A method and system for monitoring and maintaining the growth status of ancient trees based on digital twins

Abstract

The present invention relates to a method and system for monitoring and maintaining the growth status of ancient trees based on digital twins, and belongs to the field of digital twin technology. The method includes constructing a virtual map of the ancient tree management area, obtaining the location data of the ancient trees, and marking the virtual map according to the location data of the ancient trees; obtaining the point cloud data of the ancient trees, constructing the twin model of the ancient trees according to the point cloud data of the ancient trees, and integrating the twin model of the ancient trees with the marked points of the ancient trees; obtaining daily data, calculating and obtaining statistical data according to the daily data, calculating and obtaining the growth data of the ancient trees according to the statistical data, and performing parameter correction on the twin model of the ancient trees according to the growth data of the ancient trees, obtaining the image data around the ancient trees and the real-time soil moisture data; analyzing the image data around the ancient trees and the real-time soil moisture data to obtain the alarm signal and the water supply valve switch signal. The automatic and intelligent monitoring and maintenance of the growth status of the ancient trees is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital twins, and specifically relates to a method and system for monitoring and maintaining the growth status of ancient trees based on digital twins. Background Art

[0002] With the acceleration of urbanization, the protection of ancient trees, as an important part of urban cultural heritage, faces many challenges. The complexity of the growth environment of ancient trees and the unpredictability of their growth state require us to adopt more efficient and scientific methods for monitoring and maintenance.

[0003] At present, there are the following problems in the maintenance of ancient trees:

[0004] (1) The protection sites for ancient and famous trees are relatively scattered and there are many types of trees. Their growth conditions and maintenance are different, so it is difficult to achieve unified maintenance for all ancient trees.

[0005] (2) The growth data of ancient trees are collected manually on a periodic basis, and manual evaluation of the collected data is subjective.

[0006] (3) Traditional ancient tree protection mainly relies on manpower and external material protection equipment to protect ancient trees, which has defects such as inadequate protection, affecting the city appearance, and wasting manpower. Summary of the invention

[0007] In order to solve the above problems existing in the prior art, the present invention provides a method and system for monitoring and maintaining the growth status of ancient trees based on digital twins.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] Constructing a virtual map of the ancient tree management area through GIS technology, obtaining the location data of the ancient trees, and marking the virtual map according to the location data of the ancient trees to obtain the ancient tree marking points;

[0010] Acquire ancient tree point cloud data through a visual sensor, construct a model based on the ancient tree point cloud data by reconstructing a model to obtain an ancient tree twin model, and integrate the ancient tree twin model with the ancient tree landmarks based on the ancient tree position data;

[0011] Obtain daily data through the ecological one-shot communication system, obtain statistical data through statistical calculation based on the daily data, obtain ancient tree growth data through growth model calculation based on the statistical data, perform parameter correction on the ancient tree twin model based on the ancient tree growth data, obtain image data around the ancient tree through the visual sensor, and obtain real-time soil moisture data through the ecological one-shot communication system;

[0012] The abnormal result is calculated by the abnormal detection model through the image data around the ancient tree, and the real-time soil moisture data is compared with the preset moisture content threshold to obtain a comparison result. The smart terminal generates an alarm adaptive signal according to the abnormal result to trigger an alarm, and the smart terminal generates an adaptive signal according to the comparison result to open the water supply valve.

[0013] As a preferred technical solution of the present invention, the reconstruction model includes a point cloud segmentation model, a point cloud repair model, and a three-dimensional reconstruction model. The specific calculation steps of the point cloud segmentation model are as follows:

[0014] S201: obtaining preprocessed point cloud data by preprocessing the ancient tree point cloud data;

[0015] S202: performing point cloud segmentation according to the preprocessed point cloud data by using the LeWoS model to obtain a first segmented point cloud, and performing a similarity threshold test according to the first segmented point cloud to obtain a first classification label;

[0016] S203: performing point cloud segmentation based on the pre-processed point cloud data by using the HSV color segmentation method to obtain a second segmented point cloud, and performing an H threshold test on the second segmented point cloud to obtain a second classification label;

[0017] S204: merging the first classification label and the second classification label according to the point cloud coordinates to obtain a merged label, and removing conflicting points from the ancient tree point cloud data according to the merged label to obtain a leaf segmentation point cloud and a branch segmentation point cloud.

