A forest damage assessment method, device and equipment based on multi-point laser radar

Abstract

The present invention provides a forest damage assessment method, device and equipment based on multi-point laser radar, the method comprising: dividing the forest area to be tested into a cubic grid, collecting original point cloud data through multi-point monitoring and deployment of wireless synchronous laser radar equipment, and performing first and second synchronous processing to obtain first and second point cloud data, constructing a complete forest point cloud model of the forest area to be tested, and optimizing to obtain a first feature model, extracting first feature parameters and second feature parameters, combining them into the damage index by weighted summation, and dividing the damage level of the forest area to be tested, statistically analyzing the area proportion and spatial distribution characteristics of each damage level, generating a forest damage assessment report, and proposing emergency response and forest management suggestions. The present invention realizes automated damage assessment of complex forest areas through damage index assessment with spatiotemporal data synchronization and weighted feature fusion, and the assessment results are intuitive and clear.

Description

Technical Field

[0001] The present invention belongs to the technical field of forest damage monitoring and assessment, and in particular relates to a forest damage assessment method, device and equipment based on multi-point laser radar. Background Art

[0002] With the improvement of ecological protection awareness in forest areas, forest damage assessment has become a key part of monitoring and management. At present, forest damage monitoring and assessment technologies mainly include manual field surveys, satellite remote sensing monitoring and single-point lidar scanning. Each technology has its advantages and disadvantages in practical applications, and the current forest damage assessment schemes mainly have the following problems:

[0003] (1) The data collection blind area is large. For example, single-point LiDAR scanning is often limited by the device's viewing angle, making it difficult to achieve comprehensive data collection over a large area, resulting in data blind areas and areas that are easily uncovered, affecting the comprehensiveness and accuracy of the assessment;

[0004] (2) They usually rely on manual experience and judgment, and lack a standardized damage indicator system, making it difficult to conduct unified assessments across regions and equipment, and the assessments of different types of forest areas lack accuracy and consistency;

[0005] (3) When dealing with large-scale forest damage assessments, manual intervention is usually relied on, and the degree of automation is low, resulting in a cumbersome and inefficient assessment process. In particular, in the assessment of large areas, manual analysis and calculations take a long time.

[0006] Therefore, we need to develop a forest damage assessment method, device and equipment based on multi-point lidar, which can monitor forestry in forest areas without blind spots and collect data, conduct standardized assessments of damage, and present corresponding visual assessment reports. Summary of the invention

[0007] The purpose of the present invention is to provide a forest damage assessment method, device and equipment based on multi-point lidar to solve the problems mentioned in the above background technology, such as insufficient data collection, inconsistent damage indicator system and low degree of automation of existing forest damage assessment schemes.

[0008] To achieve the above objectives, on the one hand, the present invention provides a forest damage assessment method based on multi-point laser radar, and the method is specifically as follows:

[0009] Divide the forest area to be measured into a cubic grid format, scan the forest area to be measured by multiple wirelessly synchronized laser radar devices, collect point cloud data, and obtain original point cloud data;

[0010] Performing a first synchronization process on the original point cloud data to obtain first point cloud data consistent with the coordinate system of the forest area to be measured, including point cloud coordinate data, and performing a second synchronization process on the first point cloud data to obtain second point cloud data collected by each of the laser radar devices at the same time, including point cloud time data;

[0011] Based on the second point cloud data, construct a complete forest point cloud model of the forest area to be measured, and perform model optimization to obtain a first feature model corresponding to the original point cloud data;

[0012] Based on the first characteristic model, extract the first characteristic parameter and the second characteristic parameter, and combine them into a damage index by weighted summation as a quantitative indicator reflecting the degree of damage to the forest area, and divide the damage level of the forest area to be tested based on the damage index;

[0013] Based on the damage level, the area proportion and spatial distribution characteristics of each damage level in the tested forest area are counted, a forest area damage assessment report is generated, and emergency response and forest area management suggestions are proposed based on the forest area damage assessment report.

[0014] Based on the above scheme, the cubic grid division is based on the size of the forest area to be measured, and the forest area to be measured is divided into n cubic grid units, and the side length a of each cubic grid unit is adjusted according to actual needs and the performance of the laser radar device;

[0015] Based on the above scheme, the wireless synchronization is to synchronize the time of the master node and the slave node through wireless communication technology, including: selecting a laser radar device located in the center of the forest area to be measured as the master node, and the remaining laser radar devices as the slave nodes. The master node is responsible for sending synchronization signals to the slave nodes to control the data acquisition process of the entire system.

[0016] Based on the above-mentioned scheme, the first synchronization processing is coordinate synchronization processing, specifically: initializing and calibrating each of the laser radar devices to make the laser radar device consistent with the coordinate system of the forest area to be measured, and calculating the three-dimensional coordinates (x, y, z) of the target point in the coordinate system of the forest area to be measured based on the original point cloud data, and combining the intensity returned by the target point to obtain the first point cloud data including the point cloud coordinate data.

[0017] Based on the above-mentioned scheme, the second synchronization processing is time synchronization processing, specifically: based on the first point cloud data, through a precise time synchronization protocol, ensure that the clock deviation between the master node and the slave node is controlled at the sub-microsecond level, the master node sends a timestamp to the slave node, and the slave node adjusts the local clock by calculating the timestamp difference, thereby achieving time synchronization between all the lidar devices and obtaining the second point cloud data including the point cloud time data.

[0018] Based on the above scheme, the construction of a complete forest point cloud model of the forest area to be tested is specifically as follows:

[0019] The second point cloud data is converted into a unified reference coordinate system, and each point cloud data in the second point cloud data is subjected to coordinate transformation to achieve spatial registration of the second point cloud data of different laser radars;

[0020] Using a point density-based filtering method, the number of neighboring points within a certain radius around each point in the second point cloud data is calculated, and if the number of neighboring points is lower than a preset threshold, the point is removed;

[0021] Using a progressive morphological filtering method, based on the geometric constraints of the points in the second point cloud data, including the distance and angle from the points to the triangular surface, to determine whether they are ground points, and separate the ground points;

[0022] Normalizing the intensity values ​​in the second point cloud data, linearly mapping the intensity values ​​to the interval [0, 1], and obtaining the mapping coefficients based on the maximum and minimum intensity values ​​of all points in the second point cloud data.

