A forest fire monitoring method and system based on lidar and thermal infrared

Abstract

The present application provides a wildfire monitoring method and system based on lidar and thermal infrared. Among them, the method includes: combining three-dimensional point cloud data with thermal radiation distribution data to generate a target fire spread path; using the embedded real-time processing chip deployed in the detection drone to process the thermal radiation distribution data in real time, identify temperature anomaly areas and remove interference areas, and generate fire point locations and classification results; based on the target fire spread path, fire point location and classification results, dynamic scheduling instructions are sent to the fire-fighting drone cluster through a low-latency communication link to adjust the location, fire extinguishing dosage and coverage of the fire-fighting drones in the fire-fighting drone cluster. The present application collects three-dimensional point cloud and thermal radiation data of the wildfire area in real time, combines intelligent algorithms to predict the fire spread path, and dynamically schedules the fire-fighting drone cluster, which can achieve precise positioning, interference filtering and optimal spatial allocation of fire-fighting resources.

Description

Technical Field

[0001] The present application relates to the field of wildfire monitoring technology, and in particular to a wildfire monitoring method and system based on lidar and thermal infrared. Background Art

[0002] Wildfires are one of the most serious natural disasters affecting forest ecosystems. They destroy a significant amount of land worldwide each year, posing a direct threat to biodiversity and the viability of ecological conservation. Wildfire monitoring and suppression can mitigate the damage caused by these disasters. Therefore, integrating modern equipment to achieve effective wildfire monitoring, timely suppression, and rational strategies is a current research priority.

[0003] The existing solution uses drones equipped with infrared sensors and visible light cameras on fixed routes to collect two-dimensional thermal and video data of the fire scene along a preset route. This data is then used in conjunction with an edge computing platform to coarsely locate the fire point and generate a static firefighting plan based on historical fire spread models. After the data is transmitted back, a central server at a ground station is used to predict the fire spread and assign tasks.

[0004] However, existing solutions rely on two-dimensional thermal imaging data and fixed scanning routes, making it difficult to obtain three-dimensional structural information from vertical cross-sections of a fire scene, resulting in insufficient accuracy in predicting the spread of fire. Furthermore, the fusion of infrared and visible light data is limited in its ability to filter out interference from smoke and high-temperature non-fire sources. Furthermore, the centralized processing model at the ground station introduces communication delays, making it difficult to meet the timeliness requirements for rapid response to dynamic fire scenes. Summary of the Invention

[0005] The present application provides a wildfire monitoring method and system based on lidar and thermal infrared to solve the problems in the prior art such as insufficient three-dimensional fire scene structure perception capability, single multi-source data fusion dimension, weak interference suppression capability and response delay, which lead to poor wildfire monitoring effect and unsatisfactory firefighting effect.

[0006] In a first aspect, the present application provides a wildfire monitoring method based on lidar and thermal infrared, comprising:

[0007] By controlling the flight altitude and heading angle of the detection drone, the laser radar and thermal infrared sensor deployed on the detection drone collect three-dimensional point cloud data of the vertical section of the wildfire area and thermal radiation distribution data of the wildfire area from multiple angles.

[0008] Combining the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path;

[0009] Using the embedded real-time processing chip deployed in the detection drone, the thermal radiation distribution data is processed in real time to identify abnormal temperature areas and remove interference areas, generating fire point locations and classification results;

[0010] Based on the target fire spread path, the fire point location and the classification result, dynamic scheduling instructions are sent to the fire-fighting drone cluster through a low-latency communication link to adjust the location, fire-fighting dosage and coverage range of the fire-fighting drones in the fire-fighting drone cluster.

[0011] Optionally, combining the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path includes:

[0012] extracting wildfire terrain elevation features from the three-dimensional point cloud data;

[0013] extracting temperature field gradient variation characteristics from the thermal radiation distribution data;

[0014] Based on the wildfire terrain elevation characteristics and the temperature field gradient change characteristics, combined with the historical meteorological data of the wildfire area, a target fire spread path is generated.

[0015] Optionally, the wildfire terrain elevation features include: an elevation value sequence, a slope change rate set, and a slope aspect angle set; the temperature field gradient change features include: a temperature extreme value region boundary, a temperature gradient direction vector, and a temperature gradient intensity distribution; and the historical meteorological data include a wind speed vector, humidity, and vegetation type;

[0016] The target fire spread path is generated based on the wildfire terrain elevation characteristics and the temperature field gradient change characteristics, combined with historical meteorological data of the wildfire area, including:

[0017] Performing differential calculation on the elevation value sequence to obtain a distribution of terrain elevation change rates;

[0018] Performing nonlinear mapping on the slope change rates and the temperature gradient direction vectors in the slope change rate set to generate a dynamic impact intensity parameter representing the characteristics of the terrain on the temperature field gradient change;

[0019] Extracting topological structure features from the boundary of the temperature extreme value region, wherein the topological structure features include the number of connected domains and the boundary curvature radius;

[0020] The slope angle set is partitioned into azimuth zones, a covariance coefficient between the slope angle and the temperature gradient intensity distribution in each zone is calculated, and a zone having a covariance coefficient greater than a preset coefficient threshold is defined as a fire spread sensitive zone;

[0021] Calculating the thermal convection intensity of the fire front under different terrain conditions based on the terrain elevation change rate distribution, calculating the diffusion rate of the fire front based on the vegetation type, humidity, and wind speed vector, and correcting the direction of the fire front based on the dynamic impact intensity parameter to generate multiple candidate fire spread paths;

[0022] According to the spatial coupling relationship between the distribution range of the fire spread sensitive area and the topological structure characteristics, a probability simulation is performed on each candidate fire spread path to screen out a target fire spread path that meets the preset diffusion conditions.

[0023] Optionally, performing nonlinear mapping on the slope change rates in the slope change rate set and the temperature gradient direction vector to generate a dynamic impact intensity parameter characterizing the terrain's influence on the temperature field gradient change characteristics includes:

[0024] Normalizing all the slope change rates in the slope change rate set to obtain a standardized slope change rate vector;

[0025] Decompose the temperature gradient direction vector into a horizontal component and a vertical component, and calculate the first cosine similarity between the horizontal component and the normalized slope change rate vector and the second cosine similarity between the vertical component and the normalized slope change rate vector respectively;

[0026] performing exponential weighting on the first cosine similarity and the second cosine similarity to obtain a first weighted value and a second weighted value, and taking the sum of the first weighted value and the second weighted value as a directional consistency parameter;

[0027] The directional consistency parameter is normalized to obtain the dynamic impact intensity parameter.

[0028] Optionally, the heat convection intensity of the fire front under different terrain conditions is calculated based on the terrain elevation change rate distribution, the diffusion rate of the fire front is calculated based on the vegetation type, humidity, and wind speed vector, and the direction of the fire front is corrected based on the dynamic impact intensity parameter to generate multiple candidate fire spread paths, including:

[0029] Calculating heat convection intensity correction coefficients for a first location and a second location within the volcanic region according to formulas with different parameter values, respectively; the first location being a location corresponding to a terrain elevation change rate greater than a preset elevation change rate threshold in the terrain elevation change rate distribution; and the second location being a location corresponding to a terrain elevation change rate less than or equal to the preset elevation change rate threshold in the terrain elevation change rate distribution;

[0030] The thermal convection intensity at each location is obtained by multiplying the thermal convection intensity correction coefficient, the intensity reference value, and the correction coefficient corresponding to the vegetation type.

[0031] The diffusion rate is calculated using a preset diffusion rate calculation formula according to the vegetation coefficient, rate reference value, humidity and wind speed vector corresponding to the vegetation type;

[0032] Calculate the extension length of the path based on the heat convection intensity and diffusion rate at each location;

[0033] determining a direction correction coefficient corresponding to the dynamic impact intensity parameter, and correcting the direction of the firing line based on the direction correction coefficient to obtain a corrected direction;

[0034] The starting point of the path is initialized to the current coordinates of the fire front. Based on the corrected direction and the extension length of the path, combined with path screening conditions, multiple candidate fire spread paths are generated. The path screening conditions include: the overlapping area between the path and the fire spread sensitive area is greater than a preset overlapping area threshold, the diffusion rate of the path is greater than a preset rate threshold, and the diffusion duration is greater than a preset duration threshold.