[0018] Specifically, the specific calculation steps of the point cloud repair model are as follows:

[0019] S301: performing triangulation meshing according to the branch segmentation point cloud to obtain a triangulated mesh point cloud;

[0020] S302: performing hole identification on the triangular mesh point cloud according to the topological relationship of the center lines of the triangular facets to obtain the point cloud missing type;

[0021] S303: Determine a point cloud repair method according to the point cloud missing type, and repair the branch segmentation point cloud according to the point cloud repair method to obtain a repaired branch point cloud.

[0022] Preferably, the specific calculation steps of the three-dimensional reconstruction model are:

[0023] S401: Acquire color data and texture data of the repaired branch point cloud and the leaf segmentation point cloud;

[0024] S402: Obtaining a whole tree slice point cloud by slicing according to the repaired branch point cloud and the leaf segmentation point cloud;

[0025] S403: fitting and reconstructing the whole tree slice point cloud by using the RANSAC method to obtain a branch model;

[0026] S404: splicing each of the branch models to obtain an initial model, and filling the initial model according to the color data and the texture data to obtain the ancient tree twin model.

[0027] Preferably, the growth model is used to perform predictive calculations based on the statistical data to obtain leaf area prediction results, tree height prediction results, and tree breast diameter prediction results.

[0028] Preferably, the specific calculation steps of the anomaly detection model are:

[0029] S501: Obtaining a video frame sequence by frame processing according to the image data around the ancient tree;

[0030] S502: Obtain a preprocessed frame sequence by image preprocessing according to the video frame sequence;

[0031] S503: Calculate the abnormal result by using Timesformer algorithm according to the preprocessed frame sequence.

[0032] Preferably, the ancient tree twin model is integrated with the ancient tree landmark point according to the ancient tree location data, and the specific steps of the integration are: determine the ancient tree twin model corresponding to the ancient tree landmark point according to the ancient tree location data, obtain the storage location of the ancient tree twin model, and realize nesting through a jump function according to the landmark point and the storage location.

[0033] An ancient tree growth status monitoring and maintenance system based on digital twins, comprising: a map building module, a twin model building module, a data acquisition module, and an anomaly detection module;

[0034] The map building module is used to construct a virtual map of the ancient tree management area through GIS technology, obtain the ancient tree location data, and mark the virtual map according to the ancient tree location data to obtain the ancient tree marking points;

[0035] The twin model building module is used to obtain the ancient tree point cloud data through the visual sensor, construct the ancient tree twin model by reconstructing the model according to the ancient tree point cloud data, and integrate the ancient tree twin model with the ancient tree landmark according to the ancient tree position data;

[0036] The data acquisition module is used to acquire daily data through the ecological one-shot communication system, obtain statistical data through statistical calculation based on the daily data, obtain ancient tree growth data through growth model calculation based on the statistical data, perform parameter correction on the ancient tree twin model based on the ancient tree growth data, obtain image data around the ancient tree through the visual sensor, and obtain real-time soil moisture data through the ecological one-shot communication system;

[0037] The anomaly detection module is used to calculate the anomaly result through the anomaly detection model using the image data around the ancient tree, compare the real-time soil moisture data with the preset moisture content threshold to obtain a comparison result, and the smart terminal generates an alarm adaptive signal according to the anomaly result to trigger an alarm, and the smart terminal generates an adaptive signal according to the comparison result to open the water supply valve.

[0038] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the ancient tree growth status monitoring and maintenance method based on digital twins as described above is implemented.

[0039] A storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute the ancient tree growth status monitoring and maintenance method based on digital twins as described above.

[0040] The beneficial effects of the present invention are:

[0041] (1) By setting up an ecological one-pole system to obtain daily data in real time and analyzing the collected data, the manual collection of data on the growth environment of ancient trees is avoided.

[0042] (2) A virtual map of the ancient tree management area is established. According to the location of the ancient trees, points are set on the virtual map. A twin model of the ancient trees is established. The twin model of the ancient trees is integrated with the points. The twin model is changed according to the analysis results of the collected data. This realizes the visualization of ancient tree management and the targeted maintenance of the ancient trees based on the prediction of their current status through the collected data, thus avoiding real-time reliance on manual inspections.

[0043] (3) By setting up monitoring equipment near ancient trees to collect video data, abnormal behavior detection is performed on the video data and the corresponding markers are triggered to flash, thus realizing intelligent ancient tree monitoring, effectively avoiding the destruction of ancient trees and improving the city appearance.