[0023] Based on the above scheme, the damage index is calculated as follows:

[0024] Divide the first feature model into cube grids of the same size and consistent with the scale of the cube grid units, and extract the first feature parameter and the second feature parameter of the point cloud data in each of the cube grids;

[0025] The calculation formula of the first characteristic parameter is: ,in represents the total number of point cloud data points in the i-th cube grid, represents the number of vegetation point cloud data points in the i-th cube grid, represents the average vegetation coverage of the healthy forest area, and the first characteristic parameter refers to the coverage completeness of the vegetation point cloud in the cube grid, that is, the vegetation coverage;

[0026] The calculation formula of the second characteristic parameter is: ,in represents the total number of point cloud data points in the i-th cube grid, represents the side length of the cube grid; the second characteristic parameter refers to the sparseness of the point cloud data in each cube grid, that is, the point cloud density;

[0027] Based on the first characteristic parameter and the second characteristic parameter, the damage index of each cube grid is constructed, and the damage index calculation formula is as follows: ,in represents the damage index of the ith cube grid, represents the first characteristic parameter of the i-th cube mesh, represents the second characteristic parameter of the i-th cube mesh, represents the weight coefficient of the first characteristic parameter, represents the weight coefficient of the second characteristic parameter, Represents the maximum value of the second characteristic parameter in all the i-th cube grids.

[0028] Based on the above scheme, the damage level is divided based on the actual situation of the forest area to be tested, historical monitoring data and expert knowledge, and an evaluation threshold is set to quantify the continuous damage index into discrete damage levels.

[0029] Based on the above scheme, the generation of the forest damage assessment report is specifically as follows: after completing the calculation of the damage index and the classification of the damage levels, an assessment report with the forest damage distribution map as the core is generated. The forest damage distribution map uses the cube grid as the basic unit, and identifies different damage levels with different colors or patterns, intuitively showing the regional distribution and degree of damage in the forest area to be tested, and enriching the content of the forest damage distribution map in combination with the forest area topography map and vegetation type map.

[0030] On the one hand, the present invention also provides a forest damage assessment device based on a multi-point laser radar, comprising:

[0031] A data acquisition module is used to divide the forest area to be measured into a cubic grid format, scan the forest area to be measured through multiple wirelessly synchronized laser radar devices, collect point cloud data, and obtain original point cloud data;

[0032] A data synchronization module, used for performing a first synchronization process on the original point cloud data to obtain first point cloud data consistent with the coordinate system of the forest area to be measured, including point cloud coordinate data, and performing a second synchronization process on the first point cloud data to obtain second point cloud data collected by each of the laser radar devices at the same time, including point cloud time data;

[0033] A data processing and modeling module, used for constructing a complete forest point cloud model of the forest area to be measured based on the second point cloud data, and performing model optimization to obtain a first feature model corresponding to the original point cloud data;

[0034] A forest damage assessment module, used for extracting the first characteristic parameter and the second characteristic parameter based on the first characteristic model, and combining them into the damage index by weighted summation as a quantitative indicator reflecting the degree of damage to the forest area, and dividing the damage level of the forest area to be tested based on the damage index;

[0035] The result analysis and suggestion module is used to count the area proportion and spatial distribution characteristics of each damage level in the forest area to be tested based on the damage level, generate a forest area damage assessment report, and put forward emergency response and forest area management suggestions based on the forest area damage assessment report.

[0036] On the other hand, the present invention also provides a forest damage assessment device based on a multi-point laser radar, comprising:

[0037] A processor configured to execute a forest damage assessment method based on multi-point laser radar as described in the above scheme;

[0038] A memory for storing a program for executing the forest area damage assessment method;

[0039] The processor and the memory are connected via a bus, and the bus comprises an I2C bus.

[0040] Compared with the prior art, the present invention has the following advantages and effects:

[0041] (1) By deploying multiple wireless synchronized LiDAR devices, we achieved all-round 3D point cloud data collection, ensuring that there were no blind spots in forest monitoring. We also designed a spatiotemporal data synchronization algorithm based on Euclidean distance to solve the problem of poor spatiotemporal synchronization when collecting data from multiple devices, avoiding the impact of time differences between devices on the evaluation results, and ensuring the comprehensiveness and accuracy of the data.

[0042] (2) A standardized damage index assessment method based on weighted feature fusion was designed, which comprehensively considered the two characteristic parameters of vegetation coverage integrity and point cloud density, combined with a dynamically adjusted weight coefficient to accurately assess the damage status of the forest area;

[0043] (3) Through the automated assessment process and the generation of damage index based on calculation, rapid and large-scale forest damage assessment can be achieved, and the assessment results can be presented through visual assessment reports, making it easier for forestry managers to intuitively understand the assessment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0045] Figure 1 It is a flow chart of a forest damage assessment method based on multi-point laser radar provided by an embodiment of the present invention;

[0046] Figure 2 It is a structural diagram of a forest damage assessment device based on a multi-point laser radar provided in an embodiment of the present invention;

[0047] Figure 3 It is a structural diagram of a forest damage assessment device based on a multi-point laser radar provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to more clearly explain the purpose, technical solutions and advantages of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The example implementation methods can be implemented in various forms and should not be understood as being limited to the examples described herein. On the contrary, these implementation methods are provided to make the present invention more comprehensive and complete, and to fully convey the concepts of the example implementation methods to those skilled in the art.

[0049] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, those skilled in the art will appreciate that the technical solution of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.

[0050] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0051] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0052] The present invention is described in detail below in conjunction with specific embodiments:

[0053] Example 1

[0054] As attached Figure 1 As shown, Embodiment 1 of the present invention provides a forest damage assessment method based on a multi-point laser radar, and the specific steps of the method are as follows:

[0055] Step S1, dividing the forest area to be measured into a cubic grid format, scanning the forest area to be measured by multiple wirelessly synchronized laser radar devices, collecting point cloud data, and obtaining original point cloud data;

[0056] Step S2: perform a first synchronization processing on the original point cloud data to obtain first point cloud data consistent with the coordinate system of the forest area to be measured, including point cloud coordinate data; perform a second synchronization processing on the first point cloud data to obtain second point cloud data collected by each of the laser radar devices at the same time, including point cloud time data.

[0057] Step S3: constructing a complete forest point cloud model of the forest area to be measured based on the second point cloud data, and performing model optimization to obtain a first feature model corresponding to the original point cloud data.

[0058] Step S4, based on the first characteristic model, extracting the first characteristic parameter and the second characteristic parameter, combining the first characteristic parameter and the second characteristic parameter into the damage index by weighted summation as a quantitative indicator reflecting the degree of damage to the forest area, and dividing the damage level of the forest area to be tested based on the damage index;

[0059] Step S5: based on the damage level, the area proportion and spatial distribution characteristics of each damage level in the forest area to be tested are counted, a forest area damage assessment report is generated, and emergency response and forest area management suggestions are proposed based on the forest area damage assessment report.

