personnel and improving the monitoring query efficiency.

[0004] The present invention provides a dynamic tracking monitoring system based on a liquid lens, characterized in that it comprises:

[0005] The target detection module is used to detect targets in real-time panoramic video based on a preset target recognition model to determine whether a new tracking target appears in the monitoring area;

[0006] A task scheduling module is used to assign the new tracking target to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group after confirming that a new tracking target appears in the monitoring area;

[0007] A target locking module, used to control the target lens to lock the new tracking target based on the position information of the new tracking target;

[0008] The intelligent tracking module is used to control the target lens to adjust the lens parameters according to the movement of the tracked target to obtain dynamic tracking images.

[0009] Preferably, in a dynamic tracking monitoring system based on a liquid lens, the task scheduling module includes:

[0010] An information acquisition unit, a user acquires the current working status of multiple liquid lenses included in the target tracking device group, determines whether there is an idle liquid lens, and if so, adds an idle tag to the idle liquid lens;

[0011] A first allocation unit is used to obtain the first position coordinates of the new tracking target in the monitoring area, and obtain the position information of each idle liquid lens, and allocate the new tracking target to the nearest idle liquid lens according to the principle of proximity based on the real-time position information and the position coordinates, and use the nearest idle liquid lens as the target lens in real time;

[0012] The second allocation unit is used to obtain the second position coordinates of the monitoring and tracking targets corresponding to each liquid lens in the monitoring area when there are no idle liquid lenses in the target tracking device group, compare the first position coordinates with each second position coordinate respectively to obtain multiple coordinate distances, and use the liquid lens corresponding to the monitoring and tracking target corresponding to the second position coordinate corresponding to the minimum coordinate distance as the target lens, and allocate the new tracking target as the secondary tracking target to the target lens.

[0013] Preferably, in a dynamic tracking monitoring system based on a liquid lens, the second allocation unit comprises:

[0014] The tag adding subunit is used to determine the first tracking target of the target lens, add a priority tracking target tag to the first tracking target, and add a secondary tracking target tag to the remaining monitoring tracking targets of the target lens, wherein the remaining monitoring tracking targets include the new tracking target;

[0015] The label updating subunit is used to update the target labels of all monitoring and tracking targets in the current tracking picture when the number of tracking targets of the target lens changes;

[0016] When the priority tracking target leaves the monitoring area, the secondary tracking target closest to the lens center in the current tracking image is taken as the priority tracking target, and the target label is updated.

[0017] Preferably, in a dynamic tracking monitoring system based on a liquid lens, a target locking module comprises:

[0018] A task confirmation unit, used to determine the total number of tracking targets of the target lens corresponding to the new tracking target;

[0019] A locking control unit, used for, when the total number of tracking targets of the target lens is 1, taking the new tracking target as the lens focus center and locking the new tracking target based on the rectangular frame;

[0020] When the total number of tracking targets of the target lens is not 1, the target center is determined based on the coordinate positions of each tracking target corresponding to the target lens, the target center is used as the lens focusing center, and each tracking target is locked based on the rectangular frame.

[0021] Preferably, in a dynamic tracking monitoring system based on a liquid lens, a locking control unit comprises:

[0022] The position monitoring subunit is used to obtain the real-time distance between the first position coordinate of the secondary tracking target and the second position coordinate of the corresponding priority tracking target in real time when the total number of tracking targets of the target lens is not 1, and when the real-time distance is greater than the preset distance, control the second allocation unit to reallocate the secondary tracking target.

[0023] Preferably, in a dynamic tracking monitoring system based on a liquid lens, the intelligent tracking module comprises:

[0024] A parameter adjustment unit, for adjusting lens parameters of the target lens based on the number of tracking targets corresponding to the target lens and the lens focus center, with reference to preset dynamic tracking image requirements;

[0025] A first tracking unit is used to obtain a first dynamic tracking image corresponding to the priority tracking target, and send the first dynamic tracking image to a display position corresponding to the target number carried by the priority tracking target for display;

[0026] The second tracking unit is used to obtain a second dynamic tracking image corresponding to a target lens corresponding to a secondary tracking target, and send the second tracking video to a display position corresponding to a target number carried by the secondary tracking target for display.

[0027] Preferably, in a dynamic tracking monitoring system based on a liquid lens, the second tracking unit comprises:

[0028] The recording subunit is used to record the motion trajectory of the secondary tracking target in the monitoring area, generate a marked monitoring switching node according to the target lens allocation information corresponding to the secondary tracking target, and mark the monitoring switching node mark on the motion trajectory;

[0029] The acquisition subunit is used to determine the target lens currently corresponding to the secondary tracking target based on the monitoring switching node, acquire the real-time picture, and pre-process the real-time picture to obtain the current tracking sub-video with the secondary tracking target as the video center;

[0030] The splicing subunit is used to monitor that when the switching node is triggered, obtain the first background feature and the first target feature corresponding to the initial frame image of the current tracking sub-video, and the second background feature and the second target feature corresponding to the end frame image of the previous tracking sub-video;

