Remote alarm monitoring method and device, computer equipment and storage medium thereof

Abstract

The invention relates to a remote alarm monitoring method and device, computer equipment and a storage medium thereof, and the method comprises the steps: obtaining monitoring data outputted by a camera module, and the monitoring data comprises image data; identifying the image data, calibrating a three-dimensional coordinate system of a moving object presented in the image data, and forming a carried tracking coordinate system in the image data; the movement of the corresponding moving object is continuously tracked through the tracking coordinate system; based on a pre-deployed camera network and a grid map, using a convolutional neural network model to identify a moving orientation of the tracking coordinate system, and sensing a corresponding camera module in the camera network according to the moving orientation; judging whether each sensitive camera module collects a tracking coordinate system or not, generating a route path when the moving object moves based on the positioning of each camera module, and marking the route path on a grid map to form a tracking map; and uploading the tracking map to a remote alarm terminal in real time, so that a user can check the moving condition of the object in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of video data processing, and in particular to a remote alarm monitoring method, device, computer equipment and storage medium thereof. Background Art

[0002] In modern society, remote alarm monitoring systems play a vital role in helping to protect and monitor the safety of various places and facilities. For example, these systems can be applied to public transportation systems, commercial buildings, residential communities, industrial facilities, etc.

[0003] However, the current remote alarm monitoring system still has some drawbacks and technical difficulties in the industry context. First, most of the existing monitoring systems only provide basic image or video monitoring functions and lack the ability to accurately identify and track moving objects. Secondly, since traditional monitoring systems use fixed camera layouts, the coverage area is limited and cannot meet the needs of large-scale coverage. In addition, the amount of monitoring data is huge, and managing and analyzing this data is also a challenge. For certain key places, such as transportation hubs or public safety areas, the importance of real-time monitoring and accurate alarms is particularly prominent.

[0004] The existing remote alarm monitoring system has drawbacks and needs. The following are the technical difficulties: Accurate tracking and identification of moving objects: How to use the monitoring data output by the camera module to accurately identify and track moving objects so as to provide reliable monitoring and alarm functions.

[0005] Effective coverage of large areas: How to achieve effective coverage of large areas through effective camera module deployment and optimized layout to avoid dead ends and blind spots and ensure that the entire area is fully monitored.

[0006] Processing and management of monitoring data: How to process and manage huge amounts of monitoring data to achieve efficient storage, retrieval and analysis in order to provide real-time and accurate alarm information.

[0007] Real-time and reliability requirements: In key locations, such as transportation hubs or public safety areas, real-time monitoring and fast and accurate alarm response are essential. How to meet real-time and reliability requirements to ensure that the monitoring system can detect and respond to potential security threats in a timely manner. Summary of the invention

[0008] The main purpose of the present invention is to provide a remote alarm monitoring method, device, computer equipment and storage medium thereof, which can accurately identify and track the moving position of moving objects by utilizing a convolutional neural network model in combination with long short-term memory functions, combined with a pre-deployed camera network and grid map, and generate a tracking map in real time, and upload relevant information to a remote alarm terminal so that users can view the movement of objects in real time.

[0009] To achieve the above object, the present invention provides a remote alarm monitoring method, comprising the following steps: Acquire monitoring data output by the camera module, wherein the monitoring data includes image data; wherein the image data is used to display the moving object being photographed; Identify the image data, and calibrate the three-dimensional coordinate system of the moving object presented in the image data, so as to form a tracking coordinate system carried in the image data; so as to continuously track the movement of the corresponding moving object through the tracking coordinate system; Based on the pre-deployed camera network and grid map, a convolutional neural network model is used to identify the moving direction of the tracking coordinate system, and the corresponding camera module in the camera network is sensitively detected according to the moving direction; Determine whether each of the sensitive camera modules has captured the tracking coordinate system. If so, generate the route behavior of the moving object based on the positioning of each camera module, and mark the route behavior on the grid map to form a tracking map; The tracking map is uploaded to the remote alarm terminal in real time.

[0010] Furthermore, the step of obtaining monitoring data output by the camera module includes: Acquire monitoring data output by the camera module, wherein the monitoring data includes status data; the status data is used to present information on image resolution, power and signal of the camera module; Determine whether the image resolution, power and signal of the camera module are higher than the preset resolution threshold, power threshold and signal threshold; If yes, the camera module is marked as a sensitive camera module, and the sensitive camera module is loaded and constructed at a corresponding position of a preset grid map; If not, the camera module is marked as an abnormal camera module and the abnormal camera module is not called.