[0060] Preferably, in step S1, based on the size of the forest area to be measured, the forest area to be measured is divided into n cubic grid units, and the side length a of each cubic grid unit is adjusted according to actual needs and the performance of the laser radar device. In this embodiment, a is between 50 and 200 meters, and is preferably 100 meters. A laser radar device is arranged at the center of each cubic grid unit. In order to obtain the original point cloud data above the canopy, the height h is set higher than most canopies in the forest area to be measured. Preferably, h is 120% to 150% of the average height of the canopy in the forest area to be measured. For example, assuming that the average height of the canopy in the forest area to be measured is 30 meters, h can be set within 36 to 45 meters.

[0061] Specifically, the wireless synchronization is to synchronize the time of the master control node and the slave node through wireless communication technology, including: selecting a laser radar device located in the center of the forest area to be measured as the master control node, and the remaining laser radar devices as the slave nodes. The master control node is responsible for sending synchronization signals to the slave nodes to control the data acquisition process of the entire system; through the wireless communication module, reliable data transmission and command interaction can be carried out between the master control node and the slave nodes.

[0062] Preferably, all the laser radar devices in step S1 adopt the same or similar models and specifications to ensure the consistency and comparability of the original point cloud data.

[0063] It should be noted that each of the laser radar devices can optimize the point cloud data acquisition performance by adjusting the scanning point cloud parameters, and set a reasonable scanning angle range and angle resolution according to the device's ranging capability and forest environmental conditions; the higher the angle resolution, the finer the three-dimensional point cloud data obtained, but also the greater the data volume and data processing pressure, and it is necessary to find a balance between accuracy and efficiency.

[0064] Specifically, in this embodiment, the scanning angle range is preferably: 360-degree omnidirectional scanning is realized in the horizontal direction, and 90-degree range from the zenith to the horizon is covered in the vertical direction;

[0065] For example, when deploying the laser radar equipment, it is necessary to reasonably set scanning parameters to balance the quality and acquisition efficiency of the original point cloud data, wherein the vertical and horizontal scanning angles of the laser radar equipment determine the spatial coverage of the data; and are the number of vertical and horizontal scan lines, respectively. and are the vertical and horizontal angle resolutions respectively, and the vertical scanning angle range and horizontal scanning angle range The calculation formula is as follows: ; In more implementation situations, it may be preferred , , , ,but , (Close to 360°).

[0066] Preferably, the laser pulse repetition frequency (PRF) of the laser radar device is subject to an adaptive adjustment algorithm to optimize the efficiency of point cloud data acquisition in different areas and under different environmental conditions; it is subject to a multi-mode control algorithm to use a low PRF mode to scan in a longer distance area, and to achieve a higher data density by increasing the PRF in the area close to the target; it is subject to an intelligent scheduling algorithm to balance the speed of data acquisition and the system processing capacity to avoid data overload or processing delay;

[0067] Specifically, the adaptive adjustment algorithm, based on the environmental perception system, monitors the real-time environmental changes around the laser radar device in real time, and adjusts the PRF through a machine learning algorithm or a rule-based dynamic adjustment algorithm to adapt to environmental changes in different areas and optimize the acquisition efficiency and accuracy of the original point cloud data. The real-time environmental changes include changes in tree density, tree height, canopy coverage, terrain undulation, and weather conditions in the forest area to be measured; more specifically, in densely treed areas or complex terrain areas, the system automatically reduces the PRF value to reduce signal interference and improve the system's ranging accuracy, while in open areas or areas where the target is far away, the PRF value is automatically increased to increase data acquisition density;

[0068] Specifically, the multi-mode control algorithm includes: a low PRF mode suitable for long-distance scanning and a high PRF mode suitable for short-distance scanning, and the multi-mode control algorithm is used to set the corresponding low PRF mode or high PRF mode for each of the cubic grid unit areas to optimize data acquisition efficiency;

[0069] Specifically, the intelligent scheduling algorithm dynamically adjusts the PRF by real-time monitoring of the load of the lidar equipment to balance the data acquisition speed and system load capacity, automatically reduces the PRF when the load is high to avoid processing delays and data overload, and allocates a higher PRF to important areas to ensure high accuracy of data acquisition, while using a lower PRF for other areas to save computing resources and improve data acquisition efficiency; the load conditions include processor load, memory usage, and bandwidth; the important areas include fire risk areas and ecological monitoring areas;

[0070] Exemplarily, the frequency of emitting the laser beam of the lidar device is set at tens of thousands to hundreds of thousands of times per second. The specific value needs to be adjusted according to the equipment performance and actual needs. The repetition frequency (PRF) of the laser pulse emitted per second determines the amount of point cloud data obtained per second. Let t be the time interval of each pulse, then: PRF=1 / t, and the value of t needs to consider the laser performance and ranging range; if t=0.1ms, then PRF=10kHz, that is, the laser emits 10,000 laser pulses per second.

[0071] It should be noted that the laser radar device uses a frequency Send out laser pulses and receive return signals, each of which includes the distance : The distance from the laser emission point to the target point; Intensity : Returns the signal strength value, reflecting the reflection characteristics of the target point; Angle : The azimuth and elevation angles of the laser beam relative to the laser radar device coordinate system, where The selection of needs to take into account both the point cloud density and the amount of point cloud data;

[0072] Preferably, the first synchronization process in step S2 is coordinate synchronization process, specifically:

[0073] Initialize and calibrate each of the laser radar devices to make the laser radar device consistent with the coordinate system of the forest area to be measured, calculate the three-dimensional coordinates (x, y, z) of the target point in the coordinate system of the forest area to be measured based on the original point cloud data, and combine the intensity returned by the target point to obtain the first point cloud data including the point cloud coordinate data;

[0074] For example, take the , that is, 100,000 laser pulses are emitted per second. Assuming that the coordinates of the laser radar device in the coordinate system of the forest area to be measured are (X, Y, Z), the laser radar device calculates the three-dimensional coordinates (x, y, z) of each target point in the coordinate system of the forest area to be measured according to d, θ, φ. The calculation formula is: ; ; , a set of point cloud data collected by each of the laser radar devices within time T, that is, the first point cloud data is represents the three-dimensional coordinates (x, y, z) and intensity values ​​i of all target points collected by the laser radar device at time t;