[0035] Optionally, the embedded real-time processing chip deployed in the detection drone is used to process the thermal radiation distribution data in real time, identify abnormal temperature areas and remove interference areas, and generate fire point locations and classification results, including:

[0036] Using the embedded real-time processing chip deployed in the detection drone, the thermal radiation distribution data is subjected to median filtering to eliminate environmental noise interference and generate a noise-reduced temperature field matrix;

[0037] Processing the noise-reduced temperature field matrix to determine a target fire point area;

[0038] Extracting a temperature distribution feature vector from the target fire point region, wherein the temperature distribution feature vector includes the average temperature of the region, the maximum temperature gradient, the standard deviation of the temperature distribution, and the ratio of the perimeter of the region to the area;

[0039] Using the K-means clustering algorithm to perform dynamic threshold division on the temperature distribution feature vector, and generating fire point type labels and confidence parameters based on the Euclidean distance matching between the cluster center and the historical fire point samples, the fire point type labels are used as the classification results. The fire point type labels include surface fire, smoldering fire and re-ignition fire;

[0040] Based on the spatial coordinates of the target fire point area corresponding to the three-dimensional point cloud data, the three-dimensional geographic coordinates of the fire point center in the UAV navigation coordinate system are calculated, and the three-dimensional geographic coordinates are used as the fire point position.

[0041] Optionally, processing the noise-reduced temperature field matrix to determine a target fire point area includes:

[0042] Binarizing and segmenting the denoised temperature field matrix based on a preset temperature threshold to extract candidate temperature anomaly regions, and merging adjacent candidate temperature anomaly regions using a region growing algorithm to generate an initial fire point region set;

[0043] Deleting fire point regions whose areas are smaller than a preset area threshold from the initial fire point region set to obtain an intermediate fire point region set;

[0044] Fire point areas with reflectivity lower than a preset reflectivity threshold are screened out from the set of intermediate fire point areas to obtain target fire point areas.

[0045] In a second aspect, the present application provides a wildfire monitoring system based on lidar and thermal infrared, comprising:

[0046] The acquisition module is used to control the flight altitude and heading angle of the detection drone, so that the laser radar and thermal infrared sensor deployed on the detection drone can respectively collect three-dimensional point cloud data of the vertical section of the wildfire area and thermal radiation distribution data of the wildfire area from multiple angles;

[0047] A generation module, configured to combine the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path;

[0048] An identification module is used to process the thermal radiation distribution data in real time using an embedded real-time processing chip deployed in the detection drone, identify temperature anomaly areas and remove interference areas, and generate fire point locations and classification results;

[0049] The adjustment module is used to send dynamic scheduling instructions to the fire-fighting drone cluster through a low-latency communication link based on the target fire spread path, the fire point location and the classification result, so as to adjust the position, fire extinguishing dosage and coverage range of the fire-fighting drones in the fire-fighting drone cluster.

[0050] In a third aspect, the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a wildfire monitoring method based on lidar and thermal infrared as described in any one of the first aspects.

[0051] In a fourth aspect, the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a forest fire monitoring method based on lidar and thermal infrared as described in any one of the first aspects.

[0052] In an embodiment of the present application, a wildfire monitoring method based on lidar and thermal infrared is provided, the method comprising: controlling the flight altitude and heading angle of the detection drone so that the lidar and thermal infrared sensor deployed on the detection drone respectively collect three-dimensional point cloud data of the vertical section of the wildfire area and thermal radiation distribution data of the wildfire area at multiple angles; combining the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path; utilizing an embedded real-time processing chip deployed in the detection drone to perform real-time processing on the thermal radiation distribution data, identify temperature abnormality areas and remove interference areas, and generate fire point locations and classification results; based on the target fire spread path, the fire point locations and the classification results, dynamic scheduling instructions are sent to a cluster of fire-fighting drones through a low-latency communication link to adjust the position, fire extinguishing dosage and coverage range of the fire-fighting drones in the cluster.

[0053] This application dynamically adjusts the flight parameters of the detection drone, combines the multi-angle collaborative scanning of the lidar and thermal infrared sensors, and achieves high-precision reconstruction of the three-dimensional point cloud of the vertical section of the wildfire area and omnidirectional coverage of the thermal radiation field, breaking through the dimensional limitations of traditional planar data; based on the fusion modeling of three-dimensional space and thermal field data, it constructs the dynamic prediction capability of the fire spread path and enhances the credibility of fire situation deduction under complex terrain; uses airborne embedded real-time processing technology to perform online denoising and fire point classification on thermal radiation data, reduces the misjudgment rate of smoke and high-temperature non-fire source targets, and improves the accuracy of fire point positioning; relies on low-latency communication and dynamic scheduling algorithms to achieve real-time task re-planning of fire-fighting drone clusters, optimizes the spatial matching and dosage control accuracy of fire extinguishing agent delivery, forms a "perception-decision-execution" closed-loop control link, and improves the response efficiency and resource utilization of wildfire fighting.

[0054] Furthermore, by extracting terrain elevation features from 3D point cloud data, analyzing temperature field gradient characteristics from thermal radiation data, and combining them with historical meteorological data, a multidimensional fire spread prediction model was constructed. Specifically, this model generates dynamic impact intensity parameters based on a nonlinear mapping of terrain elevation change rate and temperature gradient direction vectors. Fire spread-sensitive areas are screened using the aspect angle partitioning covariance coefficient. The fire front diffusion rate and thermal convection intensity are calculated based on vegetation type, humidity, and wind speed. The fire direction is then modified to generate candidate paths. Finally, the topological structure characteristics are used to spatially couple sensitive areas, and probabilistic simulation is used to select target fire spread paths. By finely integrating the three-dimensional structure of terrain, the evolution of temperature field gradients, and meteorological dynamic parameters, a fire spread model coupling topographic, thermal, and meteorological multi-physical fields is established, thereby improving the spatiotemporal accuracy of fire front predictions in complex environments. Based on nonlinear mapping and covariant coefficient screening mechanisms, the ability to quantitatively characterize the dynamic impact of terrain on fire direction is enhanced, reducing path misjudgments caused by misalignment of slope aspect, slope, and temperature gradient. Through probabilistic simulation and topological structure coupling analysis, physical constraint screening of candidate paths is achieved, avoiding the overfitting risk of a single diffusion model, and ultimately generating a high-confidence target fire spread path, providing a reliable decision-making basis for the dynamic scheduling of fire-fighting resources.

[0055] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0057] Figure 1 A flowchart of a wildfire monitoring method based on lidar and thermal infrared provided in an embodiment of the present application;

[0058] Figure 2 A schematic diagram of the structure of a wildfire monitoring system based on lidar and thermal infrared provided in an embodiment of the present application;

[0059] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0061] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 11, 12, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0062] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0063] To address the problems in the prior art of insufficient three-dimensional fire scene structure perception, single multi-source data fusion dimension, weak interference suppression capability, and response delay, which lead to poor wildfire monitoring results and unsatisfactory firefighting effects, the present invention provides a wildfire monitoring method based on lidar and thermal infrared. The method adopts the following concepts: with multimodal sensor collaborative acquisition and real-time decision-making as the core, a dynamic flight control strategy for detection drones is designed. By adaptively adjusting the flight altitude and heading angle, multi-angle vertical scanning of the lidar and omnidirectional radiation coverage of the thermal infrared sensor are achieved, and a spatial coupling model of three-dimensional point cloud and thermal field data is constructed; an online cleaning algorithm for thermal radiation data is developed based on an airborne embedded chip, and interference suppression is achieved by combining temperature gradient threshold segmentation and morphological filtering. The terrain elevation features and meteorological parameters are simultaneously integrated to construct the fire spread dynamics equation; finally, relying on edge computing and low-latency communication technology, a dynamic task allocation mechanism for firefighting drone clusters is established, forming a full-link closed-loop system from high-precision perception, multi-physics field modeling, real-time decision-making to precise execution, breaking through the data dimension limitations and response bottlenecks of traditional solutions, thereby improving wildfire monitoring and firefighting effects.