[0031] Comparing the first background feature with the second background feature, determining the background association relationship between the current tracking sub-video and the previous tracking sub-video, determining the associated background, and generating an image change vector based on the change trend of the overlapping part of the associated background on the end frame image and the initial frame image;

[0032] At the same time, based on the first target feature generation and the second target feature comparison, the action change corresponding to the secondary tracking target on the end frame image and the initial frame image is determined, and whether the connection action of the secondary tracking target is coherent;

[0033] If they are coherent, then the associated background is spliced ​​based on the image change vector to generate a second dynamic tracking image;

[0034] Otherwise, based on the motion change feature, similar motion query is performed in the acquired tracking sub-video of the secondary tracking target to obtain the best similar motion, obtain the motion vector corresponding to the best similar motion, and perform motion supplement on the connecting motion of the secondary tracking target according to the motion vector to obtain a coherent connecting motion;

[0035] According to the graphic change vector, the related background is fused and spliced, the video splicing node is determined, and the coherent connection action is inserted into the video splicing node to replace the connection action, so as to generate a second dynamic tracking image.

[0036] Preferably, in a dynamic tracking monitoring system based on a liquid lens, the second tracking unit further includes:

[0037] The data association subunit is used to determine the priority tracking target corresponding to each movement stage of the secondary tracking target according to the marking result on the movement trajectory of the secondary tracking target as the associated tracking target;

[0038] Based on the shot switching time of each motion stage, determine the video association time period of the associated tracking target and the secondary tracking target, and based on the video association time period, establish a mapping association relationship between each tracking sub-video corresponding to the second dynamic tracking image and the target time period corresponding to the first dynamic tracking image;

[0039] The video jump subunit is used to obtain the corresponding first dynamic tracking image segment based on the mapping association relationship corresponding to the click time period corresponding to the click of the display end when the display end clicks to view the original video of the second dynamic tracking image.

[0040] Preferably, in a dynamic tracking monitoring system based on a liquid lens, it also includes:

[0041] The remote control module is used to remotely adjust the tracking parameters of each liquid lens in the target tracking device group, and generate lens parameter adjustment instructions according to the remote adjustment results of the tracking parameters, synchronously adjust the liquid lens parameters, and obtain a dynamic tracking image of the target magnification.

[0042] The present invention provides a dynamic tracking monitoring method based on a liquid lens, comprising:

[0043] Step 1: Based on the preset target recognition model, target detection is performed on the real-time panoramic video to determine whether there is a new tracking target in the monitoring area;

[0044] Step 2: after confirming that a new tracking target appears in the monitoring area, the new tracking target is assigned to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group;

[0045] Step 3: Based on the position information of the new tracking target, control the target lens to lock the new tracking target;

[0046] Step 4: Control the eyepiece lens to adjust lens parameters according to the movement of the tracking target to obtain dynamic tracking images.

[0047] Compared with the prior art, the present invention has at least the following beneficial effects:

[0048] Beneficial effects: The present invention uses a target detection module to perform target detection on real-time panoramic video based on a preset target recognition model, and determines whether there is a new tracking target appearing in the monitoring area, thereby realizing intelligent recognition of the tracking target; after confirming that a new tracking target appears in the monitoring area, the task scheduling module assigns the new tracking target to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group, thereby realizing intelligent scheduling of each liquid lens in the monitoring system, and obtaining tracking images of multiple tracking targets through limited tracking lenses, thereby improving the utilization rate of the tracking lens; then, the target locking module controls the target lens to lock the new tracking target based on the position information of the new tracking target, thereby helping to improve the accuracy of the tracking target and ensure that each tracking target is tracked by the lens, and also helps to track a single tracking target in a multi-tracking target state. The successful separation of the image, through the intelligent tracking module to control the target lens to adjust the lens parameters according to the movement of the tracking target, ensures that the video quality of the dynamic tracking image of each tracking target is maintained at a certain level, and shows good monitoring video to the management personnel. The present invention realizes the precise tracking of the tracking target in the monitoring area and the intelligent allocation of tracking tasks in the tracking equipment group, and adaptively adjusts the target lens parameters in time after completing the tracking task allocation, ensuring that each tracking target in the target lens can be clearly photographed. While accurately tracking the tracking target, the automatic separation of the tracking target image is realized. When the management personnel views the monitoring video corresponding to a single tracking target, they only need to click or enter the tracking target information to obtain the tracking image corresponding to the tracking target in the monitoring area, reducing the monitoring query task volume of the management personnel and improving the monitoring query efficiency.

[0049] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in this application document.