[0075] For example, when collecting data, the vertical scanning angle range is determined. and horizontal scanning angle range , determine the pulse repetition rate (PRE), acquisition time , and the collection cycle After determining the above parameters, the master control node sends a synchronization instruction including the acquisition time T and the acquisition cycle Δt to each slave node through the wireless communication module; after receiving the instruction, the slave node automatically enters the data acquisition mode according to the unified time reference. During the acquisition process, the scanning parameters of the slave node are kept consistent with those of the master control node to ensure the comparability and consistency of the data;

[0076] Preferably, the second synchronization process in step S2 is time synchronization process, including:

[0077] Based on the first point cloud data, a precise time synchronization protocol is used to ensure that the clock deviation between the master node and the slave node is controlled at a sub-microsecond level, the master node sends a timestamp to the slave node, and the slave node adjusts the local clock by calculating the timestamp difference, thereby achieving time synchronization among all the laser radar devices, and obtaining the second point cloud data including the point cloud time data;

[0078] It should be noted that the second point cloud data is obtained by time synchronization processing of the first point cloud data, so the second point cloud data includes the point cloud coordinate data in addition to the point cloud time data;

[0079] For example, the master node and the slave nodes can communicate using a precise time protocol to control the clock deviation of each node to the sub-microsecond level to meet the synchronization requirements of multi-site data collection. Suppose the clock of the master node is , the clock of the slave node is , the deviation between the two is ; The master node of the LiDAR device sends an accurate timestamp to the slave node, which contains Time cut-off ; The slave node records the local time of receiving , and sends a delay request message to the master node, requesting the master node to record the timestamp of the message arrival; the master node records a timestamp when receiving the delay request message, and sends the timestamp back to the slave node through a delay response message, which contains , is the timestamp of the delayed request; the slave node calculates the time deviation between it and the master node through the timestamp, and then adjusts its own clock to achieve accurate time synchronization; the self-generated clock deviation is obtained according to the following formula, Δts=[(TR−TS)−(TP−TR)] / 2; in the above way, all lidar devices in the network can maintain time consistency and synchronization, and the slave node adjusts the local clock accordingly to achieve time synchronization with the master node.

[0080] Preferably, obtaining the first feature model corresponding to the original point cloud data in step S3 specifically includes the following:

[0081] Step S301: convert the second point cloud data into a unified reference coordinate system, wherein the reference coordinate system includes a geodetic coordinate system, and establish a conversion relationship between the laser radar device coordinate system and the reference coordinate system by measuring the precise installation position and angle of each laser radar device, and perform coordinate transformation on each point cloud data in the second point cloud data to achieve spatial registration of the second point cloud data of different laser radars;

[0082] It should be noted that since each lidar device has its own independent coordinate system, before data fusion, the first point cloud data collected by all devices need to be uniformly converted into a common reference coordinate system. The geodetic coordinate system of the forest area can be selected as the reference system. By measuring the precise installation position and angle of each device, establishing the conversion relationship between the device coordinate system and the reference coordinate system, and performing coordinate transformation on each point cloud data, spatial alignment of data from different devices can be achieved.

[0083] Step S302: using a point density-based filtering method, calculate the number of neighboring points within a certain radius around each point in the second point cloud data; if the number of neighboring points is lower than a preset threshold, remove the point.

[0084] It should be noted that due to factors such as equipment accuracy and environmental interference, there are inevitably some noise points in the point cloud data. These noise points usually have characteristics such as outliers in spatial position and abnormally low density. In order to remove noise points, a point density-based filtering method is adopted. The size of the preset threshold needs to be adjusted according to the characteristics of the data to balance the filtering effect and fidelity.

[0085] Step S303: using a progressive morphological filtering method, based on the geometric constraints of the points in the second point cloud data, including the distance and angle between the points and the triangular surface, to determine whether the points are ground points, and separate the ground points;

[0086] It should be noted that since the ground is usually a continuous and smooth plane, its point cloud features are obviously different from those of ground objects. In order to highlight the information of ground trees, it is necessary to separate the ground points from the point cloud. Using the progressive morphological filtering method, each round of iteration determines whether the point belongs to the ground based on geometric constraints such as the distance and angle from the point to the triangle. After multiple rounds of iterations, the ground points can be effectively separated. In more cases, other methods can also be used to separate ground points.

[0087] Step S304, normalizing the intensity values ​​in the second point cloud data, linearly mapping the intensity values ​​to the interval [0, 1], and calculating the mapping coefficients based on the maximum and minimum intensity values ​​of all points in the second point cloud data;

[0088] It should be noted that each point data returned by the lidar contains not only three-dimensional coordinate information, but also reflection intensity attributes. The intensity attribute reflects the surface material, roughness and other characteristics of the target object, which is very useful for distinguishing different types of land objects. However, the absolute values ​​of the intensity values ​​may vary greatly under different equipment and distance conditions. In order to eliminate this difference, the intensity value needs to be normalized and the original intensity value is linearly mapped to the standard range of 0~1. The mapping coefficient is calculated based on the maximum and minimum values ​​of the intensity of all points.

[0089] Step S305, the above is the optimization of the second point cloud data of a single laser radar device. Repeat steps S301 to S304 for all the laser radar devices to obtain a point cloud data set, fuse the point cloud data set to form a complete forest point cloud model of the forest area to be tested, and perform model optimization to obtain a first feature model corresponding to the original point cloud data.

[0090] Preferably, the damage index calculation in step S4 specifically includes:

[0091] Step S401: Based on the first feature model, extract the first feature parameter and the second feature parameter, and calculate a quantitative index reflecting the degree of damage to the forest area, that is, a damage index. The calculation of the damage index needs to comprehensively consider the geometric structure characteristics and radiation attribute characteristics of the forest area point cloud, which specifically includes the following:

[0092] Step S4011, dividing the first feature model into cube grids of the same size and consistent with the scale of the cube grid unit, the point cloud in each grid can reflect the average state of the forest in the area. Extracting the first feature parameter and the second feature parameter of the point cloud data in each cube grid;

[0093] Step S4012: The calculation formula of the first characteristic parameter is: ,in represents the total number of point cloud data points in the i-th cube grid, represents the number of vegetation point cloud data points in the i-th cube grid, represents the average vegetation coverage of the healthy forest area, which can be obtained by counting the damaged areas; the first characteristic parameter refers to the coverage completeness of the vegetation point cloud in the cube grid, that is, the vegetation coverage;

[0094] Specifically, the normalized intensity values ​​of all points in the cube grid are counted, and the points with normalized intensity values ​​greater than a preset intensity threshold, such as 0.3, are regarded as vegetation points, otherwise they are non-vegetation points. The ratio of vegetation points to the total number of points is calculated and recorded as vegetation coverage. The vegetation coverage in the cube grid is subtracted from the average vegetation coverage of the healthy forest area to obtain the first characteristic parameter. The value range of the first characteristic parameter is -1~0, and the closer to 0, the more complete the vegetation coverage.