[0064] Figure 1 A flowchart of a wildfire monitoring method based on lidar and thermal infrared is provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0065] S11. By controlling the flight altitude and heading angle of the detection drone, the laser radar and thermal infrared sensor deployed on the detection drone collect three-dimensional point cloud data of the vertical section of the wildfire area and thermal radiation distribution data of the wildfire area at multiple angles.

[0066] Among them, the thermal infrared sensor may refer to a device that detects the temperature distribution of the target area based on the principle of infrared thermal radiation, and generates a temperature field image by receiving radiation energy in a specific band, which is used to reflect the thermal distribution and gradient change characteristics of the wildfire area. The specific band can be medium-wave infrared or long-wave infrared, etc. The laser radar may refer to a remote sensing device that obtains the distance and shape of the target by emitting laser pulses and measuring the time difference of the reflected signal. In the embodiment of the present application, it can generate three-dimensional point cloud data based on the time-of-flight principle for reconstructing the vertical profile structure of the terrain. Three-dimensional point cloud data refers to a spatial coordinate data set generated by a laser radar scan, which contains the elevation, slope and aspect information of the target area, and is used for terrain modeling and fire spread analysis. The thermal radiation distribution data of the wildfire area refers to the two-dimensional matrix data of the temperature field collected by the thermal infrared sensor, which reflects the thermal radiation intensity and gradient direction of the fire scene, and is used to identify areas with abnormal temperatures.

[0067] In an embodiment of the present application, first, the flight altitude and heading angle of the detection drone are adjusted through a dynamic flight control algorithm to enable it to fly in a multi-angle circle; secondly, the lidar is controlled to perform a vertical profile scan of the wildfire area, emitting laser pulses and receiving reflected signals to generate three-dimensional point cloud data containing elevation, slope and slope direction; at the same time, the thermal infrared sensor synchronously collects the thermal radiation energy of the wildfire area to generate temperature field distribution data; finally, the three-dimensional point cloud data and the thermal radiation distribution data are matched according to coordinates and timestamps through a spatiotemporal alignment algorithm to form a fused data set.

[0068] S12. Combine the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path.

[0069] Among them, the target fire spread path refers to the predicted path generated based on the fusion of terrain, temperature field and meteorological data, and the fire spread direction with high confidence is screened out through probabilistic simulation.

[0070] In an embodiment of the present application, first, terrain elevation features are extracted from three-dimensional point cloud data, including elevation value sequences, slope change rates, and aspect angle sets; secondly, gradient analysis is performed on the thermal radiation distribution data to extract the boundaries of temperature extreme areas, gradient direction vectors, and intensity distributions; subsequently, the terrain features and the temperature field gradient are input into a nonlinear mapping model to calculate the intensity parameters of the dynamic impact of the terrain on the temperature field; then, fire spread sensitive areas are screened based on the covariance coefficient of the aspect angle partition and the temperature gradient intensity; finally, a fire spread dynamics model is constructed in combination with the wind speed, humidity, and vegetation type in historical meteorological data, multiple candidate paths are generated, and the target fire spread path is determined through probabilistic simulation.

[0071] S13. Use the embedded real-time processing chip deployed in the detection drone to process the thermal radiation distribution data in real time, identify temperature anomaly areas and remove interference areas, and generate fire point locations and classification results.

[0072] The embedded real-time processing chip is a high-performance computing module integrated into the detection drone. It supports parallel processing of thermal radiation data for online denoising, fire point identification, and classification. Interference area removal uses morphological filtering and connected domain analysis to eliminate high-temperature areas that are not fire sources, thereby improving fire point detection accuracy. Fire point location and classification results refer to the geographic coordinates of fire points and their type labels, including active fire points, ember areas, and other categories, obtained through temperature threshold segmentation and gradient analysis.

[0073] In an embodiment of the present application, first, an embedded real-time processing chip is used to load a temperature gradient threshold segmentation algorithm to process the thermal radiation distribution data frame by frame to identify abnormal areas where the temperature exceeds a preset threshold; secondly, a morphological filtering algorithm is used to perform expansion and corrosion operations on the abnormal areas to eliminate isolated small interference areas; then, the boundaries of the fire points are extracted based on the connected domain analysis, and the fire points are classified based on the temperature gradient direction; finally, a classification result containing the geographic coordinates and type labels of the fire points is generated and spatially bound to the three-dimensional point cloud data.

[0074] S14. Based on the target fire spread path, fire point location and classification results, dynamic scheduling instructions are sent to the fire-fighting drone cluster through a low-latency communication link to adjust the location, fire-fighting dosage and coverage of the fire-fighting drones in the fire-fighting drone cluster.

[0075] Among them, the low-latency communication link refers to a communication channel that uses high-speed wireless transmission technology, which is used to transmit fire data to the fire-fighting drone cluster in real time to ensure the timeliness of command issuance. The fire-fighting drone cluster refers to a collaborative operation system composed of multiple drones, equipped with a fire extinguishing agent tank and a dynamic scheduling module, which supports the adjustment of the delivery strategy according to instructions. Dynamic scheduling instructions refer to optimized instructions generated based on the fire spread path and fire point classification, including drone delivery location, fire extinguishing dosage distribution and coverage range parameters. The fire extinguishing dosage refers to the fire extinguishing agent capacity carried by a single drone, and the priority is dynamically assigned according to the type of fire point. The coverage range refers to the effective area of ​​​​fire extinguishing agent spraying, and the coverage radius is controlled by adjusting the drone's flight altitude and nozzle angle.

[0076] In an embodiment of the present application, first, the target fire spread path, fire point location and classification results are transmitted to the fire-fighting drone cluster through a low-latency communication link; secondly, the spatial distance, fire extinguishing agent inventory and coverage capacity of each drone are calculated based on a dynamic scheduling algorithm to optimize the delivery priority; then, the fire extinguishing dosage is allocated according to the fire point classification results, and the coverage radius of each machine is adjusted; finally, the fire-fighting drone cluster is controlled to deliver the fire extinguishing agent according to instructions to form a spatially matched coverage area.

[0077] In an embodiment of the present application, first, the detection drone adjusts its flight altitude and heading angle according to the fire environment, the lidar generates three-dimensional point cloud data of the wildfire area, and the thermal infrared sensor synchronously collects thermal radiation distribution data. Secondly, the terrain elevation features and temperature gradient direction are integrated to screen areas sensitive to fire spread, and the target fire spread path is generated by combining wind speed and vegetation type. Subsequently, the embedded chip processes the thermal radiation data in real time, identifies active fire points and removes interference areas, and generates fire point classification results. Finally, the command is sent to the fire-fighting drone cluster through low-latency communication, dynamically allocates the fire extinguishing dose and adjusts the coverage range to achieve precise fire extinguishing agent delivery.

[0078] By executing S11 to S14, the embodiment of the present application accurately reconstructs the vertical structure of the fire scene and captures the dynamic changes of the thermal field through the fusion of multi-angle three-dimensional point cloud and thermal radiation data, thereby improving the accuracy of fire spread path prediction; embedded real-time processing technology realizes rapid identification of fire points and interference filtering, reducing the misjudgment rate; dynamic scheduling algorithm optimizes the fire extinguishing agent delivery strategy, combined with low-latency communication to form a closed-loop control from perception, decision-making to execution, greatly improving the efficiency of wildfire fighting and resource utilization.