[0011] Furthermore, the step of identifying the image data includes: The image data is monitored by a preset voxel bounding box program and a corresponding color scheme; the method for monitoring the image data is: In the range of the 1st frame to the 24th frame of the image data, all objects on the image data are marked with different colors using a pixel bounding box program and a corresponding color scheme, specifically, corresponding bounding boxes are marked with different colors; Continuously monitor whether each bounding box carrying a color in the image data is displaced compared with the previous frame, and repeat this judgment process 20 times to screen out the first displaced bounding box that is moved; Continue to monitor whether the displacement of the first displacement bounding box is greater than a preset displacement threshold. Repeat this judgment process three times to screen out a second displacement bounding box whose displacement is greater than the threshold, and mark the second displacement bounding box as a moving object.

[0012] Furthermore, the step of calibrating a three-dimensional coordinate system for the moving object presented in the image data to form a tracking coordinate system carried in the image data includes: Loading the planarization data to the camera module, and determining the absolute coordinate origin on the image data through the planarization data; An absolute coordinate system is constructed by using the absolute coordinate origin, and coordinate points of the second displacement bounding box are calibrated based on the absolute coordinate points of the absolute coordinate system, thereby forming a tracking coordinate system corresponding to the second displacement bounding box; Among them, the method of determining the absolute coordinate origin is: identifying a path plane in the image data using an image, the path plane comprising a road plane; Overlapping the path plane with the virtual plane carried by the planarization data to calculate the angle difference between the orientation angle of the camera module and the virtual angle preset in the planarization data; The image carried in the image data is uniformly adjusted using the angle difference correspondence to overlap the image carried in the image data with the virtual image carried by the planarization data; The absolute coordinate origin is calibrated on the path plane after the uniform adjustment.

[0013] Furthermore, based on the pre-deployed camera network and grid map, a convolutional neural network model is used to identify the moving position of the tracking coordinate system, and the step of sensitively identifying the corresponding camera module in the camera network according to the moving position includes: Retrieving the grid map, and inputting the grid map into a convolutional neural network model, so as to convolve the grid map through a subconvolution kernel of the convolutional neural network model to generate a corresponding grid structure, wherein the grid structure includes grid connection lines and grid nodes; The camera module information corresponding to the pre-deployed camera network is input into the convolutional neural network model to calibrate and associate the camera module with the grid node through positioning; Using the long short-term memory of the convolutional neural network model, the changes of the tracking coordinate system within the range of 48 frames are identified, and based on the changes, the X-axis or Y-axis of the tracking coordinate system is adjusted to be consistent with the moving direction of the object; The next sensitive camera module is determined in the grid structure through the grid nodes, grid line connections and movement directions corresponding to the camera network currently stored in the tracking coordinate system.

[0014] Further, it is determined whether each sensitive camera module has captured the tracking coordinate system. If so, based on the positioning of each camera module, a route behavior of the moving object is generated, and the route behavior is marked on a grid map to form a tracking map, including: Determine whether the tracking coordinate system is captured by the sensitive camera module; If yes, the sensitive camera module is marked on the grid structure, and the entire grid structure continues to perform this judgment process; The marked grid nodes in the grid structure are reversed to mark the corresponding route paths on the grid map.

[0015] Furthermore, after the step of uploading the tracking map to the remote alarm terminal in real time, the following steps are included: Obtaining a viewing instruction of the remote alarm terminal; The object movement image data collected by the camera module corresponding to the node viewing information is displayed on the tracking map through the node viewing information carried by the viewing instruction.

[0016] The present invention also provides a remote alarm monitoring device, comprising: An acquisition unit, used to acquire monitoring data output by the camera module, wherein the monitoring data includes image data; wherein the image data is used to display the moving object being photographed; A coordinate unit, used to identify the image data, and calibrate the three-dimensional coordinate system of the moving object presented in the image data, so as to form a tracking coordinate system carried in the image data; so as to continuously track the movement of the corresponding moving object through the tracking coordinate system; A model unit, for identifying the moving orientation of the tracking coordinate system using a convolutional neural network model based on a pre-deployed camera network and a grid map, and sensitively deriving a corresponding camera module in the camera network according to the moving orientation; A behavior unit, used to determine whether each of the sensitive camera modules has captured the tracking coordinate system. If so, based on the positioning of each camera module, a route behavior of the moving object is generated, and the route behavior is marked on a grid map to form a tracking map; The feedback unit is used to upload the tracking map to the remote alarm terminal in real time.