[0095] Step S4013: The calculation formula of the second characteristic parameter is: ,in represents the total number of point cloud data points in the i-th cube grid, Represents the side length of the cube grid; the second characteristic parameter refers to the sparsity of the point cloud data in each cube grid, that is, the point cloud density, which is equal to the total number of points in the cube grid divided by the volume of the cube grid, and the unit is point / cubic meter. The larger the value, the denser the point cloud data and the more luxuriant the vegetation. Otherwise, there may be a lack of trees or sparse vegetation.

[0096] Step S4014: construct the damage index of each cube grid based on the first characteristic parameter and the second characteristic parameter. The damage index calculation formula is as follows: ,in represents the damage index of the ith cube grid, represents the first characteristic parameter of the i-th cube mesh, represents the second characteristic parameter of the i-th cube mesh, represents the weight coefficient of the first characteristic parameter, represents the weight coefficient of the second characteristic parameter, Represents the maximum value of the second characteristic parameter in all the i-th cube grids.

[0097] It should be noted that the first characteristic parameter and the second characteristic parameter are transformed respectively so that their changing directions are consistent, that is, the larger the value, the lighter the damage; for the first characteristic parameter, the value is between [-1,0], and the larger the value, the more complete the vegetation, so it is directly adopted In the form of, the larger the value of the first characteristic parameter, the more serious the damage; for the second characteristic parameter, divide it by the maximum value of the second characteristic parameter for normalization, and then use ( ) The larger the value of the second characteristic parameter, the lighter the degree of damage. The normalization process makes the numerical range of the first characteristic parameter and the second characteristic parameter consistent, which is convenient for subsequent weight assignment and result interpretation.

[0098] It should be noted that, in this embodiment, the weighted summation method is adopted to combine the two indicators of the first characteristic parameter and the second characteristic parameter into a comprehensive damage index; weighted summation is a commonly used multi-indicator evaluation method, and the importance of different indicators can be flexibly adjusted by setting weight coefficients; preferably, in this embodiment, the first characteristic parameter and the second characteristic parameter respectively reflect the vegetation coverage and the point cloud density, and the weight coefficients are set to 0.6 and 0.4 respectively, which means that the influence of vegetation coverage on the degree of damage to the forest area is slightly greater than that of the point cloud density; the reason for this setting is based on the following ecological considerations: the first characteristic parameter directly reflects the integrity of the canopy and is an intuitive expression of the degree of damage to the forest area, while the second characteristic parameter mainly reflects the standing tree density. Although it is related to the damage condition to a certain extent, it is not directly corresponding. For example, in natural sparse areas such as forest open spaces and forest windows, the second characteristic parameter value is lower, but it does not necessarily mean that it has been damaged. Therefore, giving the first characteristic parameter a higher weight can more accurately judge the damage condition of the forest area.

[0099] It should be noted that the weighted sum of the first characteristic parameter and the second characteristic parameter is used as the damage index value, which comprehensively considers the damage degree of the forest area based on the vegetation coverage and the standing tree density. The damage index ranges from [0,1]. The larger the value, the higher the degree of damage. That is, a damage index of 0 corresponds to a completely intact healthy forest area, and a damage index of 1 corresponds to a completely damaged forest area.

[0100] It should be noted that and are the weight coefficients of the two characteristics, reflecting their contribution to the damage index. In specific applications, the weight coefficients can be appropriately adjusted according to the actual situation of the forest area through on-site investigations and expert scoring. For example, in forest areas with more combustibles under the forest, the weight of the second characteristic parameter can be appropriately increased to highlight the impact of forest stand density on the risk of fire spread.

[0101] Step S402: setting an assessment threshold based on the actual situation of the forest area to be tested, historical monitoring data and expert knowledge, and quantifying the continuous damage index into discrete damage levels;

[0102] Specifically, M healthy forest areas are selected and the damage index is calculated. Mean of the values and standard deviation The terms "undamaged" and "healthy" are relative and need to be determined based on the actual situation of the local forest area and expert judgment. Usually, a well-protected mature forest stand away from human interference can be selected as a reference. Assume that the damage index of the healthy forest area is The value follows a normal distribution , calculate the following three evaluation thresholds: .

[0103] Specifically, based on the normal distribution , about 68.3% of the healthy forest stands Values ​​fall within the range [ f 0 - σ 0 , f 0 + σ 0 ] About 95.4% of the forest stands are healthy. Values ​​fall within the range [ f 0 - 2 σ 0 , f 0 +2 σ 0 ] About 99.7% of the forest stands are healthy. Values ​​fall within the range [ f 0 - 3 σ 0 , f 0 +3 σ 0 ] Inside.

[0104] Step S403: Based on the evaluation threshold, the damage level classification standard is obtained as follows: Lossless: ,Right now Values ​​are within normal fluctuation range for healthy stands; Slightly impaired: ,Right now The value is 1~2 standard deviations higher than the mean of healthy forest area, which indicates a high possibility of damage; moderate damage: ,Right now If the value is 2 to 3 standard deviations higher than the mean of healthy forest areas, the possibility of damage is high; severe damage: ,Right now The value is more than 3 standard deviations higher than the mean of healthy forest areas, indicating a high possibility of damage.

[0105] Preferably, in step S5, after completing the calculation of the damage index and the classification of the damage level, an intuitive and comprehensive forest damage assessment report is generated, the core of which is a forest damage distribution map, which uses grids as basic units and identifies different damage levels through different colors or patterns, intuitively showing the regional distribution and degree of damage in the forest area, and combining auxiliary information such as forest area topographic maps and vegetation type maps to enrich the content of the forest damage distribution map;

[0106] Based on the forest damage assessment report, the forest damage status is quantitatively analyzed through text and tables to analyze the damage trend of the forest area to be tested. The forest damage status includes the number and area of ​​grids of each damage level and their proportion in the total number of grids and total area, the spatial distribution characteristics of the damaged area, the comparison of the damage status of different vegetation types, and the longitudinal comparison with historical data.