[0079] In a possible embodiment, S12, combining the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path, includes:

[0080] Step 121: Extracting wildfire terrain elevation features from the three-dimensional point cloud data.

[0081] Among them, the wildfire terrain elevation characteristics refer to a set of terrain parameters extracted from three-dimensional point cloud data, including elevation value sequences, slope change rate sets, and slope angle sets, which are used to quantify the dynamic impact of terrain on fire spread.

[0082] In an embodiment of the present application, first, the three-dimensional point cloud data is preprocessed using a point cloud density optimization algorithm to remove noise and fill in missing areas; secondly, the elevation value sequence of the wildfire area is extracted using elevation interpolation technology to generate a continuous terrain elevation surface; then, based on the curvature calculation model, a slope change rate set is extracted from the elevation surface, and the terrain normal vector is projected onto the horizontal plane using the aspect angle calculation formula to obtain an aspect angle set; finally, the elevation value sequence, slope change rate set, and aspect angle set are integrated to form a wildfire terrain elevation feature dataset, thereby improving the accuracy of terrain modeling.

[0083] Step 122: extract temperature field gradient change characteristics from the thermal radiation distribution data.

[0084] Among them, the temperature field gradient change characteristics refer to the dynamic parameters of the thermal field based on the analysis of thermal radiation distribution data, including the temperature gradient direction vector, gradient intensity distribution and temperature extreme area boundary, which are used to identify the fire spread trend.

[0085] In an embodiment of the present application, first, the thermal radiation distribution data is Gaussian filtered to eliminate sensor noise; second, the gradient operator is used to calculate the gradient change of the temperature field in the horizontal direction, and the temperature gradient direction vector and intensity distribution are extracted; then, the region growing algorithm is used to segment the temperature extreme value region from the temperature field, and its boundary contour is extracted by edge detection technology; finally, the temperature gradient change rate is quantified in combination with time series analysis, and the temperature field gradient change feature matrix is ​​generated, thereby improving the sensitivity of temperature anomaly detection.

[0086] Step 123: Based on the wildfire terrain elevation characteristics and temperature field gradient change characteristics, combined with historical meteorological data of the wildfire area, generate a target fire spread path.

[0087] Among them, the historical meteorological data of the wildfire area refers to the set of historical environmental parameters of the area where the fire is located, including wind speed vectors, such as those affecting the direction and speed of fire spread, humidity and vegetation type, which are used to correct the dynamic boundary conditions of the fire spread model.

[0088] In an embodiment of the present application, first, the set of slope angles in the wildfire terrain elevation features is divided into azimuth partitions, and the covariance coefficient of the slope angle mean and the temperature gradient intensity in each partition is calculated, and the covariance coefficient is used as the fire spread sensitive area; secondly, the thermal convection intensity at the front of the fire line is calculated based on the distribution of terrain elevation change rate, and the wind speed vector in the historical meteorological data is multiplied by the flammability coefficient of the vegetation type to correct the diffusion rate; then, a plurality of candidate fire spread paths are generated through Monte Carlo simulation, and the probability weight of each path is determined by the spatial coupling degree of the sensitive area distribution and the topological structure characteristics; finally, the path with a probability of more than 80% is selected as the target fire spread path, thereby reducing the prediction error.

[0089] In an embodiment of the present application, first, the three-dimensional point cloud data collected by the detection drone is interpolated to generate an elevation value sequence, and the slope change rate and slope angle are extracted to improve the accuracy of terrain modeling; secondly, the thermal infrared data is subjected to gradient analysis to extract the temperature gradient direction vector, and the region growing algorithm is used to segment three temperature extreme value areas; then, the slope angle partition and the two partitions with a covariance coefficient of the temperature gradient intensity greater than 0.7 are marked as sensitive areas, and the fire line diffusion rate correction value is calculated based on the historical wind speed of 8m / s, humidity of 20% and coniferous forest vegetation type; finally, five candidate paths are generated through Monte Carlo simulation, and the target path with a probability of more than 80% is screened to guide the fire-fighting drone cluster to accurately deliver the fire extinguishing agent and shorten the firefighting response time.

[0090] By executing steps 121 to 123, the embodiment of the present application constructs a multi-dimensional fire spread prediction model by integrating terrain elevation features, temperature field gradient changes and historical meteorological data, and performs coupled analysis of terrain and thermal fields to improve the accuracy of path prediction; based on the screening of sensitive areas based on covariance coefficients and Monte Carlo probability simulation, the robustness of fire direction prediction in complex environments is enhanced and the misjudgment rate is reduced; the diffusion rate and thermal convection intensity calculation are dynamically corrected to improve the spatial matching of fire extinguishing resource scheduling.

[0091] In one possible embodiment, the wildfire terrain elevation features include: an elevation value sequence, a slope change rate set, and a slope aspect angle set; the temperature field gradient change features include: the boundary of the temperature extreme value region, the temperature gradient direction vector, and the temperature gradient intensity distribution; and the historical meteorological data includes wind speed vector, humidity, and vegetation type. Step 123: Based on the wildfire terrain elevation features and the temperature field gradient change features, combined with the historical meteorological data of the wildfire area, generate a target fire spread path, including:

[0092] Step a1: perform differential calculation on the elevation value sequence to obtain the terrain elevation change rate distribution.

[0093] Difference calculations involve subtracting adjacent data points in a terrain elevation sequence to extract local changes in elevation. Terrain elevation change rate distributions are two-dimensional spatial matrices derived through difference calculations, reflecting the trend of terrain elevation changes within a unit distance.

[0094] In this embodiment, a continuous sequence of terrain elevation values ​​is first differentiated. This involves calculating the elevation change by taking the difference between adjacent elevation values, thereby generating a distribution of terrain elevation change rates. This process can be performed point by point using a first-order difference algorithm, converting the elevation sequence into a rate of change matrix. This matrix is ​​used to characterize the local steepness and spatial continuity of the terrain.

[0095] Step a2: performing nonlinear mapping on the slope change rate and the temperature gradient direction vector in the slope change rate set to generate a dynamic impact intensity parameter representing the terrain's effect on the temperature field gradient change characteristics.

[0096] The slope change rate is a parameter calculated from the directional derivative of the elevation change rate, representing the severity of slope changes with spatial position. The temperature gradient direction vector is a spatial vector field that describes the direction of maximum temperature change in the temperature field and contains magnitude and orientation information. Nonlinear mapping operations are the process of mapping two sets of heterogeneous data into a common feature space using radial basis functions or neural networks. The dynamic impact intensity parameter is a scalar field used to quantify the degree of dynamic influence of terrain on the temperature gradient and is used to correct the direction of fire spread.

[0097] In this embodiment, the slope change rate set and the temperature gradient direction vector are first input into a nonlinear mapping model. A kernel function is used to correlate the two in a high-dimensional space. The mapping parameters are then optimized using gradient descent to generate a dynamic impact intensity parameter. This parameter, normalized using an activation function, quantifies the nonlinear strengthening or weakening effect of terrain slope on temperature field gradient changes.

[0098] Step a3: extracting topological structural features from the boundary of the temperature extreme value region, where the topological structural features include the number of connected domains and the boundary curvature radius.

[0099] Among them, the topological structure characteristics include the number of connected domains and the set of boundary curvature radius, which reflects the morphological complexity of the temperature extreme area.

[0100] In this example, a morphological closing operation is first performed on the boundaries of the temperature extremes to eliminate noise. A region growing algorithm is then used to extract connected domains. The number of connected domains is counted, and the radius of curvature of each boundary point is calculated. The boundary curve is then fitted using the least squares method, ultimately outputting a description matrix that contains the topological distribution of the connected domains and the curvature variation characteristics.

[0101] Step a4: partition the slope angle set by azimuth angle, calculate the covariance coefficient between the slope angle and the temperature gradient intensity distribution in each partition, and define the partition with a covariance coefficient greater than a preset coefficient threshold as a fire spread sensitive area.