[0044] (4) By obtaining the daily soil moisture content in real time and automatically switching the irrigation system on and off according to the daily soil moisture content, the water required by the ancient trees is supplied. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0046] Figure 1 It is a schematic flow chart of the ancient tree growth status monitoring and maintenance method based on digital twins of the present invention. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0048] The inventors of this case found that the current maintenance of ancient trees mainly relies on periodic manual data collection and manual growth index recording. However, the protection sites of ancient trees and famous trees are relatively scattered, and manual inspections one by one require a lot of time and energy, resulting in problems such as untimely and incomplete protection of ancient trees and famous trees and waste of resources. This embodiment provides the following solution:

[0049] See also Figure 1 , a method for monitoring and maintaining the growth status of ancient trees based on digital twins;

[0050] Constructing a virtual map of the ancient tree management area through GIS technology, obtaining the location data of the ancient trees, and marking the virtual map according to the location data of the ancient trees to obtain the ancient tree marking points;

[0051] It should be noted that a virtual map of the ancient tree management area is constructed through GIS technology, and the ancient trees are marked on the virtual map according to their locations. This is used to uniformly manage the ancient trees within the jurisdiction and to connect the digital twin models of the ancient trees at the marked points.

[0052] Acquire ancient tree point cloud data through a visual sensor, construct a model based on the ancient tree point cloud data by reconstructing a model to obtain an ancient tree twin model, and integrate the ancient tree twin model with the ancient tree landmarks based on the ancient tree position data;

[0053] It should be noted that this step is used to establish the initial digital twin model of ancient trees. Currently, the establishment of digital twin models of ancient trees mainly relies on collecting point cloud data of ancient trees, and performing processing steps such as cleaning, segmentation, filling, and construction on the point cloud data to construct the digital twin model of ancient trees.

[0054] Obtain daily data through the ecological one-shot communication system, obtain statistical data through statistical calculation based on the daily data, obtain ancient tree growth data through growth model calculation based on the statistical data, perform parameter correction on the ancient tree twin model based on the ancient tree growth data, obtain image data around the ancient tree through the visual sensor, and obtain real-time soil moisture data through the ecological one-shot communication system;

[0055] It should be noted that this step is used to obtain daily data in real time through the ecological one-pole system, summarize the daily data to obtain statistical data, and predict the tree height prediction results, tree breast diameter prediction results, and leaf area prediction results of the ancient trees through the growth model based on the monthly statistical data. The digital twin model of the ancient trees is adjusted according to the tree height prediction results, tree breast diameter prediction results, and leaf area prediction results to realize the visualization of the growth of the ancient trees. The digital twin model of the ancient trees can be viewed by clicking the target points on the virtual map to understand the growth status of the ancient trees in real time.

[0056] The abnormal result is calculated by the abnormal detection model through the image data around the ancient tree, and the real-time soil moisture data is compared with the preset moisture content threshold to obtain a comparison result. The smart terminal generates an alarm adaptive signal according to the abnormal result to trigger an alarm, and the smart terminal generates an adaptive signal according to the comparison result to open the water supply valve.

[0057] It should be noted that this step is used to monitor the growth environment of ancient trees in real time and to identify whether there are any behaviors of harming the ancient trees among the people moving around the ancient trees. If any behaviors of harming the ancient trees are identified, the points marked on the virtual map will be triggered to flash according to the actual coordinates of the ancient trees, which makes it easier for the staff to monitor the ancient trees and to calculate based on the real-time moisture content data to realize automated irrigation of the ancient trees.

[0058] In this embodiment, the alarm is triggered by the location of the ancient tree, which triggers the corresponding mark point on the virtual map to flash.

[0059] The daily data include daily ecological data, daily light data, and daily precipitation data. The daily ecological data include soil pH value, carbon dioxide concentration, trace element content, and management data; the management data include pest and disease level data, ground damage level data, dead branch level data, initial measured tree breast diameter, initial measured tree height, and initial measured leaf area; the statistical data include the average soil pH value of the previous month, the average trace element content value of the previous month, the management data of the previous month, the temperature of the previous day, the average precipitation of the previous month, the monthly average temperature, the highest temperature of the previous month, and the highest precipitation of the previous month.

[0060] In this embodiment, the ecological one-pole communication system is composed of a solar panel, a solar bracket, a waterproof box, a display screen, and a sensor, which realizes autonomous data collection and recording.