[0017] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned remote alarm monitoring method when executing the computer program.

[0018] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned remote alarm monitoring method are implemented.

[0019] The remote alarm monitoring method, device, computer equipment and storage medium provided by the present invention have the following beneficial effects: Monitoring data acquisition and processing: This method can acquire the monitoring data output by the camera module and identify moving objects from it. Through the calibration of the three-dimensional coordinate system and the formation of the tracking coordinate system, the movement of the moving object can be continuously tracked.

[0020] Camera module selection and deployment: Based on the pre-deployed camera network and grid map, the convolutional neural network model is used to identify the moving direction of the tracking coordinate system, and the corresponding camera module is selected according to the moving direction. In this way, the sensitive camera module can be determined to accurately monitor the moving object.

[0021] Route generation and annotation: Based on whether the sensitive camera module captures the tracking coordinate system, the route of the moving object can be generated and annotated on the grid map. This forms a tracking map that can display the trajectory of the moving object.

[0022] Real-time upload and remote viewing: The tracking map can be uploaded to the remote alarm terminal in real time. Through the viewing command of the remote terminal, the object movement image data collected by the camera module corresponding to the command can be displayed on the tracking map. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flowchart of a remote alarm monitoring method in one embodiment of the present invention; Figure 2 It is a structural block diagram of remote alarm monitoring in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0024] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0026] Reference Figure 1 The present invention provides a flow chart of a remote alarm monitoring method, which includes the following steps: S1, obtaining monitoring data output by a camera module, wherein the monitoring data includes image data; wherein the image data is used to display the moving object being photographed; S2, identifying the image data, and calibrating a three-dimensional coordinate system for the moving object presented in the image data, so as to form a tracking coordinate system carried in the image data; so as to continuously track the movement of the corresponding moving object through the tracking coordinate system; S3, based on the pre-deployed camera network and grid map, using the convolutional neural network model to identify the moving direction of the tracking coordinate system, and sensitively identify the corresponding camera module in the camera network according to the moving direction; S4, determining whether each of the sensitive camera modules has captured the tracking coordinate system, and if so, generating a route of the moving object based on the positioning of each of the camera modules, and marking the route on a grid map to form a tracking map; S5, uploading the tracking map to the remote alarm terminal in real time.

[0027] In S1, the monitoring data output by the camera module is obtained, which includes image data, and the image data can display the moving object being photographed. Through this technology, the image data can be extracted from the monitoring data and the image of the moving object can be displayed, thereby realizing visual monitoring of the moving object.

[0028] In S2, the image data is identified and the three-dimensional coordinate system is calibrated to form a tracking coordinate system. Through this technology, the three-dimensional coordinates of the moving object can be calibrated in the image data to form a tracking coordinate system. Through the tracking coordinate system, the movement of the corresponding moving object can be continuously tracked.

[0029] In S3, based on the pre-deployed camera network and grid map, a convolutional neural network model is used to identify the moving direction of the tracking coordinate system, and the corresponding camera module in the camera network is selected according to the moving direction. Through this technology, the corresponding camera module can be selected for monitoring according to the moving direction of the moving object in the tracking coordinate system. This ensures that the sensitive camera module can accurately monitor the moving object and provide more accurate monitoring data.

[0030] In S4, it is determined whether each camera module that is sensitive has captured the tracking coordinate system. Based on the positioning information of each camera module, the route of the moving object is generated, and the route is marked on the grid map to form a tracking map. The tracking map can clearly display the trajectory and movement path of the moving object.

[0031] In S5, the tracking map is uploaded to the remote alarm terminal in real time. Through this technology, the tracking map can be transmitted to the remote alarm terminal in a timely manner, allowing monitoring personnel to view the data on the tracking map in real time.