[0102] Among them, the azimuth partition divides the 360-degree azimuth into several sectors, which is used to calculate the distribution law of slope angle. The covariance coefficient is an indicator of the degree of linear correlation between two sets of spatial data, which is calculated by the ratio of covariance to standard deviation. The calculation formula is: ,in, For the The set of aspect angles within the azimuth partitions, For the The corresponding temperature gradient intensity distribution within each azimuth partition is: is the slope angle Temperature gradient strength The covariance of For the The standard deviation of the aspect angle within each partition, For the The standard deviation of the temperature gradient strength within a partition.

[0103] In this embodiment, the aspect angle set is first divided into eight sector-shaped zones by azimuth. The covariance between the aspect angle distribution and the temperature gradient intensity distribution within each zone is calculated, and then divided by the standard deviation to obtain the covariance coefficient. Zones with strong positive correlations are screened using a preset coefficient threshold (e.g., 0.8 or 0.85). These zones are then overlaid with geographic raster data and marked as areas susceptible to fire spread.

[0104] Step a5: Calculate the thermal convection intensity of the fire front under different terrain conditions based on the distribution of terrain elevation change rate, calculate the diffusion rate of the fire front based on vegetation type, humidity, and wind speed vector, and correct the direction of the fire front based on the dynamic impact intensity parameter to generate multiple candidate fire diffusion paths.

[0105] Among them, the thermal convection intensity refers to the heat transfer rate parameter calculated based on the air flow acceleration effect caused by terrain undulation.

[0106] In this application's example, a heat convection intensity model is first established based on the distribution of terrain elevation change rates. The heat flow acceleration effect under different terrain conditions is calculated using a set of fluid dynamics equations. Next, the spatial interpolation results of vegetation type combustion index, humidity attenuation factor, and wind speed vector are integrated to construct a diffusion rate field. Finally, the dynamic impact intensity parameter is used as a directional correction weight to iteratively generate multiple candidate fire spread paths.

[0107] Step a6: Based on the spatial coupling relationship between the distribution range of the fire spread sensitive area and the topological structure characteristics, a probability simulation is performed on each candidate fire spread path to screen out the target fire spread path that meets the preset spread conditions.

[0108] In this application example, a coupling evaluation function between fire spread-sensitive regions and topological structural features was first constructed. A Monte Carlo simulation was then used to probabilistically sample candidate paths. The spatiotemporal overlap of each path with high-curvature boundaries and multiply connected regions was then calculated. A Bayesian network was then used to select target fire spread paths that met pre-defined spread criteria.

[0109] In an embodiment of the present application, a differential calculation is first performed on the elevation data of a certain mountain area to generate a terrain elevation change rate distribution map, and then the slope change rate and the temperature gradient direction vector provided by the meteorological station are input into a pre-trained nonlinear mapping model to output a dynamic impact intensity parameter field. Then, the boundaries of the temperature extreme value areas identified by the forest fire monitoring system are extracted, and the number of connected domains and the radius of curvature are analyzed to form topological features. After azimuth partitioning, the covariance coefficient of the slope angle and the temperature gradient in each sector is calculated, and three fire spread sensitive areas are delineated. Based on the elevation change rate and meteorological data, the thermal convection intensity and diffusion rate field are calculated, and five candidate paths are generated in combination with dynamic parameter correction. Finally, the coupling probability of the candidate paths with sensitive areas and topological features is evaluated through Monte Carlo simulation, and two target paths that meet the diffusion conditions are screened out for emergency evacuation planning.

[0110] By executing steps a1 through a6, this embodiment of the application achieves high-precision prediction of fire spread paths through multi-source data fusion and multi-stage modeling. By analyzing the dynamic impact of terrain elevation changes on thermodynamic fields, this method effectively captures the evolution of fire lines in complex environments. By coupling topological features with sensitive area assessments, the physical interpretability of path screening is enhanced. The resulting prediction model combines spatial resolution with computational efficiency, providing reliable decision-making support for disaster prevention and control.

[0111] In a possible embodiment, step a2, performing nonlinear mapping on the slope change rate in the slope change rate set and the temperature gradient direction vector to generate a dynamic impact intensity parameter characterizing the terrain's influence on the temperature field gradient change characteristics, includes:

[0112] Step a21: normalize all the slope change rates in the slope change rate set to obtain a standardized slope change rate vector.

[0113] Among them, the standardized slope change rate vector refers to the slope change rate data set after normalization processing, which reflects the relative change degree of terrain slope and is used to eliminate dimensional differences and enhance data comparability.

[0114] In an embodiment of the present application, first, all values ​​in the slope change rate set are normalized to their maximum and minimum values, and the minimum value of each slope change rate is subtracted from the minimum value and then divided by the difference between the maximum and minimum values ​​to obtain a standardized slope change rate vector; secondly, the numerical range is constrained to between 0 and 1 through linear scaling to eliminate dimensional differences; finally, a standardized vector is generated for subsequent directional consistency analysis to improve data comparability.

[0115] Step a22: Decompose the temperature gradient direction vector into a horizontal component and a vertical component, and calculate the first cosine similarity between the horizontal component and the normalized slope change rate vector and the second cosine similarity between the vertical component and the normalized slope change rate vector.

[0116] The horizontal component refers to the projection of the temperature gradient vector onto the terrain plane and is used to analyze the correlation between horizontal heat transfer and terrain slope. The vertical component refers to the component of the temperature gradient vector perpendicular to the terrain plane and reflects the vertical convection effect of heat. The first cosine similarity measures the similarity of the cosine angle between the normalized slope change rate vector and the horizontal component and is used to quantify the directional consistency between the two. The second cosine similarity measures the similarity of the cosine angle between the normalized slope change rate vector and the vertical component and is used to quantify the strength of the correlation between vertical thermal convection and terrain.

[0117] In an embodiment of the present application, first, the temperature gradient direction vector is decomposed into a horizontal component and a vertical component. The horizontal component is calculated by projecting it onto the terrain plane, and the vertical component is obtained by subtracting the horizontal component modulus from the vector modulus. Secondly, the first cosine similarity of the normalized slope change rate vector and the horizontal component is calculated, that is, the dot product of the two vectors divided by the product of their respective moduli. Subsequently, the second cosine similarity of the normalized slope change rate vector and the vertical component is calculated. Finally, the spatial correlation between the terrain and the temperature gradient is quantified by a two-dimensional similarity metric.

[0118] Step a23: perform exponential weighting on the first cosine similarity and the second cosine similarity to obtain a first weighted value and a second weighted value, and use the sum of the first weighted value and the second weighted value as a directional consistency parameter.

[0119] Exponential weighting applies an exponential weighting function to the cosine similarity, amplifying the contribution of key components by adjusting the coefficients. The first weighted value refers to the exponentially weighted result of the first cosine similarity and reflects the strength of the synergistic effect between horizontal terrain and temperature gradient. The second weighted value refers to the exponentially weighted result of the second cosine similarity and reflects the strength of the coupling effect between vertical terrain and thermal convection. The directional consistency parameter is a composite parameter of the horizontal and vertical weighted values, representing the degree of spatial synergy between terrain slope and temperature gradient direction.

[0120] In an embodiment of the present application, first, an exponential weight is applied to the first cosine similarity, and the weight coefficient is calculated based on the contribution ratio of the horizontal component to obtain a first weighted value; secondly, an independent exponential weight is applied to the second cosine similarity, and the weight coefficient is related to the thermal convection effect of the vertical component to obtain a second weighted value; then, the first weighted value and the second weighted value are added to generate a directional consistency parameter; finally, this parameter comprehensively reflects the degree of spatial coordination between the terrain slope and the temperature gradient direction.

[0121] Step a24: normalize the directional consistency parameter to obtain a dynamic influence intensity parameter.