[0061] This embodiment also provides a method for monitoring and maintaining the state of ancient ginkgo trees based on digital twins, including the following steps:

[0062] Step 1: The staff collects the vector data of the management area and the longitude and latitude coordinates of the ancient ginkgo tree, cleans the vector data, converts the coordinates, and extracts the data to obtain unified data, imports the unified data into the GIS project, creates a map template for the GIS project through GIS software, makes the ancient ginkgo tree symbol, and marks the map template according to the longitude and latitude coordinates of the ancient ginkgo tree and the ancient ginkgo tree symbol;

[0063] Step 2: Use a binocular camera to take images of the ancient ginkgo tree from different angles, use computer vision technology to perform feature matching and depth estimation to obtain point cloud data of the ancient ginkgo tree, and construct a twin model of the ancient ginkgo tree based on the point cloud data of the ancient ginkgo tree through a reconstruction model. Connect the twin model of the ancient ginkgo tree with the ancient ginkgo tree symbol at the corresponding position on the virtual map according to the longitude and latitude coordinates of the ancient ginkgo tree, and store the twin model of the ancient ginkgo tree in a cloud database;

[0064] It should be noted that the administrator clicks on the ancient ginkgo tree symbol on the virtual map on the front-end page, picks up the mouse and clicks to obtain the longitude and latitude coordinates of the ancient ginkgo tree to be viewed, retrieves the corresponding twin model of the ancient ginkgo tree from the cloud database based on the longitude and latitude coordinates of the ginkgo tree, and views the twin model of the ancient ginkgo tree through the components provided by EnOS, thereby realizing the digitization of the growth status of the ancient ginkgo tree. By viewing the twin model of the ancient ginkgo tree, the growth status of the ancient ginkgo tree can be understood in real time.

[0065] Step 3: Use the Eco-One-Pole to obtain real-time daily data on the growth environment of the ancient ginkgo tree, compile statistics on the daily data to obtain monthly statistical data, and use the growth model to predict the leaf area, tree diameter at breast height, and tree height of the ancient ginkgo tree based on the monthly statistical data to obtain leaf area prediction results, tree diameter at breast height prediction results, and tree height prediction results. Adjust the leaf area, tree diameter at breast height, and tree height of the twin model of the ancient ginkgo tree based on the leaf area prediction results, tree diameter at breast height prediction results, and tree height prediction results;

[0066] Step 4: Obtain image data around the ancient ginkgo tree through a visual sensor, obtain real-time soil moisture data of the ancient ginkgo tree through the ecological one-pole communication system, transmit the image data around the ancient ginkgo tree and the real-time soil moisture data to the smart terminal through wireless signals, the smart terminal obtains abnormal results based on the image data around the ancient ginkgo tree through an abnormality detection model, compares the real-time soil moisture data with a preset moisture content threshold to obtain a comparison result, the smart terminal generates an alarm adaptive signal based on the abnormal result to trigger an alarm, and the smart terminal generates an adaptive signal based on the comparison result to open the water supply valve.

[0067] It should be noted that the main implementation method of the intelligent terminal generating an alarm adaptive signal according to the abnormal result to trigger the alarm is as follows: the computer program judges the abnormal result, if the abnormal result value is 1, there is an abnormality, the longitude and latitude coordinates of the ancient ginkgo tree are retrieved, and the corresponding ancient ginkgo tree symbol on the virtual map is triggered to flash according to the longitude and latitude coordinates of the ancient ginkgo tree, if the abnormal result value is 0, the ancient ginkgo tree symbol is not triggered to flash;

[0068] The main implementation method of the smart terminal generating an adaptive signal to open the water supply valve according to the comparison result is: the ecological one-pole transmits the monitored moisture content data to the smart terminal wirelessly, the control unit of the smart terminal processes the data according to a preset program, and sends a control signal, and the actuator on the irrigation system turns on or off the water pump according to the control signal to realize automatic irrigation.

[0069] Specifically, the reconstruction model includes a point cloud segmentation model, a point cloud repair model, and a 3D reconstruction model. The specific calculation steps of the point cloud segmentation model are as follows:

[0070] S201: obtaining preprocessed point cloud data by preprocessing the ancient tree point cloud data;

[0071] The preprocessing is to convert and splice the ancient tree point cloud data through Scene software to obtain spliced ​​data, and obtain the preprocessed point cloud data through filtering processing based on the spliced ​​data;

[0072] The filtering process includes CSF filtering and SOR filtering. First, initial filtering data is calculated based on the spliced ​​data through the CSF filtering algorithm, and the pre-processed point cloud data is calculated based on the initial filtering data through the SOR filter.