[0050] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0052] Figure 1 is a schematic diagram of a dynamic tracking monitoring system based on a liquid lens according to the present invention;

[0053] Figure 2 It is a schematic diagram of a task scheduling module of a dynamic tracking monitoring system based on a liquid lens of the present invention;

[0054] Figure 3 It is a schematic diagram of a target locking module of a dynamic tracking monitoring system based on a liquid lens of the present invention;

[0055] Figure 4 It is a schematic diagram of an intelligent tracking module of a dynamic tracking monitoring system based on a liquid lens of the present invention;

[0056] Figure 5 The present invention is a flow chart of a dynamic tracking and monitoring method based on a liquid lens. DETAILED DESCRIPTION

[0057] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0058] Embodiment 1:

[0059] The present invention provides a dynamic tracking monitoring system based on a liquid lens, such as Figure 1 As shown, including:

[0060] The target detection module is used to detect targets in real-time panoramic video based on a preset target recognition model to determine whether a new tracking target appears in the monitoring area;

[0061] A task scheduling module is used to assign the new tracking target to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group after confirming that a new tracking target appears in the monitoring area;

[0062] A target locking module, used to control the target lens to lock the new tracking target based on the position information of the new tracking target;

[0063] The intelligent tracking module is used to control the target lens to adjust the lens parameters according to the movement of the tracked target to obtain dynamic tracking images.

[0064] In this embodiment, the preset target recognition model refers to a trained target recognition model, and recognition training of different tracking targets can be performed according to the target model corresponding to the usage environment, so as to achieve the purpose of recognizing multiple tracking targets.

[0065] In this embodiment, the dynamic tracking system includes a wide-angle monitoring device group and a target tracking device group. The wide-angle monitoring device group is used to perform regional monitoring of the monitoring area, and the target tracking device group is used to accurately track the tracking target.

[0066] In this embodiment, the target lens refers to a liquid lens that matches the tracking target.

[0067] The beneficial effects of the above technical solution are as follows: the present invention performs target detection on real-time panoramic video based on a preset target recognition model, determines whether there is a new tracking target appearing in the monitoring area, and realizes intelligent recognition of the tracking target; after confirming that a new tracking target appears in the monitoring area, based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group, the new tracking target is assigned to the target lens for tracking, thereby realizing intelligent scheduling of each liquid lens in the monitoring system, and obtaining tracking images of multiple tracking targets through limited tracking lenses, thereby improving the utilization rate of the tracking lens; then, based on the position information of the new tracking target, the target lens is controlled to lock on the new tracking target, which is beneficial to improving the accuracy of the tracking target and ensuring that each tracking target is tracked by the lens. At the same time, it is also beneficial to the successful separation of a single tracking target from the image in the state of multiple tracking targets. The intelligent tracking module controls the target lens to adjust the lens parameters according to the movement of the tracking target, ensuring that the video quality of the dynamic tracking image of each tracking target is maintained at a certain level, providing good monitoring video for management personnel to display. The present invention realizes the precise tracking of the tracking target in the monitoring area and the intelligent allocation of tracking tasks in the tracking equipment group, and adaptively adjusts the target lens parameters in time after completing the tracking task allocation, ensuring that each tracking target in the target lens can be clearly photographed. While accurately tracking the tracking target, the automatic separation of the tracking target image is realized. When the management personnel view the monitoring video corresponding to a single tracking target, they only need to click or enter the tracking target information to obtain the tracking image corresponding to the tracking target in the monitoring area, reducing the monitoring query task volume of the management personnel and improving the monitoring query efficiency.

[0068] Embodiment 2:

[0069] Based on Example 1, the task scheduling module, such as Figure 2 As shown, including:

[0070] An information acquisition unit, a user acquires the current working status of multiple liquid lenses included in the target tracking device group, determines whether there is an idle liquid lens, and if so, adds an idle tag to the idle liquid lens;

[0071] A first allocation unit is used to obtain the first position coordinates of the new tracking target in the monitoring area, and obtain the position information of each idle liquid lens, and allocate the new tracking target to the nearest idle liquid lens according to the principle of proximity based on the real-time position information and the position coordinates, and use the nearest idle liquid lens as the target lens in real time;

[0072] The second allocation unit is used to obtain the second position coordinates of the monitoring and tracking targets corresponding to each liquid lens in the monitoring area when there are no idle liquid lenses in the target tracking device group, compare the first position coordinates with each second position coordinate respectively to obtain multiple coordinate distances, and use the liquid lens corresponding to the monitoring and tracking target corresponding to the second position coordinate corresponding to the minimum coordinate distance as the target lens, and allocate the new tracking target as the secondary tracking target to the target lens.

[0073] In this embodiment, the current working state includes both a state of having a tracking target and a state of not having a tracking target.

[0074] In this embodiment, the idle liquid lens refers to a liquid lens in the target tracking device group that does not track a target.

[0075] In this embodiment, the first position coordinates refer to the coordinates of the new tracking target in the preset coordinate system within the monitoring area. The second position coordinates refer to the coordinates of each liquid lens in the target tracking device group in the preset coordinate system within the monitoring area.

[0076] In this embodiment, the coordinate distance refers to the position distance between the new tracking target and each liquid lens in the target tracking device group.

[0077] In this embodiment, the minimum coordinate distance refers to the minimum value among multiple coordinate distances.