[0122] Among them, in the embodiment of the present application, first, the directional consistency parameter is input into the S function for nonlinear normalization processing, and the parameter range is mapped to between 0 and 1; secondly, the interference of extreme values ​​is eliminated through dynamic threshold adjustment; finally, the dynamic impact intensity parameter is output to quantify the influence weight of the terrain on the direction of fire spread.

[0123] In an embodiment of the present application, first, the slope change rate data collected by the drone is normalized to generate a standardized vector, and the temperature gradient direction is decomposed into horizontal and vertical components; secondly, the cosine similarity between the standardized vector and the horizontal component is calculated, and combined with exponential weighting to generate a first weighted value, and the similarity with the vertical component is calculated and weighted to obtain a second weighted value; then, the two are added together to generate a direction consistency parameter, which is normalized to a dynamic influence intensity parameter using an S function; finally, the parameter is input into the fire spread model to correct the fire line direction prediction, guiding the fire-fighting drone cluster to optimize the coverage range.

[0124] By executing steps a21 to a24, the embodiment of the present application enhances the quantitative characterization capability of the coupling effect between terrain and thermal field through two-dimensional similarity analysis of the normalized slope change rate and the temperature gradient component; integrates the horizontal and vertical component contributions based on the exponentially weighted directional consistency parameter to improve the physical consistency of the fire spread direction prediction model; and dynamically normalizes the processing to reduce parameter sensitivity and ensure the robustness of the model in different fire environments.

[0125] In one possible embodiment, step a5, calculating the thermal convection intensity of the fire front under different terrain conditions based on the distribution of terrain elevation change rate, calculating the diffusion rate of the fire front based on vegetation type, humidity, and wind speed vector, and correcting the direction of the fire front based on the dynamic impact intensity parameter to generate multiple candidate fire spread paths, includes:

[0126] Step a51: Calculate the thermal convection intensity correction coefficient for a first location and a second location within the volcanic region using formulas with different parameter values. The first location is a location in the terrain elevation change rate distribution corresponding to a terrain elevation change rate greater than a preset elevation change rate threshold, and the second location is a location in the terrain elevation change rate distribution corresponding to a terrain elevation change rate less than or equal to the preset elevation change rate threshold.

[0127] Among them, the thermal convection intensity correction coefficient refers to the adjustment parameter calculated in sections according to the difference in terrain elevation change rate, which is used to quantify the local enhancement or weakening effect of terrain undulation on thermal convection intensity.

[0128] In an embodiment of the present application, first, a first position and a second position are screened out according to the distribution of terrain elevation change rate, wherein the first position corresponds to an area where the terrain elevation change rate is greater than a preset change rate threshold, and the second position corresponds to an area where the terrain elevation change rate is less than or equal to the threshold; secondly, a piecewise function is used to calculate the thermal convection intensity correction coefficients of the two types of areas, respectively. The correction coefficient formula for the first position is the terrain elevation change rate multiplied by the temperature gradient intensity, and the correction coefficient formula for the second position is the terrain elevation change rate divided by the temperature gradient intensity; finally, a set of correction coefficients for the two types of areas is generated to quantify the differentiated effects of terrain on thermal convection.

[0129] Step a52: The thermal convection intensity at each location is obtained by multiplying the thermal convection intensity correction coefficient at each location, the intensity reference value, and the correction coefficient corresponding to the vegetation type as the thermal convection intensity at each location.

[0130] The intensity reference value refers to a predefined thermal convection intensity reference, which is used to generate an actual intensity value in combination with a correction factor.

[0131] In an embodiment of the present application, first, the thermal convection intensity correction coefficient, the predefined intensity reference value, and the correction coefficient corresponding to the vegetation type at each location are obtained; secondly, the three are multiplied point by point according to the location to obtain the thermal convection intensity at each location; finally, a continuous thermal convection intensity distribution map is generated through spatial interpolation to reflect the local enhancement or weakening effect of heat transfer at the front of the fire line.

[0132] Step a53: Calculate the diffusion rate using a preset diffusion rate calculation formula according to the vegetation coefficient, rate reference value, humidity, and wind speed vector corresponding to the vegetation type.

[0133] The vegetation coefficient refers to the quantitative parameter of flammability corresponding to different vegetation types, which is obtained from experimental data or historical fire statistics. The rate reference value refers to the reference benchmark of fire spread rate under conditions without external interference, which is used to calculate the actual rate in combination with environmental parameters. Preset spread rate The calculation formula is: ,in, is the rate reference value, is the vegetation coefficient, For humidity, is the humidity attenuation coefficient, is the wind speed vector, is the wind speed intensification coefficient, It is the angle between the wind direction and the direction of the obstacle.

[0134] In an embodiment of the present application, first, the corresponding vegetation coefficient is found according to the vegetation type, and the preset diffusion rate calculation formula is input in combination with the rate reference value, the inverse of the humidity and the modulus of the wind speed vector; secondly, the diffusion rate calculation formula is defined as the vegetation coefficient multiplied by the rate reference value, and then multiplied by the product of the inverse of the humidity and the wind speed vector; finally, the spatial distribution of the diffusion rate is generated to quantify the spread speed of the fire under different vegetation and meteorological conditions.

[0135] Step a54: Calculate the extension length of the path based on the heat convection intensity and diffusion rate at each position.

[0136] The extension length of the path refers to the spatial distance that the fire spreads in a specific direction per unit time, which is calculated jointly by the thermal convection intensity and the diffusion rate.

[0137] In the embodiment of the present application, first, the thermal convection intensity and diffusion rate at each location are superimposed according to the location to calculate the path extension length; second, the extension length formula is the thermal convection intensity multiplied by the diffusion rate and then divided by the reference time constant; finally, a path extension length matrix is ​​generated to evaluate the spatial expansion potential of the fire in unit time.

[0138] Step a55: Determine a direction correction coefficient corresponding to the dynamic impact intensity parameter, and correct the direction of the fire front based on the direction correction coefficient to obtain a corrected direction.

[0139] Among them, the corrected direction refers to the spreading direction vector of the fire front after adjustment by the dynamic impact intensity parameter, reflecting the coupling effect of the terrain and the thermal field.

[0140] In an embodiment of the present application, first, the direction correction coefficient of the current fire front is determined based on the dynamic impact intensity parameter and the preset direction correction coefficient mapping table; secondly, the original direction vector of the fire front is multiplied by the correction coefficient to obtain the corrected direction vector; finally, the physical rationality of the direction correction is ensured through vector normalization processing.

[0141] Step a56: Initialize the path starting point to the current coordinates of the fire front. Based on the corrected direction and the extension length of the path, combined with the path screening conditions, generate multiple candidate fire spread paths. The path screening conditions include: the overlapping area between the path and the fire spread sensitive area is greater than the preset overlapping area threshold, the diffusion rate of the path is greater than the preset rate threshold, and the diffusion duration is greater than the preset duration threshold.

[0142] Among them, the initialization path starting point refers to the initial spatial position generated by the current coordinates of the fire front as the candidate path. The path screening condition refers to the set of constraints that the candidate path must meet, including the threshold limits of the overlapping area with the sensitive area, the diffusion rate and the diffusion duration. The overlapping area of ​​the fire spread sensitive area refers to the spatial intersection area of ​​the candidate path and the fire spread sensitive area, which is used to evaluate the rationality of the path. The preset rate threshold refers to the minimum allowable value of the fire spread rate. Paths below this value will be eliminated. The preset duration threshold refers to the lower limit of the fire spread duration, which is used to filter out short-term invalid paths.

[0143] In an embodiment of the present application, first, the current coordinates of the fire front are used as the starting point of the path, and an initial path segment is generated based on the corrected direction vector and the path extension length; second, all paths that meet the screening conditions are traversed, and the screening conditions include that the overlapping area of ​​the path and the fire spread sensitive area exceeds a preset threshold, the diffusion rate is greater than a preset threshold, and the diffusion time is greater than a preset threshold; finally, multiple candidate fire spread path sets are output for further screening by probability simulation.