[0073] S202: performing point cloud segmentation according to the preprocessed point cloud data by using the LeWoS model to obtain a first segmented point cloud, and performing a similarity threshold test according to the first segmented point cloud to obtain a first classification label;

[0074] S203: performing point cloud segmentation based on the pre-processed point cloud data by using the HSV color segmentation method to obtain a second segmented point cloud, and performing an H threshold test on the second segmented point cloud to obtain a second classification label;

[0075] The H threshold test is to preset the H threshold when dividing the point cloud data to determine whether the points in the point cloud data belong to the leaf point cloud and the trunk point cloud.

[0076] S204: merging the first classification label and the second classification label according to the point cloud coordinates to obtain a merged label, and removing conflicting points from the ancient tree point cloud data according to the merged label to obtain a leaf segmentation point cloud and a branch segmentation point cloud.

[0077] Specifically, the specific calculation steps of the point cloud repair model are as follows:

[0078] S301: performing triangulation meshing according to the branch segmentation point cloud to obtain a triangulated mesh point cloud;

[0079] S302: identifying holes in the triangular mesh point cloud according to the topological relationship of the center lines of the triangular facets to obtain point cloud missing types; the point cloud missing types include symmetrical missing, broken missing, and porous missing;

[0080] S303: Determine a point cloud repair method according to the point cloud missing type, and repair the branch segmentation point cloud according to the point cloud repair method to obtain a repaired branch point cloud.

[0081] The specific steps of determining the point cloud repair method according to the point cloud missing type are: assigning the point cloud missing type to obtain a numerical label, and determining the point cloud repair method according to the numerical label. If the numerical label is 1, the cloud repair method is symmetrical repair; if the numerical label is 2, the cloud repair method is fracture repair; if the numerical label is 3, the cloud repair method is repair based on deep learning.

[0082] In this embodiment, the point cloud repair method includes symmetric repair, fracture repair, and deep learning-based repair. The symmetric repair is used to repair the symmetric missing point cloud data. First, the point cloud data is sliced ​​to obtain a slice point cloud. Then, edge features are extracted from the slice point cloud to obtain edge center points and vectors. The central axis of the point cloud data is generated according to the edge center points and the vectors. The symmetric points of the intersections of the horizontal planes of all points in the point cloud data and the central axis are calculated. The symmetric points are copied to obtain the repaired branch point cloud.

[0083] The fracture repair is used to repair the fracture-type missing point cloud data, slice the point cloud data to obtain a slice point cloud, extract feature parameters of the front and back slice point clouds of the slice point cloud to obtain feature parameters, and repair the point cloud data according to the feature parameters to obtain the repaired branch point cloud;

[0084] The deep learning-based repair is used to repair the missing point cloud data of the porous type, and feature learning of the point cloud data is performed through a convolutional neural network to obtain point cloud features, and a point cloud is directly generated according to the point cloud features to obtain the repaired branch point cloud.

[0085] Specifically, the specific calculation steps of the three-dimensional reconstruction model are:

[0086] S401: Acquire color data and texture data of the repaired branch point cloud and the leaf segmentation point cloud;

[0087] S402: Obtaining a whole tree slice point cloud by slicing according to the repaired branch point cloud and the leaf segmentation point cloud;

[0088] S403: fitting and reconstructing the whole tree slice point cloud by using the RANSAC method to obtain a branch model;

[0089] S404: splicing each of the branch models to obtain an initial model, and filling the initial model according to the color data and the texture data to obtain the ancient tree twin model.

[0090] Specifically, the growth model is used to perform prediction calculations based on the statistical data to obtain leaf area prediction results, tree height prediction results, and tree breast diameter prediction results;

[0091] The leaf area prediction result calculation formula is:

[0092] ,

[0093] in, Y t+1 is the leaf area prediction result, Y jt is the leaf area of ​​the previous month, Q is the climate factor function, θ is a random perturbation function, x, C 0 、C 1 、C 2 、C 3 、C 4 、C 5 is the generalized least squares estimate parameter, b 1 、b 2 、b 3 、b 4 is the weight coefficient, A 1 For tree age, A 2 The average soil pH value of last month. A 3 The average content of trace elements in the previous month, A 4 To manage data for the previous month, T is the temperature of the previous day, P is the average precipitation last month, M 3is the monthly mean temperature, M 4 The highest temperature last month, M 5 The highest precipitation in the previous month. t is a time variable, and its unit is month.