[0078] The beneficial effects of the above technical solution are as follows: when a new tracking target is found, the present invention first determines the working state of each liquid lens in the target tracking device group, and preferably allocates the new tracking target to an idle liquid lens, effectively shortening the response efficiency of tracking task scheduling while improving the tracking accuracy of the tracking target, and when there is no idle liquid lens in the target tracking device group, the new tracking target is allocated according to the coordinate distance between the new tracking target and the liquid lens, ensuring the acquisition of high-quality dynamic tracking images of the new tracking target while reducing the impact of the addition of the tracking task of the new tracking target on the tracking task of the original tracking target of the target lens, and realizing the joint monitoring of multiple tracking targets. .

[0079] Embodiment 3:

[0080] On the basis of Embodiment 2, the second allocation unit includes:

[0081] The tag adding subunit is used to determine the first tracking target of the target lens, add a priority tracking target tag to the first tracking target, and add a secondary tracking target tag to the remaining monitoring tracking targets of the target lens, wherein the remaining monitoring tracking targets include the new tracking target;

[0082] The label updating subunit is used to update the target labels of all monitoring and tracking targets in the current tracking picture when the number of tracking targets of the target lens changes;

[0083] When the priority tracking target leaves the monitoring area, the secondary tracking target closest to the lens center in the current tracking image is taken as the priority tracking target, and the target label is updated.

[0084] In this embodiment, the first tracking target refers to the tracking target that is first determined as the tracking object of the target lens in the current tracking picture when there are multiple tracking targets for the target lens.

[0085] In this embodiment, the target tag includes a priority tracking target and a secondary tracking target.

[0086] In this embodiment, the remaining monitoring tracking targets refer to other tracking targets in the target shot that process the first tracking target.

[0087] The beneficial effects of the above technical solution are as follows: the present invention determines the first tracking target of the target lens, adds a priority tracking target tag to the first tracking target, and adds a secondary tracking target tag to the remaining monitoring tracking targets of the target lens, determines a main tracking object (priority tracking target) for the tracking of the target lens, ensures that in time when the running speeds or motion trajectories of the first tracking target and the secondary tracking target are different, the target lens can also independently complete the selection of the tracking target, ensures that the running trajectory of the main tracking target in the monitoring area can be completely tracked, and also facilitates the switching of the target lens corresponding to the secondary tracking target; through the label updating subunit, when the number of tracking targets of the target lens changes, the target tags of all monitoring tracking targets in the current tracking screen are updated; when the priority tracking target leaves the monitoring area, the secondary tracking target closest to the lens center in the current tracking screen is used as the priority tracking target, and the target tag is updated, thereby reducing the number of switching times of the target lens corresponding to the secondary tracking target and reducing the system data processing pressure.

[0088] Embodiment 4:

[0089] Based on Example 1, the target locking module, such as Figure 3 As shown, including:

[0090] A task confirmation unit, used to determine the total number of tracking targets of the target lens corresponding to the new tracking target;

[0091] A locking control unit, used for, when the total number of tracking targets of the target lens is 1, taking the new tracking target as the lens focus center and locking the new tracking target based on the rectangular frame;

[0092] When the total number of tracking targets of the target lens is not 1, the target center is determined based on the coordinate positions of each tracking target corresponding to the target lens, the target center is used as the lens focusing center, and each tracking target is locked based on the rectangular frame.

[0093] In this embodiment, the total number of tracking targets refers to the number of tracking targets allocated to the target lens.

[0094] In this embodiment, the target center refers to the center of a line segment or a geometric figure formed by connecting the tracking targets in the target lens.

[0095] In this embodiment, the lens focusing center refers to the shooting center of the liquid lens.

[0096] The beneficial effect of the above technical solution is as follows: the present invention determines the lens focus center of the target lens according to the total number of tracking targets of the target lens corresponding to the new tracking target, and locks the tracking target in the lens to ensure that the target lens can capture all tracking targets.

[0097] Embodiment 5:

[0098] On the basis of Embodiment 3, the locking control unit comprises:

[0099] The position monitoring subunit is used to obtain the real-time distance between the first position coordinate of the secondary tracking target and the second position coordinate of the corresponding priority tracking target in real time when the total number of tracking targets of the target lens is not 1, and when the real-time distance is greater than the preset distance, control the second allocation unit to reallocate the secondary tracking target.

[0100] The beneficial effect of the above technical solution is as follows: when the total number of tracking targets of the target lens is not 1, the present invention obtains in real time the real-time distance between the first position coordinate of the secondary tracking target and the second position coordinate of its corresponding priority tracking target; when the real-time distance is greater than the preset distance, the second allocation unit is controlled to reallocate the secondary tracking target, providing a basis for switching the tracking screen of the secondary tracking target, ensuring that when the running speed and trajectory of the secondary tracking target are greatly different from those of the preferred tracking target, a complete dynamic image can be obtained for the secondary tracking target in the monitoring area.