[0144] In an embodiment of the present application, first, the first position and the second position are divided according to the terrain elevation change rate threshold, and the thermal convection intensity correction coefficient is calculated respectively; secondly, the correction coefficient, the intensity reference value and the coniferous forest vegetation coefficient are multiplied to generate the thermal convection intensity distribution; then, the diffusion rate is calculated based on the vegetation coefficient, the rate reference value, the humidity and the wind speed, and the thermal convection intensity is superimposed to obtain the path extension length; then, the fire line direction is corrected according to the dynamic impact intensity parameter, and the initial path is generated with the fire line front coordinate as the starting point; finally, the paths with overlapping areas exceeding the threshold, diffusion rates and durations exceeding the standards are screened, and multiple candidate fire spread paths are output to guide the drone cluster for precise interception.

[0145] By executing steps a51 to a56, the embodiment of the present application calculates the thermal convection intensity correction coefficient in segments to accurately quantify the differentiated effects of terrain undulation on fire thermal convection; combines the vegetation coefficient and the diffusion rate calculation model of meteorological parameters to enhance the environmental adaptability of the fire spread speed prediction; based on the dynamic direction correction and multi-condition path screening mechanism, improves the physical rationality and spatial matching of the candidate path generation, and provides a reliable basis for the dynamic scheduling of fire extinguishing resources.

[0146] In one possible embodiment, S13 utilizes an embedded real-time processing chip deployed in a detection drone to perform real-time processing on thermal radiation distribution data, identify abnormal temperature areas, remove interference areas, and generate fire location and classification results, including:

[0147] Step 131: Use the embedded real-time processing chip deployed in the detection drone to perform median filtering on the thermal radiation distribution data to eliminate environmental noise interference and generate a noise-reduced temperature field matrix.

[0148] Median filtering is an image denoising method that uses a sliding window to replace the current pixel value with the median of neighboring pixels. This method is used to eliminate salt and pepper noise in thermal radiation data. The denoised temperature field matrix is ​​a two-dimensional temperature data matrix processed through median filtering, preserving the true temperature distribution in the fire area while suppressing environmental interference.

[0149] In this embodiment, the embedded real-time processing chip deployed in the detection drone first performs median filtering on the collected thermal radiation distribution data. This involves traversing each pixel using a sliding window, sorting the radiation values ​​of all pixels in its neighborhood by size, and then taking the median value to replace the original pixel value to eliminate impulse noise and interference from isolated outliers. After processing, a de-noised temperature field matrix is ​​generated, which preserves the temperature distribution characteristics of the fire area and suppresses environmental noise.

[0150] Step 132: Process the temperature field matrix after noise reduction to determine the target fire point area.

[0151] Among them, the target fire point area is a continuous high-temperature area extracted by threshold segmentation and morphological processing, which represents the spatial range of the potential fire point location.

[0152] In this embodiment, adaptive threshold segmentation is first performed on the denoised temperature field matrix. The segmentation threshold is dynamically adjusted by calculating the mean and variance of the global temperature distribution, screening out continuous regions with temperature values ​​above the threshold. Morphological dilation and erosion operations are then performed on the candidate regions to eliminate small holes. Adjacent regions are then merged to output the spatial extent of the target fire point region.

[0153] Step 133: extract a temperature distribution feature vector from the target fire point region. The temperature distribution feature vector includes the regional average temperature, the maximum temperature gradient, the temperature distribution standard deviation, and the ratio of the regional contour perimeter to the area.

[0154] Among them, the temperature distribution feature vector includes a four-dimensional vector of regional average temperature, maximum temperature gradient, standard deviation of temperature distribution and ratio of contour perimeter to area, which is used to quantify the thermodynamic and morphological characteristics of the fire point. The regional average temperature refers to the arithmetic mean of the temperature values ​​of all pixels in the target fire point area, reflecting the overall thermal radiation intensity. The maximum temperature gradient refers to the extreme value of the temperature change rate calculated by the gradient operator, which characterizes the degree of temperature mutation at the edge of the fire point. The standard deviation of the temperature distribution refers to the degree of dispersion of the temperature values ​​in the fire point area from the average temperature, reflecting the uniformity of the temperature distribution. The regional contour perimeter refers to the total length of the closed curve of the fire point area boundary, which is used to calculate the shape complexity in combination with the area.

[0155] In this embodiment, temperature data within the target fire area is first collected pixel by pixel, and the average temperature of the area is calculated as an indicator of overall heat intensity. Next, the temperature gradient field is calculated using the Sobel operator, and the maximum gradient is extracted to characterize the degree of local temperature change. The standard deviation of the temperature values ​​within the area is then calculated to reflect the discreteness of the temperature distribution. The area outline is then extracted using an edge detection algorithm, and the shape complexity ratio is calculated by dividing the outline perimeter by the area. Finally, these four features are integrated into a temperature distribution feature vector.

[0156] Step 134: Use the K-means clustering algorithm to perform dynamic threshold division on the temperature distribution feature vector, and generate fire point type labels and confidence parameters based on the Euclidean distance matching between the cluster center and the historical fire point samples. The fire point type labels are used as classification results. The fire point type labels include surface fire, hidden fire and re-ignition fire.

[0157] The K-means clustering algorithm is an unsupervised classification algorithm based on a distance metric. It groups feature vectors by iteratively optimizing cluster centers. The cluster center is the mean vector of the feature vectors for each category in the K-means algorithm and represents the typical characteristics of that category. The Euclidean distance of historical fire samples refers to the straight-line distance in feature space between the current fire feature vector and the historical sample, and is used for category matching. The fire type label refers to the fire category classified based on the results of cluster matching with historical data. It includes three burning states: surface fire, smoldering fire, and rekindling fire. The confidence parameter is a probability indicator calculated based on the inverse of the Euclidean distance, reflecting the confidence level of the classification result. Surface fire refers to a fire with visible and intense flames, corresponding to high temperature, high gradient, and high standard deviation. Smoldering fire refers to a fire with no open flame but continuous internal smoldering, characterized by medium temperature and low gradient. Rekindling fire refers to a fire that reignites in an area where a fire has been extinguished, characterized by small area and high contour complexity.

[0158] In this embodiment, the K-means clustering algorithm is first used to perform unsupervised classification on the temperature distribution feature vectors. The feature vectors are divided into several categories by iteratively updating the cluster centers until convergence. Next, the Euclidean distance between each cluster center and the feature vector of the historical fire point samples is calculated. The historical sample category with the smallest distance is selected as the type label for the current fire point, and a confidence parameter is generated based on the inverse of the distance. The final output includes classification results for three categories: surface fire, smoldering fire, and rekindled fire.

[0159] Step 135: Based on the spatial coordinates of the target fire point area in the three-dimensional point cloud data, calculate the three-dimensional geographic coordinates of the fire point center in the UAV navigation coordinate system, and use the three-dimensional geographic coordinates as the fire point position.

[0160] In this embodiment, the spatial coordinates of the fire center are first calculated using a centroid algorithm based on the boundary coordinates of the target fire area in the three-dimensional point cloud data. Then, the fire center coordinates are converted into three-dimensional geographic coordinates representing latitude, longitude, and altitude, using the georeferencing parameters of the drone navigation coordinate system. Finally, the converted coordinates are output as the fire location to the disaster emergency command system.

[0161] In the embodiment of the present application, the thermal radiation data of a certain forest area is first collected by a thermal imaging sensor carried by a drone, and the embedded chip is used to perform median filtering to generate a temperature field matrix after noise reduction. Subsequently, adaptive threshold segmentation and morphological optimization are used to extract three target fire areas, and the temperature distribution feature vectors of each area are calculated. Based on the K-means clustering algorithm, the feature vectors are divided into two categories. Combined with the Euclidean distance matching of the historical fire database, it is determined that two of them are surface fires and one is a hidden fire, and the confidence parameters are output. Finally, the center coordinates of the fire point are calculated based on the three-dimensional point cloud data, converted into longitude and latitude and altitude information, and transmitted to the fire command platform in real time for dynamic fire monitoring and fire fighting path planning.