[0107] Preferably, the forest damage assessment report is used to comprehensively evaluate the overall damage status and potential risks of the forest area to be tested, and corresponding forest management suggestions are put forward, which include: giving priority to on-site inspections of severely damaged areas, determining the priority of forest restoration based on the damaged area and regional ecological importance, selecting suitable afforestation tree species for artificial afforestation and stand improvement, setting up monitoring plots around severely damaged areas to carry out medium- and long-term ecological monitoring and assessment, strengthening daily patrols and fire source management, regularly conducting satellite remote sensing monitoring, and improving forest resource archives and databases; these measures will provide important references for forestry management decisions, accelerate forest vegetation recovery, and reduce human interference.

[0108] In this embodiment, the forest area to be tested is divided into a cubic grid, wireless synchronous laser radar equipment is deployed through multi-point monitoring to collect original point cloud data, and the first and second synchronous processing are performed to obtain the first and second point cloud data, a complete forest area point cloud model of the forest area to be tested is constructed, and the first feature model is optimized, the first feature parameter and the second feature parameter are extracted, and the weighted summation method is used to combine them into the damage index, and the damage level of the forest area to be tested is divided, the area proportion and spatial distribution characteristics of each damage level are statistically analyzed, a forest area damage assessment report is generated, and emergency response and forest area management suggestions are proposed, so as to realize automated damage assessment of complex forest areas, and the assessment results are intuitive and clear, and can accurately distinguish forest areas with different degrees of damage, providing a scientific basis for the protection and management of forestry resources.

[0109] Example 2

[0110] As attached Figure 2 As shown, Embodiment 2 of the present invention provides a forest damage assessment device 200 based on a multi-point laser radar, the system comprising:

[0111] The data acquisition module 201 is used to divide the forest area to be measured into a cubic grid format, scan the forest area to be measured through multiple wirelessly synchronized laser radar devices, collect point cloud data, and obtain original point cloud data;

[0112] The data synchronization module 202 is used to perform a first synchronization process on the original point cloud data to obtain first point cloud data consistent with the coordinate system of the forest area to be measured, including point cloud coordinate data, and perform a second synchronization process on the first point cloud data to obtain second point cloud data collected by each of the laser radar devices at the same time, including point cloud time data;

[0113] The data processing and modeling module 203 is used to construct a complete forest point cloud model of the forest area to be measured based on the second point cloud data, and optimize the model to obtain a first feature model corresponding to the original point cloud data;

[0114] The forest damage assessment module 204 is used to extract the first characteristic parameter and the second characteristic parameter based on the first characteristic model, and combine them into the damage index by weighted summation as a quantitative indicator reflecting the degree of damage to the forest area, and divide the damage level of the forest area to be tested based on the damage index;

[0115] The result analysis and suggestion module 205 is used to count the area proportion and spatial distribution characteristics of each damage level in the forest area to be tested based on the damage level, generate a forest area damage assessment report, and put forward emergency response and forest area management suggestions based on the forest area damage assessment report.

[0116] In some implementations, the data acquisition module 201 performs:

[0117] Multiple lidar scanning devices are deployed in the forest area to be tested, and the area is divided into regular cubic grid units. The device at the center is selected as the master control node, and the remaining devices are slave nodes. The master control node performs synchronous control and data transmission with the slave nodes through the wireless communication module. All devices use the same scanning parameters, including reasonable scanning angles, angular resolution, laser pulse frequency, etc., to ensure the consistency and comparability of the acquired original point cloud data.

[0118] In some implementations, the data synchronization module 202 performs:

[0119] By calibrating the coordinate system of each device and unifying it to the forest coordinate system, the first point cloud data of spatial registration, including point cloud coordinate data, is obtained; at the same time, the master and slave nodes use a precise time synchronization protocol to communicate and control the clock deviation between the nodes at the sub-microsecond level, thereby obtaining the second point cloud data collected synchronously, including point cloud time data.

[0120] In some embodiments, the data processing and modeling module 203 performs:

[0121] The point cloud data collected by each device are transformed into coordinates and spatially registered to be unified into a common reference coordinate system. Then, a point density-based filtering method is used to remove noise points, and a progressive morphological filtering algorithm is used to separate ground points from the point cloud. Normalization is performed to account for the differences in laser intensity values ​​caused by different devices and different distances. Finally, the optimized point cloud data of all devices are fused to construct a complete three-dimensional point cloud model of the forest area, and the first feature model is obtained after optimization.

[0122] In some embodiments, the forest damage assessment module 204 performs:

[0123] Based on the first characteristic model, a forest damage assessment was carried out. The assessment adopted the same cubic grid scale as data collection, and two characteristic parameters were extracted from the point cloud data in each grid: the first characteristic parameter characterized the completeness of vegetation coverage, which was obtained by calculating the proportion of vegetation points to the total number of points and comparing it with the average value of healthy forest areas; the second characteristic parameter reflected the standing tree density, which was obtained by dividing the total number of points in the grid by the grid volume.

[0124] Further, according to the point cloud data in each grid, the first characteristic parameter and the second characteristic parameter are extracted, and the damage index of each grid is constructed based on the extracted first characteristic parameter and the second characteristic parameter, and the damage index is obtained according to the following formula:

[0125]

[0126] represents the damage index of the ith grid, represents the first feature parameter extracted by the first feature model, represents the second feature parameter extracted by the first feature model, represents the weight coefficient of the first characteristic parameter, represents the weight coefficient of the second characteristic parameter, It represents the maximum value of the second characteristic parameter in all grids; the two characteristic parameters are weightedly summed to construct a comprehensive damage index with a value between 0 and 1, and the larger the value, the higher the degree of damage; referring to the damage index threshold, the forest area is divided into four levels: no damage, slight damage, moderate damage, and severe damage.

[0127] In some embodiments, the result analysis and suggestion module 205 performs:

[0128] Based on the damage level assessment results of the forest grid, the area proportions and spatial distribution characteristics of different damage levels are statistically analyzed to generate a forest damage assessment report. The report intuitively displays the damage status of the entire forest area in the form of a damage distribution map, and quantitatively analyzes the proportion of damaged area, differences in damage conditions in different regions and vegetation types, and historical change trends through text and data. Based on a comprehensive assessment of the overall damage status and potential risks of the forest area, practical and feasible forest management suggestions are put forward, including giving priority to checking and repairing severely damaged areas, carrying out artificial afforestation and replanting, strengthening daily monitoring and fire source management, and improving forest resource archives, providing a scientific basis for forestry management decisions.