[0032] In one embodiment, the step of obtaining monitoring data output by the camera module includes: Acquire monitoring data output by the camera module, wherein the monitoring data includes status data; the status data is used to present information on image resolution, power and signal of the camera module; Determine whether the image resolution, power and signal of the camera module are higher than the preset resolution threshold, power threshold and signal threshold; If yes, the camera module is marked as a sensitive camera module, and the sensitive camera module is loaded and constructed at a corresponding position of a preset grid map; If not, the camera module is marked as an abnormal camera module and the abnormal camera module is not called.

[0033] Specifically, the monitoring data output by the camera module is obtained, including status data. The status data is used to display the image resolution, power and signal information of the camera module. It is determined whether the image resolution, power and signal of the camera module are higher than the preset resolution threshold, power threshold and signal threshold. If the conditions are met, the camera module is calibrated as a sensitive camera module, and the sensitive camera module is loaded to the corresponding position of the preset grid map. If the conditions are not met, the camera module is calibrated as an abnormal camera module, and the abnormal camera module is not used. Through this technology, the availability of the camera module can be judged according to its status data, and the camera module that meets the conditions can be calibrated as a sensitive camera module, and then loaded onto the preset grid map. This helps to determine the sensitive camera module in order to accurately monitor moving objects. For abnormal camera modules, they will not be used to ensure the reliability and accuracy of the monitoring system.

[0034] In one embodiment, the step of identifying the image data comprises: The image data is monitored by a preset voxel bounding box program and a corresponding color scheme; the method for monitoring the image data is: In the range of the 1st frame to the 24th frame of the image data, all objects on the image data are marked with different colors using a pixel bounding box program and a corresponding color scheme, specifically, corresponding bounding boxes are marked with different colors; Continuously monitor whether each bounding box carrying a color in the image data is displaced compared with the previous frame, and repeat this judgment process 20 times to screen out the first displaced bounding box that is moved; Continue to monitor whether the displacement of the first displacement bounding box is greater than a preset displacement threshold. Repeat this judgment process three times to screen out a second displacement bounding box whose displacement is greater than the threshold, and mark the second displacement bounding box as a moving object.

[0035] Specifically, the marking and coloring of the pixel bounding box are completed within 24 frames. During this process, 20 frames are used to screen out the first displacement bounding box with movement, and 3 frames are used to determine the displacement amount. The first frame is ignored here.

[0036] In one embodiment, the step of calibrating a three-dimensional coordinate system for a moving object presented in the image data and forming a tracking coordinate system carried in the image data includes: Loading the planarization data to the camera module, and determining the absolute coordinate origin on the image data through the planarization data; An absolute coordinate system is constructed by using the absolute coordinate origin, and coordinate points of the second displacement bounding box are calibrated based on the absolute coordinate points of the absolute coordinate system, thereby forming a tracking coordinate system corresponding to the second displacement bounding box; Among them, the method of determining the absolute coordinate origin is: identifying a path plane in the image data using an image, the path plane comprising a road plane; Overlapping the path plane with the virtual plane carried by the planarization data to calculate the angle difference between the orientation angle of the camera module and the virtual angle preset in the planarization data; The image carried in the image data is uniformly adjusted using the angle difference correspondence to overlap the image carried in the image data with the virtual image carried by the planarization data; The absolute coordinate origin is calibrated on the path plane after the uniform adjustment.

[0037] Specifically, the planarization data is loaded into the camera module, and the absolute coordinate origin is determined on the image data through the planarization data. An absolute coordinate system is constructed based on the absolute coordinate origin, and the absolute coordinate points of the absolute coordinate system are used to calibrate the coordinate points of the second displacement bounding box, thereby forming a tracking coordinate system corresponding to the second displacement bounding box. The method for determining the absolute coordinate origin is as follows: the path plane in the image data is identified using image recognition technology, including the road plane. The path plane is overlapped with the virtual plane carried by the planarization data, and the angle difference between the orientation angle of the camera module and the virtual angle preset by the planarization data is calculated. The image in the image data is uniformly adjusted using the angle difference so that it coincides with the virtual image carried by the planarization data. The absolute coordinate origin is calibrated on the uniformly adjusted path plane. Through this step, the path plane and the planarization data can be used to determine the absolute coordinate origin, and a tracking coordinate system can be constructed based on the origin, thereby calibrating the three-dimensional coordinates of the moving object in the image data. This helps in the subsequent tracking and accurate positioning of moving objects.