[0094] The calculation formula for the tree breast diameter prediction result is:

[0095] ,

[0096] in, Y t+1 is the tree breast diameter prediction result, Y jt is the breast diameter of the tree in the previous month. BA t is the basal area at breast height of the forest stand last month, Q is the climate factor function, θ is a random perturbation function, x, C 0 、C 1 、C 2 、C 3 、C 4 、C 5 is the generalized least squares estimate parameter, b 1 、b 2 、b 3 、 b 4 is the weight coefficient, A 1 For tree age, A 2 The average soil pH value of last month. A 3 The average content of trace elements in the previous month, A 4 To manage data for the previous month, T is the temperature of the previous day, P is the average precipitation last month, M 3 is the monthly mean temperature, M 4 The highest temperature last month. M 5 The highest precipitation in the previous month. t is a time variable, and its unit is month.

[0097] The tree height prediction result calculation formula is:

[0098] ,

[0099] in, Y t+1 is the tree height prediction result, C For the height of the tree on the moon, A is the maximum growth, r is the maximum growth rate, b is a parameter related to the assimilation power index, Q is the climate factor function, θ is a random perturbation function, x, C 0 、C 1 、C 2 、C 3 、C 4 、C 5 is the generalized least squares estimate parameter, b 1 、b 2 、b 3 、b 4 is the weight coefficient, A 1 For tree age, A 2 The average soil pH value of last month. A 3 The average content of trace elements in the previous month, A 4 To manage data for the previous month, T is the temperature of the previous day, P is the average precipitation last month, M 3 is the monthly mean temperature, M 4 The highest temperature last month, M 5 The highest precipitation in the previous month. t is a time variable, and its unit is month.

[0100] In this embodiment, the leaf area prediction results, tree breast diameter prediction results and tree height prediction results are obtained by summarizing and analyzing the data collected in the previous month, and the twin model is adjusted according to the leaf area prediction results, tree breast diameter prediction results and tree height prediction results to realize feedback on ancient trees with months as the time threshold. When the parameters of the twin model are adjusted for the first time, the tree breast diameter of the previous month, the tree height prediction result, and the leaf area of ​​the previous month in the calculation formula of the leaf area prediction results, the tree breast diameter prediction results and the tree height prediction results are the initially measured tree breast diameter, the initially measured tree height, and the initially measured leaf area.

[0101] Specifically, the specific calculation steps of the anomaly detection model are:

[0102] S501: Obtaining a video frame sequence by frame processing according to the image data around the ancient tree;

[0103] S502: Obtain a preprocessed frame sequence by image preprocessing according to the video frame sequence;

[0104] The image preprocessing includes grayscale, noise removal, histogram equalization, normalization, and edge detection.

[0105] S503: Calculate the abnormal result by using Timesformer algorithm according to the preprocessed frame sequence.

[0106] Preferably, the ancient tree twin model is integrated with the ancient tree landmark point according to the ancient tree location data, and the specific steps of the integration are: determine the ancient tree twin model corresponding to the ancient tree landmark point according to the ancient tree location data, obtain the storage location of the ancient tree twin model, and realize nesting through a jump function according to the ancient tree landmark point and the storage location.

[0107] In this embodiment, an event listener is added to each of the ancient tree mark points. The mouse is used to click on different ancient tree mark points on the virtual map. The click data is captured by the event listener and the jump function is triggered, so that the twin model of the ancient tree can be viewed by clicking the mark point on the electronic map.