[0129] Example 3

[0130] As attached Figure 3 As shown, Embodiment 3 of the present invention provides a forest damage assessment device 300 based on a multi-point laser radar, the device comprising:

[0131] The processor 301 is used to provide computing and control capabilities to support the operation of the entire terminal device, and is configured to execute a forest damage assessment method based on a multi-point laser radar;

[0132] The memory 302 is used to store a program for executing the forest damage assessment method;

[0133] The processor 301 and the memory 302 are connected via a bus 303 , and the bus 303 includes an I2C bus.

[0134] Specifically, the processor 301 includes a central processing unit (CPU), other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components; among which, the general-purpose processor includes a microprocessor and any conventional processor.

[0135] Specifically, the memory 302 includes a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0136] Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a partial structure related to the embodiment of the present invention, and does not constitute a limitation on the terminal device to which the embodiment of the present invention is applied. The specific server may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0137] The processor is used to run a computer program stored in a memory, and implement any one of the forest damage assessment methods using wireless synchronous multi-point detection provided by an embodiment of the present invention when executing the computer program.

[0138] In one embodiment, the processor is used to run a computer program stored in the memory, and implements the following steps when executing the computer program:

[0139] Divide the forest area to be measured into a cubic grid format, scan the forest area to be measured by multiple wirelessly synchronized laser radar devices, collect point cloud data, and obtain original point cloud data;

[0140] Performing a first synchronization process on the original point cloud data to obtain first point cloud data consistent with the coordinate system of the forest area to be measured, including point cloud coordinate data, and performing a second synchronization process on the first point cloud data to obtain second point cloud data collected by each of the laser radar devices at the same time, including point cloud time data;

[0141] Based on the second point cloud data, construct a complete forest point cloud model of the forest area to be measured, and perform model optimization to obtain a first feature model corresponding to the original point cloud data;

[0142] Based on the first characteristic model, extract the first characteristic parameter and the second characteristic parameter, and combine them into the damage index by weighted summation as a quantitative indicator reflecting the degree of damage to the forest area, and divide the damage level of the forest area to be tested based on the damage index;

[0143] Based on the damage level, the area proportion and spatial distribution characteristics of each damage level in the tested forest area are counted, a forest area damage assessment report is generated, and emergency response and forest area management suggestions are proposed based on the forest area damage assessment report.

[0144] In some implementations, the processor 301 executes:

[0145] Multiple lidar scanning devices are deployed in the forest area to be tested, and the area is divided into regular cubic grid units. The device at the center is selected as the master control node, and the remaining devices are slave nodes. The master control node performs synchronous control and data transmission with the slave nodes through the wireless communication module. All devices use the same scanning parameters, including reasonable scanning angles, angular resolution, laser pulse frequency, etc., to ensure the consistency and comparability of the acquired original point cloud data.

[0146] In some implementations, the processor 301 executes:

[0147] By calibrating the coordinate system of each device and unifying it to the forest coordinate system, the first point cloud data of spatial registration, including point cloud coordinate data, is obtained; at the same time, the master and slave nodes use a precise time synchronization protocol to communicate and control the clock deviation between the nodes at the sub-microsecond level, thereby obtaining the second point cloud data collected synchronously, including point cloud time data.

[0148] In some implementations, the processor 301 executes:

[0149] The point cloud data collected by each device are transformed into coordinates and spatially registered to be unified into a common reference coordinate system. Then, a point density-based filtering method is used to remove noise points, and a progressive morphological filtering algorithm is used to separate ground points from the point cloud. Normalization is performed to account for the differences in laser intensity values ​​caused by different devices and different distances. Finally, the optimized point cloud data of all devices are fused to construct a complete three-dimensional point cloud model of the forest area, and the first feature model is obtained after optimization.

[0150] In some embodiments, the processor 301 executes:

[0151] Based on the first characteristic model, a forest damage assessment was carried out. The assessment adopted the same cubic grid scale as data collection, and two characteristic parameters were extracted from the point cloud data in each grid: the first characteristic parameter characterized the completeness of vegetation coverage, which was obtained by calculating the proportion of vegetation points to the total number of points and comparing it with the average value of healthy forest areas; the second characteristic parameter reflected the standing tree density, which was obtained by dividing the total number of points in the grid by the grid volume.

[0152] Further, according to the point cloud data in each grid, the first characteristic parameter and the second characteristic parameter are extracted, and the damage index of each grid is constructed based on the extracted first characteristic parameter and the second characteristic parameter, and the damage index is obtained according to the following formula:

[0153]

[0154] represents the damage index of the ith grid, represents the first feature parameter extracted by the first feature model, represents the second feature parameter extracted by the first feature model, represents the weight coefficient of the first characteristic parameter, represents the weight coefficient of the second characteristic parameter, It represents the maximum value of the second characteristic parameter in all grids; the two characteristic parameters are weightedly summed to construct a comprehensive damage index with a value between 0 and 1, and the larger the value, the higher the degree of damage; referring to the damage index threshold, the forest area is divided into four levels: no damage, slight damage, moderate damage, and severe damage.

[0155] In some embodiments, the processor 301 executes:

[0156] Based on the damage level assessment results of the forest grid, the area proportions and spatial distribution characteristics of different damage levels are statistically analyzed to generate a forest damage assessment report. The report intuitively displays the damage status of the entire forest area in the form of a damage distribution map, and quantitatively analyzes the proportion of damaged area, differences in damage conditions in different regions and vegetation types, and historical change trends through text and data. Based on a comprehensive assessment of the overall damage status and potential risks of the forest area, practical and feasible forest management suggestions are put forward, including giving priority to checking and repairing severely damaged areas, carrying out artificial afforestation and replanting, strengthening daily monitoring and fire source management, and improving forest resource archives, providing a scientific basis for forestry management decisions.