[0038] In one embodiment, based on a pre-deployed camera network and a grid map, a convolutional neural network model is used to identify the moving position of the tracking coordinate system, and the steps of sensitively identifying the corresponding camera module in the camera network according to the moving position include: Retrieving the grid map, and inputting the grid map into a convolutional neural network model, so as to convolve the grid map through a subconvolution kernel of the convolutional neural network model to generate a corresponding grid structure, wherein the grid structure includes grid connection lines and grid nodes; The camera module information corresponding to the pre-deployed camera network is input into the convolutional neural network model to calibrate and associate the camera module with the grid node through positioning; Using the long short-term memory of the convolutional neural network model, the changes of the tracking coordinate system within the range of 48 frames are identified, and based on the changes, the X-axis or Y-axis of the tracking coordinate system is adjusted to be consistent with the moving direction of the object; The next sensitive camera module is determined in the grid structure through the grid nodes, grid line connections and movement directions corresponding to the camera network currently stored in the tracking coordinate system.

[0039] In the specific implementation process, the preset grid map is retrieved and input into the convolutional neural network model. The grid map is convolved by the sub-convolution kernel of the convolutional neural network model to generate a corresponding grid structure, including grid connection lines and grid nodes. The information of the camera module in the pre-deployed camera network is input into the convolutional neural network model, and its location is calibrated and associated with the corresponding grid node. Using the long and short-term memory function of the convolutional neural network model, the change of the tracking coordinate system is identified within the range of 48 frames, and the X-axis or Y-axis of the tracking coordinate system is adjusted to be consistent with the moving direction of the object according to the change. Based on the grid nodes, grid connection lines and moving direction of the object corresponding to the camera network where the tracking coordinate system is currently located, the next sensitive camera module is determined in the grid structure. Through this step, the convolutional neural network model can be used to identify the moving direction of the tracking coordinate system and sensitively determine the next camera module to be monitored. This helps to ensure that the tracking system and the camera network keep track of and monitor moving objects in sync.

[0040] The long short-term memory process is as follows: input a 48-frame image data sequence into the convolutional neural network model. In the convolutional neural network, LSTM units are used to process the image data sequence. The LSTM units process the image data frame by frame and memorize the previous hidden state and unit state. During the processing of each frame, the LSTM unit updates the hidden state and unit state according to the input of the current frame and the hidden state of the previous frame, and predicts the changes in the current tracking coordinate system. According to the output of the LSTM unit, it is determined whether the X-axis or Y-axis of the current tracking coordinate system needs to be adjusted to maintain its consistency with the moving direction of the object. If it is determined that adjustment is required, the adjustment amplitude and direction are determined according to the output of the LSTM unit, and the X-axis or Y-axis of the tracking coordinate system is adjusted accordingly. According to the updated tracking coordinate system, the image data of the next frame continues to be processed, and steps 2 to 5 are repeated until all 48 frames of image data are processed.

[0041] In one embodiment, the steps of determining whether each of the sensitive camera modules has captured a tracking coordinate system, and if so, generating a route behavior of the moving object based on the positioning of each of the camera modules, and marking the route behavior on a grid map to form a tracking map include: Determine whether the tracking coordinate system is captured by the sensitive camera module; If yes, the sensitive camera module is marked on the grid structure, and the entire grid structure continues to perform this judgment process; The marked grid nodes in the grid structure are reversed to mark the corresponding route paths on the grid map.

[0042] Specifically, determine whether the sensitive camera module has collected the tracking coordinate system based on the change of the tracking coordinate system. If a camera module collects the tracking coordinate system, mark the corresponding position of the camera module on the grid structure, and continue to execute this judgment process to detect the collection status of other camera modules. Reverse the marked grid nodes, that is, reversely search from the camera module position, and mark the route of the moving object on the grid map along the grid connection line. According to the route marked on the tracking map, the movement and path of the moving object can be more intuitively understood. Through the above steps, the route of the moving object can be generated according to the tracking coordinate system and positioning information collected by the sensitive camera module, and marked on the grid map to form a tracking map. This helps to better understand the trajectory and action route of the moving object.