[0108] An ancient tree growth status monitoring and maintenance system based on digital twins, comprising: a map building module, a twin model building module, a data acquisition module, and an anomaly detection module;

[0109] The map building module is used to construct a virtual map of the ancient tree management area through GIS technology, obtain the ancient tree location data, and mark the virtual map according to the ancient tree location data to obtain the ancient tree marking points;

[0110] The twin model building module is used to obtain the ancient tree point cloud data through the visual sensor, construct the ancient tree twin model by reconstructing the model according to the ancient tree point cloud data, and integrate the ancient tree twin model with the ancient tree landmark according to the ancient tree position data;

[0111] The data acquisition module is used to acquire daily data through the ecological one-shot communication system, obtain statistical data through statistical calculation based on the daily data, obtain ancient tree growth data through growth model calculation based on the statistical data, perform parameter correction on the ancient tree twin model based on the ancient tree growth data, obtain image data around the ancient tree through the visual sensor, and obtain real-time soil moisture data through the ecological one-shot communication system;

[0112] The anomaly detection module is used to calculate the anomaly result through the anomaly detection model using the image data around the ancient tree, compare the real-time soil moisture data with the preset moisture content threshold to obtain a comparison result, and the smart terminal generates an alarm adaptive signal according to the anomaly result to trigger an alarm, and the smart terminal generates an adaptive signal according to the comparison result to open the water supply valve.

[0113] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the ancient tree growth status monitoring and maintenance method based on digital twins as described above is implemented.

[0114] A storage medium containing computer-executable instructions, characterized in that the computer-executable instructions, when executed by a computer processor, are used to execute the ancient tree growth status monitoring and maintenance method based on digital twins as described above.

[0115] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0116] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0117] The program code contained on the computer-readable medium can be transmitted with any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or a combination thereof, and the programming language includes an object-oriented programming language-such as Java, Smalltalk, C++, and also includes a conventional procedural programming language-such as "C" language or similar programming language. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0118] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A method for monitoring and maintaining the growth status of ancient trees based on digital twins, characterized in that: include: Constructing a virtual map of the ancient tree management area through GIS technology, obtaining the location data of the ancient trees, and marking the virtual map according to the location data of the ancient trees to obtain the ancient tree marking points; Acquire ancient tree point cloud data through a visual sensor, construct a model based on the ancient tree point cloud data by reconstructing a model to obtain an ancient tree twin model, and integrate the ancient tree twin model with the ancient tree landmarks based on the ancient tree position data; The specific calculation steps of the point cloud segmentation model are as follows: S201: obtaining preprocessed point cloud data by preprocessing the ancient tree point cloud data; S202: performing point cloud segmentation according to the preprocessed point cloud data by using the LeWoS model to obtain a first segmented point cloud, and performing a similarity threshold test according to the first segmented point cloud to obtain a first classification label; S203: performing point cloud segmentation according to the pre-processed point cloud data by using the HSV color segmentation method to obtain a second segmented point cloud, and performing an H threshold test according to the second segmented point cloud to obtain a second classification label; S204: merging the first classification label and the second classification label according to the point cloud coordinates to obtain a merged label, and removing conflicting points from the ancient tree point cloud data according to the merged label to obtain a leaf segmentation point cloud and a branch segmentation point cloud; The specific calculation steps of the point cloud repair model are as follows: S301: performing triangulation meshing according to the branch segmentation point cloud to obtain a triangulated mesh point cloud; S302: performing hole identification on the triangular mesh point cloud according to the topological relationship of the center lines of the triangular facets to obtain the point cloud missing type; S303: determining a point cloud repair method according to the point cloud missing type, and repairing the branch segmentation point cloud according to the point cloud repair method to obtain a repaired branch point cloud; The specific calculation steps of the three-dimensional reconstruction model are: S401: Acquire color data and texture data of the repaired branch point cloud and the leaf segmentation point cloud; S402: Obtaining a whole tree slice point cloud by slicing according to the repaired branch point cloud and the leaf segmentation point cloud; S403: fitting and reconstructing the whole tree slice point cloud by using the RANSAC method to obtain a branch model; S404: splicing each of the branch models to obtain an initial model, and filling the initial model according to the color data and the texture data to obtain the ancient tree twin model; Obtain daily data through the ecological one-shot communication system, obtain statistical data through statistical calculation based on the daily data, obtain ancient tree growth data through growth model calculation based on the statistical data, perform parameter correction on the ancient tree twin model based on the ancient tree growth data, obtain image data around the ancient tree through the visual sensor, and obtain real-time soil moisture data through the ecological one-shot communication system; The abnormal result is calculated by the abnormal detection model through the image data around the ancient tree, and the real-time soil moisture data is compared with the preset moisture content threshold to obtain a comparison result. The smart terminal generates an alarm adaptive signal according to the abnormal result to trigger an alarm, and the smart terminal generates an adaptive signal according to the comparison result to open the water supply valve.