[0101] Embodiment 6:

[0102] Based on Example 1, the intelligent tracking module, such as Figure 4 As shown, including:

[0103] A parameter adjustment unit, for adjusting lens parameters of the target lens based on the number of tracking targets corresponding to the target lens and the lens focus center, with reference to preset dynamic tracking image requirements;

[0104] A first tracking unit is used to obtain a first dynamic tracking image corresponding to the priority tracking target, and send the first dynamic tracking image to a display position corresponding to the target number carried by the priority tracking target for display;

[0105] The second tracking unit is used to obtain a second dynamic tracking image corresponding to a target lens corresponding to a secondary tracking target, and send the second tracking video to a display position corresponding to a target number carried by the secondary tracking target for display.

[0106] In this embodiment, the preset dynamic tracking image requirements include but are not limited to default requirements for the clarity and scaling of the tracking image, which can be set by the management end according to actual needs.

[0107] In this embodiment, the first tracking image refers to a tracking video corresponding to a priority tracking target. When the target tag of the current secondary tracking target is updated to the priority tracking target, the secondary tracking target is sent to the first tracking unit, and the tracking of the secondary tracking target continues, and the display position does not change.

[0108] In this embodiment, the second dynamic tracking image refers to a tracking video corresponding to a secondary tracking target.

[0109] The beneficial effects of the above technical solution are as follows: the present invention uses a parameter adjustment unit to adjust the lens parameters of the target lens based on the number of tracking targets corresponding to the target lens and the lens focusing center, and refers to the preset dynamic tracking image requirements to ensure that the dynamic tracking image obtained meets the default requirements, provide high-quality tracking pictures for the monitoring management end, and separately collect the corresponding priority tracking targets and secondary tracking targets, providing a basis for separating the tracking target stars.

[0110] Embodiment 7:

[0111] On the basis of Embodiment 6, the second tracking unit includes:

[0112] The recording subunit is used to record the motion trajectory of the secondary tracking target in the monitoring area, generate a marked monitoring switching node according to the target lens allocation information corresponding to the secondary tracking target, and mark the monitoring switching node mark on the motion trajectory;

[0113] The acquisition subunit is used to determine the target lens currently corresponding to the secondary tracking target based on the monitoring switching node, acquire the real-time picture, and pre-process the real-time picture to obtain the current tracking sub-video with the secondary tracking target as the video center;

[0114] The splicing subunit is used to monitor that when the switching node is triggered, obtain the first background feature and the first target feature corresponding to the initial frame image of the current tracking sub-video, and the second background feature and the second target feature corresponding to the end frame image of the previous tracking sub-video;

[0115] Comparing the first background feature with the second background feature, determining the background association relationship between the current tracking sub-video and the previous tracking sub-video, determining the associated background, and generating an image change vector based on the change trend of the overlapping part of the associated background on the end frame image and the initial frame image;

[0116] At the same time, based on the first target feature generation and the second target feature comparison, the action change corresponding to the secondary tracking target on the end frame image and the initial frame image is determined, and whether the connection action of the secondary tracking target is coherent;

[0117] If they are coherent, then the associated background is spliced ​​based on the image change vector to generate a second dynamic tracking image;

[0118] Otherwise, based on the motion change feature, similar motion query is performed in the acquired tracking sub-video of the secondary tracking target to obtain the best similar motion, obtain the motion vector corresponding to the best similar motion, and perform motion supplement on the connecting motion of the secondary tracking target according to the motion vector to obtain a coherent connecting motion;

[0119] According to the graphic change vector, the related background is fused and spliced, the video splicing node is determined, and the coherent connection action is inserted into the video splicing node to replace the connection action, so as to generate a second dynamic tracking image.

[0120] In this embodiment, the monitoring switching node refers to a trajectory node of the target lens switching corresponding to the secondary tracking target.

[0121] In this embodiment, the target lens allocation information refers to the switching information of the corresponding target lens during the movement of the secondary tracking target in the monitoring area.

[0122] In this embodiment, the real-time picture refers to the dynamic picture collected by the secondary tracking target in the current corresponding target lens.

[0123] In this embodiment, the current tracking sub-video refers to a video segment obtained after preprocessing of the target shot currently corresponding to the secondary tracking target, and the video segment takes the secondary tracking target as the center of the video screen.

[0124] In this embodiment, preprocessing refers to processing the real-time picture so that the level tracking target is the center of the video picture.

[0125] In this embodiment, the initial frame image refers to an image corresponding to the first video frame of the current tracking sub-video.

[0126] In this embodiment, the end frame image refers to an image corresponding to the first video frame of the previous tracking sub-video.

[0127] In this embodiment, the associated background refers to other identical objects in the initial frame image and the end frame image except for the tracking target.

[0128] In this embodiment, the change trend refers to the change in the form of the associated background during the process of switching from the end frame image to the initial frame image;

[0129] The image change vector is used to represent the direction in which the morphology of the associated background changes during the process of switching from the ending frame image to the initial frame image.