[0043] In one embodiment, after the step of uploading the tracking map to the remote alarm terminal in real time, the following steps are included: Obtaining a viewing instruction of the remote alarm terminal; The object movement image data collected by the camera module corresponding to the node viewing information is displayed on the tracking map through the node viewing information carried by the viewing instruction.

[0044] Specifically, obtain the viewing instruction of the remote alarm terminal, which may be a command transmitted through network communication or other means. Parse the viewing instruction and extract the node viewing information carried in the instruction. According to the node viewing information, locate the corresponding node on the tracking map, that is, the position of the camera module. Obtain the image data of the object movement from the corresponding camera module, which may have been collected and stored in the camera module. Display the object movement image data collected by the camera module on the tracking map and mark it based on the corresponding node. Display the object movement image data on the tracking map in a timely manner so that the user of the remote alarm terminal can view the object movement of the specific node in real time. Through the above steps, according to the viewing instruction of the remote alarm terminal, the object movement image data collected by the camera module corresponding to the node viewing information can be displayed on the tracking map, thereby realizing real-time monitoring and viewing of the specific node. This can help users better understand the real-time movement of objects. refer to Figure 2 , is a structural block diagram of a remote alarm monitoring device proposed by the present invention, the device comprises: The acquisition unit 10 is used to acquire the monitoring data output by the camera module, wherein the monitoring data includes image data; wherein the image data is used to display the moving object being photographed; A coordinate unit 20, for identifying the image data, and calibrating a three-dimensional coordinate system for the moving object presented in the image data, and forming a tracking coordinate system carried in the image data; so as to continuously track the movement of the corresponding moving object through the tracking coordinate system; The model unit 30 is used to identify the moving direction of the tracking coordinate system using a convolutional neural network model based on the pre-deployed camera network and grid map, and to sense the corresponding camera module in the camera network according to the moving direction; The path unit 40 is used to determine whether each of the sensitive camera modules has captured the tracking coordinate system. If so, based on the positioning of each camera module, it generates the path of the moving object when it moves, and marks the path on the grid map to form a tracking map; The feedback unit 50 is used to upload the tracking map to the remote alarm terminal in real time.

[0045] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to the above method embodiment, which will not be repeated here.

[0046] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0047] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0048] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0049] In summary, this is the embodiment of the present invention.

[0050] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0051] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0052] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A remote alarm monitoring method, characterized in that: The following steps are involved: Acquire monitoring data output by the camera module, wherein the monitoring data includes image data; wherein the image data is used to display the moving object being photographed; Identify the image data, and calibrate the three-dimensional coordinate system of the moving object presented in the image data, so as to form a tracking coordinate system carried in the image data; so as to continuously track the movement of the corresponding moving object through the tracking coordinate system; Based on the pre-deployed camera network and grid map, a convolutional neural network model is used to identify the moving direction of the tracking coordinate system, and the corresponding camera module in the camera network is sensitively detected according to the moving direction; Determine whether each of the sensitive camera modules has captured the tracking coordinate system. If so, generate the route of the moving object based on the positioning of each camera module, and mark the route on the grid map to form a tracking map; The tracking map is uploaded to the remote alarm terminal in real time.

2. The remote alarm monitoring method according to claim 1, characterized in that: The steps of obtaining monitoring data output by the camera module include: Acquire monitoring data output by the camera module, wherein the monitoring data includes status data; the status data is used to present information on image resolution, power and signal of the camera module; Determine whether the image resolution, power and signal of the camera module are higher than the preset resolution threshold, power threshold and signal threshold; If yes, the camera module is marked as a sensitive camera module, and the sensitive camera module is loaded and constructed at a corresponding position of a preset grid map; If not, the camera module is marked as an abnormal camera module and the abnormal camera module is not called.

3. The remote alarm monitoring method according to claim 1, characterized in that: The step of identifying the image data comprises: The image data is monitored by a preset voxel bounding box program and a corresponding color scheme; the method for monitoring the image data is: In the range of the 1st frame to the 24th frame of the image data, all objects on the image data are marked with different colors using a pixel bounding box program and a corresponding color scheme, specifically, corresponding bounding boxes are marked with different colors; Continuously monitor whether each bounding box carrying a color in the image data is displaced compared with the previous frame, and repeat this judgment process 20 times to screen out the first displaced bounding box that is moved; Continue to monitor whether the displacement of the first displacement bounding box is greater than a preset displacement threshold. Repeat this judgment process three times to screen out a second displacement bounding box whose displacement is greater than the threshold, and mark the second displacement bounding box as a moving object.