2. The ancient tree growth status monitoring and maintenance method based on digital twins according to claim 1 is characterized in that: The growth model is used to perform prediction calculations based on the statistical data to obtain leaf area prediction results, tree height prediction results, and tree breast diameter prediction results.

3. The ancient tree growth status monitoring and maintenance method based on digital twins according to claim 1 is characterized in that: The specific calculation steps of the anomaly detection model are: S501: Obtaining a video frame sequence through frame processing according to the image data around the ancient tree; S502: Obtain a preprocessed frame sequence by image preprocessing according to the video frame sequence; S503: Calculate the abnormal result by using Timesformer algorithm according to the preprocessed frame sequence.

4. The ancient tree growth status monitoring and maintenance method based on digital twins according to claim 1 is characterized in that: The ancient tree twin model is integrated with the ancient tree landmark point according to the ancient tree location data, and the specific steps of the integration are: determining the ancient tree twin model corresponding to the ancient tree landmark point according to the ancient tree location data, obtaining the storage location of the ancient tree twin model, and realizing nesting through a jump function according to the landmark point and the storage location.

5. An ancient tree growth status monitoring and maintenance system based on digital twins, characterized in that: include: Map building module, twin model building module, data acquisition module, anomaly detection module; The map building module is used to construct a virtual map of the ancient tree management area through GIS technology, obtain the ancient tree location data, and mark the virtual map according to the ancient tree location data to obtain the ancient tree marking points; The twin model building module is used to obtain the ancient tree point cloud data through the visual sensor, construct the ancient tree twin model by reconstructing the model according to the ancient tree point cloud data, and integrate the ancient tree twin model with the ancient tree landmark according to the ancient tree position data; The specific calculation steps of the point cloud segmentation model are as follows: S201: obtaining preprocessed point cloud data by preprocessing the ancient tree point cloud data; S202: performing point cloud segmentation according to the preprocessed point cloud data by using the LeWoS model to obtain a first segmented point cloud, and performing a similarity threshold test according to the first segmented point cloud to obtain a first classification label; S203: performing point cloud segmentation according to the pre-processed point cloud data by using the HSV color segmentation method to obtain a second segmented point cloud, and performing an H threshold test according to the second segmented point cloud to obtain a second classification label; S204: merging the first classification label and the second classification label according to the point cloud coordinates to obtain a merged label, and removing conflicting points from the ancient tree point cloud data according to the merged label to obtain a leaf segmentation point cloud and a branch segmentation point cloud; The specific calculation steps of the point cloud repair model are as follows: S301: performing triangulation meshing according to the branch segmentation point cloud to obtain a triangulated mesh point cloud; S302: performing hole identification on the triangular mesh point cloud according to the topological relationship of the center lines of the triangular facets to obtain the point cloud missing type; S303: determining a point cloud repair method according to the point cloud missing type, and repairing the branch segmentation point cloud according to the point cloud repair method to obtain a repaired branch point cloud; The specific calculation steps of the three-dimensional reconstruction model are: S401: Acquire color data and texture data of the repaired branch point cloud and the leaf segmentation point cloud; S402: Obtaining a whole tree slice point cloud by slicing according to the repaired branch point cloud and the leaf segmentation point cloud; S403: fitting and reconstructing the whole tree slice point cloud by using the RANSAC method to obtain a branch model; S404: splicing each of the branch models to obtain an initial model, and filling the initial model according to the color data and the texture data to obtain the ancient tree twin model; The data acquisition module is used to acquire daily data through the ecological one-shot communication system, obtain statistical data through statistical calculation based on the daily data, obtain ancient tree growth data through growth model calculation based on the statistical data, perform parameter correction on the ancient tree twin model based on the ancient tree growth data, obtain image data around the ancient tree through the visual sensor, and obtain real-time soil moisture data through the ecological one-shot communication system; The anomaly detection module is used to calculate the anomaly result through the anomaly detection model using the image data around the ancient tree, compare the real-time soil moisture data with the preset moisture content threshold to obtain a comparison result, and the smart terminal generates an alarm adaptive signal according to the anomaly result to trigger an alarm, and the smart terminal generates an adaptive signal according to the comparison result to open the water supply valve.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, it implements the ancient tree growth status monitoring and maintenance method based on digital twins as described in any one of claims 1-4.

7. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, are used to execute the ancient tree growth status monitoring and maintenance method based on digital twins as described in any one of claims 1 to 4.

Images