[0130] In this embodiment, the first background feature refers to the shape, color, etc. of other objects in the initial frame image and the ending frame image except the tracking target; the second background feature refers to the shape, color, etc. of other objects in the ending frame image and the ending frame image except the tracking target.

[0131] In this embodiment, the first target feature refers to the motion feature of the secondary tracking target in the initial frame image; the second target feature refers to the motion feature of the secondary tracking target in the ending frame image.

[0132] In this embodiment, the connection action refers to the action of the secondary tracking target during the process of switching from the end frame image to the initial frame image.

[0133] In this embodiment, the motion change feature refers to the change feature of the secondary tracking target motion during the process of switching from the end frame image to the initial frame image.

[0134] In this embodiment, the video mosaic node refers to the time node corresponding to the fusion of the end frame image and the initial frame image.

[0135] In this embodiment, the continuous connecting action refers to a new continuous action obtained by supplementing the connecting action with an action transformation process according to a similar action with the highest similarity to the connecting action in the existing tracking image of the secondary tracking target.

[0136] The beneficial effects of the above technical solution are as follows: the present invention records the movement trajectory of the secondary tracking target in the monitoring area through the recording subunit, generates a marked monitoring switching node according to the target lens allocation information corresponding to the secondary tracking target, and marks the monitoring switching node on the movement trajectory, thereby realizing the recording of the monitoring lens switching of the secondary tracking target, facilitating the establishment of subsequent mapping association relationships and original video query, and providing a basis for managers to manually query the original video. The monitoring videos corresponding to the secondary tracking targets collected by different target lenses are spliced ​​by the acquisition subunit and the splicing subunit to obtain the second dynamic tracking image, thereby realizing the autonomous separation of the monitoring videos of the tracking targets, facilitating the managers to conduct targeted viewing of all tracking targets in the monitoring area, effectively shortening the managers' search time for the targeted viewing targets, and improving the monitoring query efficiency.

[0137] Embodiment 8:

[0138] Based on Embodiment 7, the second tracking unit further includes:

[0139] The data association subunit is used to determine the priority tracking target corresponding to each movement stage of the secondary tracking target according to the marking result on the movement trajectory of the secondary tracking target as the associated tracking target;

[0140] Based on the shot switching time of each motion stage, determine the video association time period of the associated tracking target and the secondary tracking target, and based on the video association time period, establish a mapping association relationship between each tracking sub-video corresponding to the second dynamic tracking image and the target time period corresponding to the first dynamic tracking image;

[0141] The video jump subunit is used to obtain the corresponding first dynamic tracking image segment based on the mapping association relationship corresponding to the click time period corresponding to the click of the display end when the display end clicks to view the original video of the second dynamic tracking image.

[0142] In this embodiment, the associated tracking target refers to a priority tracking target that has used the same target lens as a secondary tracking target (including a secondary tracking target whose target tag is changed to a priority tracking target) in the same time period.

[0143] In this embodiment, the tracking sub-video refers to video segments acquired by different target lenses in different second dynamic tracking images.

[0144] In this embodiment, one target lens corresponds to one operation stage, that is, the motion trajectory between two monitoring switching nodes is one operation stage.

[0145] In this embodiment, the shot switching time includes the initial time and the end time of the motion phase.

[0146] In this embodiment, the video-associated time period refers to a time period corresponding to the difference between the initial time and the end time of the motion phase.

[0147] The beneficial effects of the above technical solution are as follows: the present invention determines the priority tracking target corresponding to each movement stage of the secondary tracking target according to the marking results on the motion trajectory of the secondary tracking target through the data association subunit, as the associated tracking target; based on the lens switching time of each movement stage, the video association time period of the associated tracking target and the secondary tracking target is determined, and based on the video association time period, a mapping association relationship is established between each tracking sub-video corresponding to the second dynamic tracking image and the target time period corresponding to the first dynamic tracking image, so as to establish a fast query method for managers to quickly view the original video through the video jump subunit, and at the same time, it also realizes the alternative retention of the original tracking video of the secondary tracking target, avoiding redundant backup to increase the storage pressure of the system, and when the manager has doubts about the tracking image, the original video can be quickly retrieved through the mapping association relationship, which shortens the time required for the manager to find the video and improves the efficiency of monitoring video management.

[0148] Embodiment 9:

[0149] Based on Example 1, a dynamic tracking monitoring system based on a liquid lens is characterized by further comprising:

[0150] The remote control module is used to remotely adjust the tracking parameters of each liquid lens in the target tracking device group, and generate lens parameter adjustment instructions according to the remote adjustment results of the tracking parameters, synchronously adjust the liquid lens parameters, and obtain a dynamic tracking image of the target magnification.

[0151] In this embodiment, the tracking parameters refer to various parameters required to be included in the preset dynamic tracking image.