4. The remote alarm monitoring method according to claim 3, characterized in that: The step of calibrating a three-dimensional coordinate system for a moving object presented in the image data to form a tracking coordinate system carried in the image data comprises: Loading the planarization data to the camera module, and determining the absolute coordinate origin on the image data through the planarization data; An absolute coordinate system is constructed by using the absolute coordinate origin, and coordinate points of the second displacement bounding box are calibrated based on the absolute coordinate points of the absolute coordinate system, thereby forming a tracking coordinate system corresponding to the second displacement bounding box; Among them, the method of determining the absolute coordinate origin is: identifying a path plane in the image data using an image, the path plane comprising a road plane; Overlapping the path plane with the virtual plane carried by the planarization data to calculate the angle difference between the orientation angle of the camera module and the virtual angle preset in the planarization data; The image carried in the image data is uniformly adjusted using the angle difference correspondence to overlap the image carried in the image data with the virtual image carried by the planarization data; The absolute coordinate origin is calibrated on the path plane after the uniform adjustment.

5. The remote alarm monitoring method according to claim 1, characterized in that: Based on the pre-deployed camera network and grid map, a convolutional neural network model is used to identify the moving direction of the tracking coordinate system, and the steps of sensitively deriving the corresponding camera module in the camera network according to the moving direction include: Retrieving the grid map, and inputting the grid map into a convolutional neural network model, so as to convolve the grid map through a subconvolution kernel of the convolutional neural network model to generate a corresponding grid structure, wherein the grid structure includes grid connection lines and grid nodes; The camera module information corresponding to the pre-deployed camera network is input into the convolutional neural network model to calibrate and associate the camera module with the grid node through positioning; Using the long short-term memory of the convolutional neural network model, the changes of the tracking coordinate system within the range of 48 frames are identified, and based on the changes, the X-axis or Y-axis of the tracking coordinate system is adjusted to be consistent with the moving direction of the object; The next sensitive camera module is determined in the grid structure through the grid nodes, grid line connections and movement directions corresponding to the camera network currently stored in the tracking coordinate system.

6. The remote alarm monitoring method according to claim 5, characterized in that: The steps of determining whether each of the sensitive camera modules has captured the tracking coordinate system, and if so, generating the route behavior of the moving object when moving based on the positioning of each of the camera modules, and marking the route behavior on the grid map to form a tracking map include: Determine whether the tracking coordinate system is captured by the sensitive camera module; If yes, the sensitive camera module is marked on the grid structure, and the entire grid structure continues to perform this judgment process; The marked grid nodes in the grid structure are reversed to mark the corresponding route paths on the grid map.

7. The remote alarm monitoring method according to claim 1, characterized in that: After the step of uploading the tracking map to the remote alarm terminal in real time, the method includes: Obtaining a viewing instruction of the remote alarm terminal; The object movement image data collected by the camera module corresponding to the node viewing information is displayed on the tracking map through the node viewing information carried by the viewing instruction.

8. A remote alarm monitoring device, characterized in that: include: An acquisition unit, used to acquire monitoring data output by the camera module, wherein the monitoring data includes image data; wherein the image data is used to display the moving object being photographed; A coordinate unit, used to identify the image data, and calibrate the three-dimensional coordinate system of the moving object presented in the image data, so as to form a tracking coordinate system carried in the image data; so as to continuously track the movement of the corresponding moving object through the tracking coordinate system; A model unit, for identifying the moving orientation of the tracking coordinate system using a convolutional neural network model based on a pre-deployed camera network and a grid map, and sensitively deriving a corresponding camera module in the camera network according to the moving orientation; A behavior unit, used to determine whether each of the sensitive camera modules has captured the tracking coordinate system. If so, based on the positioning of each camera module, a route behavior of the moving object is generated, and the route behavior is marked on a grid map to form a tracking map; The feedback unit is used to upload the tracking map to the remote alarm terminal in real time.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the remote alarm monitoring method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the remote alarm monitoring method according to any one of claims 1 to 7 are implemented.

Images