[0157] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present invention, which follows the general principles of the present invention and includes common knowledge or customary techniques in the art that are not disclosed by the present invention. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present invention are indicated by the claims. It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A forest damage assessment method based on multi-point laser radar, characterized in that: include: Divide the forest area to be measured into a cubic grid format, scan the forest area to be measured by multiple wirelessly synchronized laser radar devices, collect point cloud data, and obtain original point cloud data; Performing a first synchronization process on the original point cloud data to obtain first point cloud data consistent with the coordinate system of the forest area to be measured, including point cloud coordinate data, and performing a second synchronization process on the first point cloud data to obtain second point cloud data collected by each of the laser radar devices at the same time, including point cloud time data; Based on the second point cloud data, construct a complete forest point cloud model of the forest area to be measured, and perform model optimization to obtain a first feature model corresponding to the original point cloud data; Based on the first characteristic model, extract the first characteristic parameter and the second characteristic parameter, and combine them into a damage index by weighted summation as a quantitative indicator reflecting the degree of damage to the forest area, and divide the damage level of the forest area to be tested based on the damage index; Based on the damage level, the area proportion and spatial distribution characteristics of each damage level in the tested forest area are counted, a forest area damage assessment report is generated, and emergency response and forest area management suggestions are put forward based on the forest area damage assessment report; The damage index calculation is specifically as follows: Divide the first feature model into cube grids of the same size and consistent with the scale of the cube grid units, and extract the first feature parameter and the second feature parameter of the point cloud data in each of the cube grids; The calculation formula of the first characteristic parameter is: ,in represents the total number of point cloud data points in the i-th cube grid, represents the number of vegetation point cloud data points in the i-th cube grid, represents the average vegetation coverage of the healthy forest area, and the first characteristic parameter refers to the coverage completeness of the vegetation point cloud in the cube grid, that is, the vegetation coverage; The calculation formula of the second characteristic parameter is: ,in represents the total number of point cloud data points in the i-th cube grid, represents the side length of the cube grid; the second characteristic parameter refers to the sparseness of the point cloud data in each cube grid, that is, the point cloud density; Based on the first characteristic parameter and the second characteristic parameter, the damage index of each cube grid is constructed, and the damage index calculation formula is as follows: ,in represents the damage index of the ith cube grid, represents the first characteristic parameter of the i-th cube mesh, represents the second characteristic parameter of the i-th cube mesh, represents the weight coefficient of the first characteristic parameter, represents the weight coefficient of the second characteristic parameter, Represents the maximum value of the second characteristic parameter in all the i-th cube grids.

2. The forest damage assessment method based on multi-point laser radar according to claim 1 is characterized in that: The cubic grid division is based on the size of the forest area to be measured, dividing the forest area to be measured into n cubic grid units, and the side length a of each cubic grid unit is adjusted according to actual needs and the performance of the laser radar device; The wireless synchronization is to synchronize the time of the master node and the slave node through wireless communication technology, including: selecting a laser radar device located in the center of the forest area to be measured as the master node, and the remaining laser radar devices as the slave nodes. The master node is responsible for sending synchronization signals to the slave nodes to control the data collection process of the entire system.

3. The forest damage assessment method based on multi-point laser radar according to claim 1 is characterized in that: The first synchronization processing is coordinate synchronization processing, specifically: initializing and calibrating each of the laser radar devices to make the laser radar device consistent with the coordinate system of the forest area to be measured, and calculating the three-dimensional coordinates (x, y, z) of the target point in the coordinate system of the forest area to be measured based on the original point cloud data, and combining the intensity returned by the target point to obtain the first point cloud data including the point cloud coordinate data.

4. The forest damage assessment method based on multi-point laser radar according to claim 2 is characterized in that: The second synchronization processing is time synchronization processing, specifically: based on the first point cloud data, through a precise time synchronization protocol, ensure that the clock deviation between the master node and the slave node is controlled at the sub-microsecond level, the master node sends a timestamp to the slave node, and the slave node adjusts the local clock by calculating the timestamp difference, thereby achieving time synchronization between all the lidar devices and obtaining the second point cloud data including the point cloud time data.

5. The forest damage assessment method based on multi-point laser radar according to claim 1 is characterized in that: The step of constructing a complete forest point cloud model of the forest area to be tested is as follows: The second point cloud data is converted into a unified reference coordinate system, and each point cloud data in the second point cloud data is subjected to coordinate transformation to achieve spatial registration of the second point cloud data of different laser radars; Using a point density-based filtering method, the number of neighboring points within a certain radius around each point in the second point cloud data is calculated, and if the number of neighboring points is lower than a preset threshold, the point is removed; Using a progressive morphological filtering method, based on the geometric constraints of the points in the second point cloud data, including the distance and angle from the points to the triangular surface, to determine whether they are ground points, and separate the ground points; Normalizing the intensity values ​​in the second point cloud data, linearly mapping the intensity values ​​to the interval [0, 1], and obtaining the mapping coefficients based on the maximum and minimum intensity values ​​of all points in the second point cloud data.

6. The forest damage assessment method based on multi-point laser radar according to claim 1 is characterized in that: The damage level is divided based on the actual situation of the forest area to be tested, historical monitoring data and expert knowledge, and an evaluation threshold is set to quantify the continuous damage index into discrete damage levels.

7. The forest damage assessment method based on multi-point laser radar according to claim 1 is characterized in that: The generation of the forest damage assessment report is specifically as follows: after completing the calculation of the damage index and the classification of the damage levels, an assessment report with a forest damage distribution map as the core is generated. The forest damage distribution map uses the cube grid as a basic unit, and identifies different damage levels by different colors or patterns, intuitively showing the regional distribution and degree of damage in the forest area to be tested, and enriching the content of the forest damage distribution map in combination with the forest topographic map and vegetation type map.

8. A forest damage assessment device based on multi-point laser radar, used to implement a forest damage assessment method based on multi-point laser radar as claimed in any one of claims 1 to 7, characterized in that: include: A data acquisition module is used to divide the forest area to be measured into a cubic grid format, scan the forest area to be measured through multiple wirelessly synchronized laser radar devices, collect point cloud data, and obtain original point cloud data; A data synchronization module, used for performing a first synchronization process on the original point cloud data to obtain first point cloud data consistent with the coordinate system of the forest area to be measured, including point cloud coordinate data, and performing a second synchronization process on the first point cloud data to obtain second point cloud data collected by each of the laser radar devices at the same time, including point cloud time data; A data processing and modeling module, used for constructing a complete forest point cloud model of the forest area to be measured based on the second point cloud data, and performing model optimization to obtain a first feature model corresponding to the original point cloud data; A forest damage assessment module, used for extracting the first characteristic parameter and the second characteristic parameter based on the first characteristic model, and combining them into a damage index by weighted summation as a quantitative indicator reflecting the degree of damage to the forest area, and dividing the damage level of the forest area to be tested based on the damage index; The result analysis and suggestion module is used to count the area proportion and spatial distribution characteristics of each damage level in the forest area to be tested based on the damage level, generate a forest area damage assessment report, and put forward emergency response and forest area management suggestions based on the forest area damage assessment report.

9. A forest damage assessment device based on multi-point laser radar, characterized in that: include: A processor configured to execute a forest damage assessment method based on multi-point laser radar as described in any one of claims 1 to 7; A memory for storing a program for executing the forest area damage assessment method; The processor and the memory are connected via a bus, and the bus comprises an I2C bus.

Images