[0152] The beneficial effects of the above technical solution are as follows: the present invention remotely adjusts the tracking parameters of each liquid lens in the target tracking device group through a remote control module, generates lens parameter adjustment instructions according to the remote adjustment results of the tracking parameters, synchronously adjusts the liquid lens parameters, obtains a dynamic tracking image of the target magnification, realizes remote control, completes system interaction, and facilitates the system to adjust the lens parameters in a timely manner according to the actual requirements of the management end, thereby improving the user experience.

[0153] Embodiment 10:

[0154] The present invention provides a dynamic tracking monitoring method based on a liquid lens, such as Figure 5 As shown, including:

[0155] Step 1: Based on the preset target recognition model, target detection is performed on the real-time panoramic video to determine whether there is a new tracking target in the monitoring area;

[0156] Step 2: after confirming that a new tracking target appears in the monitoring area, the new tracking target is assigned to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group;

[0157] Step 3: Based on the position information of the new tracking target, control the target lens to lock the new tracking target;

[0158] Step 4: Control the eyepiece lens to adjust lens parameters according to the movement of the tracking target to obtain dynamic tracking images.

[0159] The beneficial effects of the above technical solution are as follows: the present invention uses a target detection module to perform target detection on real-time panoramic video based on a preset target recognition model, and determines whether there is a new tracking target appearing in the monitoring area, thereby realizing intelligent recognition of the tracking target; after confirming that a new tracking target appears in the monitoring area, the task scheduling module allocates the new tracking target to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group, thereby realizing intelligent scheduling of each liquid lens in the monitoring system, and obtaining tracking images of multiple tracking targets through limited tracking lenses, thereby improving the utilization rate of tracking lenses; then, the target locking module controls the target lens to lock the new tracking target based on the position information of the new tracking target, thereby helping to improve the accuracy of the tracking target and ensure that each tracking target is tracked by the lens, and is also beneficial in the case of multiple tracking targets. A single tracking target is successfully separated from the image, and the target lens is controlled to adjust the lens parameters according to the movement of the tracking target, ensuring that the video quality of the dynamic tracking image of each tracking target is maintained at a certain level, and showing good monitoring videos to managers. The present invention realizes the precise tracking of tracking targets in the monitoring area and the intelligent allocation of tracking tasks within the tracking equipment group, and adaptively adjusts the target lens parameters in time after completing the tracking task allocation, ensuring that each tracking target in the target lens can be clearly photographed, and realizes automatic separation of the tracking target image while accurately tracking the tracking target. When viewing the monitoring video corresponding to a single tracking target, the manager only needs to click or enter the tracking target information to obtain the tracking image corresponding to the tracking target in the monitoring area, reducing the monitoring query task volume of the manager and improving the monitoring query efficiency.

[0160] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A dynamic tracking monitoring system based on liquid lens, characterized in that: include: The target detection module is used to detect targets in real-time panoramic video based on a preset target recognition model to determine whether a new tracking target appears in the monitoring area; A task scheduling module is used to assign the new tracking target to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group after confirming that a new tracking target appears in the monitoring area; A target locking module, used to control the target lens to lock the new tracking target based on the position information of the new tracking target; Intelligent tracking module, used to control the target lens to adjust the lens parameters according to the movement of the tracked target to obtain dynamic tracking images; Among them, the intelligent tracking module includes: A parameter adjustment unit, for adjusting lens parameters of the target lens based on the number of tracking targets corresponding to the target lens and the lens focus center, with reference to preset dynamic tracking image requirements; A first tracking unit is used to obtain a first dynamic tracking image corresponding to the priority tracking target, and send the first dynamic tracking image to a display position corresponding to the target number carried by the priority tracking target for display; A second tracking unit is used to obtain a second dynamic tracking image corresponding to a target lens corresponding to a secondary tracking target, and send the second tracking video to a display position corresponding to a target number carried by the secondary tracking target for display; The second tracking unit comprises: The recording subunit is used to record the motion trajectory of the secondary tracking target in the monitoring area, generate a marked monitoring switching node according to the target lens allocation information corresponding to the secondary tracking target, and mark the monitoring switching node mark on the motion trajectory; The acquisition subunit is used to determine the target lens currently corresponding to the secondary tracking target based on the monitoring switching node, acquire the real-time picture, and pre-process the real-time picture to obtain the current tracking sub-video with the secondary tracking target as the video center; The splicing subunit is used to monitor that when the switching node is triggered, obtain the first background feature and the first target feature corresponding to the initial frame image of the current tracking sub-video, and the second background feature and the second target feature corresponding to the end frame image of the previous tracking sub-video; Comparing the first background feature with the second background feature, determining the background association relationship between the current tracking sub-video and the previous tracking sub-video, determining the associated background, and generating an image change vector based on the change trend of the overlapping part of the associated background on the end frame image and the initial frame image; At the same time, based on the first target feature generation and the second target feature comparison, the action change corresponding to the secondary tracking target on the end frame image and the initial frame image is determined, and whether the connection action of the secondary tracking target is coherent; If they are coherent, then the associated background is spliced ​​based on the image change vector to generate a second dynamic tracking image; Otherwise, based on the motion change feature, similar motion query is performed in the acquired tracking sub-video of the secondary tracking target to obtain the best similar motion, obtain the motion vector corresponding to the best similar motion, and perform motion supplement on the connecting motion of the secondary tracking target according to the motion vector to obtain the coherent connecting motion; According to the graphic change vector, the associated background is fused and spliced, the video splicing node is determined, and the coherent connection action is inserted into the video splicing node to replace the connection action, so as to generate a second dynamic tracking image; The data association subunit is used to determine the priority tracking target corresponding to each movement stage of the secondary tracking target according to the marking result on the movement trajectory of the secondary tracking target as the associated tracking target; Based on the shot switching time of each motion stage, determine the video association time period of the associated tracking target and the secondary tracking target, and based on the video association time period, establish a mapping association relationship between each tracking sub-video corresponding to the second dynamic tracking image and the target time period corresponding to the first dynamic tracking image; The video jump subunit is used to obtain the corresponding first dynamic tracking image segment based on the mapping association relationship corresponding to the click time period corresponding to the click of the display end when the display end clicks to view the original video of the second dynamic tracking image.

2. A liquid lens-based dynamic tracking monitoring system according to claim 1, characterized in that: Task scheduling module, including: An information acquisition unit, a user acquires the current working status of multiple liquid lenses included in the target tracking device group, determines whether there is an idle liquid lens, and if so, adds an idle tag to the idle liquid lens; A first allocation unit is used to obtain the first position coordinates of the new tracking target in the monitoring area, and obtain the position information of each idle liquid lens, and allocate the new tracking target to the nearest idle liquid lens according to the principle of proximity based on the real-time position information and the position coordinates, and use the nearest idle liquid lens as the target lens in real time; The second allocation unit is used to obtain the second position coordinates of the monitoring and tracking targets corresponding to each liquid lens in the monitoring area when there are no idle liquid lenses in the target tracking device group, compare the first position coordinates with each second position coordinate respectively to obtain multiple coordinate distances, and use the liquid lens corresponding to the monitoring and tracking target corresponding to the second position coordinate corresponding to the minimum coordinate distance as the target lens, and allocate the new tracking target as the secondary tracking target to the target lens.

3. A liquid lens-based dynamic tracking monitoring system according to claim 2, characterized in that: The second allocation unit comprises: The tag adding subunit is used to determine the first tracking target of the target lens, add a priority tracking target tag to the first tracking target, and add a secondary tracking target tag to the remaining monitoring tracking targets of the target lens, wherein the remaining monitoring tracking targets include the new tracking target; The label updating subunit is used to update the target labels of all monitoring and tracking targets in the current tracking picture when the number of tracking targets of the target lens changes; When the priority tracking target leaves the monitoring area, the secondary tracking target closest to the lens center in the current tracking image is taken as the priority tracking target, and the target label is updated.

4. The dynamic tracking monitoring system based on liquid lens according to claim 1, characterized in that: Target locking module, including: A task confirmation unit, used to determine the total number of tracking targets of the target lens corresponding to the new tracking target; A locking control unit, used for, when the total number of tracking targets of the target lens is 1, taking the new tracking target as the lens focus center and locking the new tracking target based on the rectangular frame; When the total number of tracking targets of the target lens is not 1, the target center is determined based on the coordinate positions of each tracking target corresponding to the target lens, the target center is used as the lens focusing center, and each tracking target is locked based on the rectangular frame.

5. The dynamic tracking monitoring system based on liquid lens according to claim 3, characterized in that: Locking control unit, comprising: The position monitoring subunit is used to obtain the real-time distance between the first position coordinate of the secondary tracking target and the second position coordinate of the corresponding priority tracking target in real time when the total number of tracking targets of the target lens is not 1, and when the real-time distance is greater than the preset distance, control the second allocation unit to reallocate the secondary tracking target.

6. The dynamic tracking monitoring system based on liquid lens according to claim 1, characterized in that: Also includes: The remote control module is used to remotely adjust the tracking parameters of each liquid lens in the target tracking device group, and generate lens parameter adjustment instructions according to the remote adjustment results of the tracking parameters, synchronously adjust the liquid lens parameters, and obtain a dynamic tracking image of the target magnification.

7. A liquid lens-based dynamic tracking monitoring method is applied to a liquid lens-based dynamic tracking monitoring system according to any one of claims 1 to 6, characterized in that: include: Step 1: Based on the preset target recognition model, target detection is performed on the real-time panoramic video to determine whether there is a new tracking target in the monitoring area; Step 2: after confirming that a new tracking target appears in the monitoring area, the new tracking target is assigned to the target lens for tracking based on the actual progress of the current tasks of multiple liquid lenses in the target tracking device group; Step 3: Based on the position information of the new tracking target, control the target lens to lock the new tracking target; Step 4: Control the eyepiece lens to adjust lens parameters according to the movement of the tracking target to obtain dynamic tracking images.

Images