[0162] By executing steps 131 through 135, this embodiment of the present application achieves efficient and accurate fire point detection and classification through multimodal data fusion and hierarchical feature analysis. Combining real-time noise reduction processing with dynamic threshold segmentation effectively improves the accuracy of target fire point positioning. Multidimensional feature extraction and cluster matching enhance the physical interpretability of fire point type identification. The resulting three-dimensional geographic coordinates and type labels provide a reliable basis for decision-making in fire assessment and resource scheduling.

[0163] In a possible embodiment, step 132, processing the noise-reduced temperature field matrix to determine the target fire point area, includes:

[0164] Step b1: Binarize and segment the denoised temperature field matrix based on a preset temperature threshold to extract candidate temperature anomaly regions, and merge adjacent candidate temperature anomaly regions using a region growing algorithm to generate an initial fire point region set.

[0165] Among them, binary segmentation refers to the process of dividing the temperature field matrix into high-temperature anomaly points and background areas based on the temperature threshold, with the high-temperature points assigned a value of 1 and the background assigned a value of 0. The candidate temperature anomaly area refers to the discrete high-temperature pixel set formed after binary segmentation, which characterizes the spatial distribution of potential fire point locations. The region growing algorithm refers to a spatial clustering algorithm that iteratively merges adjacent similar pixels to generate continuous closed areas. The initial fire point area set refers to the set of fire point candidate areas merged by the region growing algorithm, which contains complete candidate targets that have not been screened.

[0166] In the embodiment of the present application, the denoised temperature field matrix is ​​first binarized and segmented according to a preset temperature threshold. Pixels above the threshold are marked as high-temperature anomalies, and pixels below the threshold are classified as background areas, generating candidate temperature anomaly regions consisting of discrete high-temperature pixels. Secondly, a region growing algorithm is used, with each high-temperature pixel as a seed point. Region expansion is performed based on the temperature similarity of adjacent pixels, and spatially continuous candidate temperature anomaly regions are merged to ultimately form an initial fire point region set, which contains multiple independent and closed fire point candidate regions.

[0167] Step b2: Delete the fire point regions whose areas are smaller than a preset area threshold in the initial fire point region set to obtain an intermediate fire point region set.

[0168] The intermediate fire point area set refers to the fire point area set after area threshold filtering, which is the intermediate result after eliminating small area noise.

[0169] In this embodiment, each fire point region in the initial set of fire point regions is first traversed, the number of pixels covered is calculated, and the value is converted into an actual area. Next, the area of ​​each region is compared with a preset area threshold, and regions with an area smaller than the threshold are deleted to eliminate isolated noise interference. Fire point regions that meet the area criteria are then retained to generate an intermediate fire point region set, which serves as the basic dataset for subsequent refined screening.

[0170] Step b3: Filter out the fire point areas whose reflectivity is lower than a preset reflectivity threshold from the set of intermediate fire point areas to obtain the target fire point areas.

[0171] Among them, the target fire point area refers to the final fire point area after the reflectivity threshold screening, excluding the high-confidence result of non-fire point interference.

[0172] In this embodiment, multispectral image data corresponding to a set of intermediate fire regions is first acquired, and the mean reflectivity of each fire region is extracted. Next, the mean reflectivity is compared with a preset reflectivity threshold to filter out regions with reflectivity below the threshold, eliminating false positives caused by high-reflectivity objects. Finally, a set of target fire regions is output, ensuring that the result contains only true fires that meet combustion characteristics.

[0173] In the embodiment of the present application, the satellite thermal infrared image is firstly binarized and segmented, and pixels with temperatures higher than a preset threshold are extracted to form candidate temperature anomaly regions. Adjacent high-temperature pixels are merged through a region growing algorithm to generate a set of six initial fire point regions. The area of ​​each region is then calculated and the three regions with a temperature lower than the preset threshold are deleted, retaining the three intermediate fire point regions. Finally, combined with the visible light image reflectivity data, two regions with a reflectivity lower than the threshold are screened out, and interference from a high reflectivity water area is eliminated. The two target fire point regions are output to the forest fire monitoring system for real-time dynamic tracking of the fire situation and emergency response.

[0174] By executing steps b1 to b3, the embodiment of the present application effectively improves the accuracy of fire point detection through a multi-level screening mechanism, the binary segmentation and region growing algorithm ensures the integrity of the fire point area, the area threshold filtering eliminates noise interference, and the reflectivity screening enhances the target recognition specificity, ultimately achieving accurate positioning of real fire points and false alarm suppression in complex environments.

[0175] Figure 2 A schematic diagram of the structure of a wildfire monitoring system based on lidar and thermal infrared is provided in an embodiment of the present application, as shown in FIG. Figure 2 As shown, the system includes:

[0176] The acquisition module 21 is used to control the flight altitude and heading angle of the detection drone so that the laser radar and thermal infrared sensor deployed on the detection drone can collect three-dimensional point cloud data of the vertical section of the wildfire area and thermal radiation distribution data of the wildfire area from multiple angles.

[0177] The generation module 22 is used to combine the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path.

[0178] The identification module 23 is used to use the embedded real-time processing chip deployed in the detection drone to process the thermal radiation distribution data in real time, identify temperature abnormality areas and remove interference areas, and generate fire point locations and classification results.

[0179] The adjustment module 24 is used to send dynamic scheduling instructions to the fire-fighting drone cluster through a low-latency communication link based on the target fire spread path, fire point location and classification results, so as to adjust the location, fire extinguishing dosage and coverage range of the fire-fighting drones in the fire-fighting drone cluster.

[0180] Figure 2 The aforementioned mountain fire monitoring system based on laser radar and thermal infrared can perform Figure 1 The implementation principles and technical effects of the laser radar and thermal infrared-based wildfire monitoring method described in the illustrated embodiment are not further elaborated. The specific manner in which each module and unit performs operations in the laser radar and thermal infrared-based wildfire monitoring system described in the above embodiment has been described in detail in the relevant embodiments of the method and will not be elaborated on here.

[0181] In one possible design, Figure 2 A mountain fire monitoring system based on laser radar and thermal infrared in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32 .

[0182] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0183] The processing component 32 is used to: control the flight altitude and heading angle of the detection drone, so that the laser radar and thermal infrared sensor deployed on the detection drone respectively collect three-dimensional point cloud data of the vertical section of the wildfire area and the thermal radiation distribution data of the wildfire area at multiple angles; combine the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path; use the embedded real-time processing chip deployed in the detection drone to process the thermal radiation distribution data in real time, identify temperature abnormality areas and remove interference areas, and generate fire point locations and classification results; based on the target fire spread path, the fire point location and the classification results, send dynamic scheduling instructions to the fire-fighting drone cluster through a low-latency communication link to adjust the location, fire extinguishing dosage and coverage range of the fire-fighting drones in the fire-fighting drone cluster.

[0184] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0185] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0186] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0187] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0188] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0189] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0190] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment is a method for monitoring wildfires based on lidar and thermal infrared.

[0191] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0192] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0193] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0194] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A forest fire monitoring method based on laser radar and thermal infrared, characterized in that: include: By controlling the flight altitude and heading angle of the detection drone, the laser radar and thermal infrared sensor deployed on the detection drone collect three-dimensional point cloud data of the vertical section of the wildfire area and thermal radiation distribution data of the wildfire area from multiple angles. Combining the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path; Using the embedded real-time processing chip deployed in the detection drone, the thermal radiation distribution data is processed in real time to identify abnormal temperature areas and remove interference areas, generating fire point locations and classification results; Based on the target fire spread path, the fire point location, and the classification result, dynamic scheduling instructions are sent to the fire-fighting drone cluster via a low-latency communication link to adjust the location, amount, and coverage of the fire-fighting agent released by each fire-fighting drone in the fire-fighting drone cluster; The step of combining the three-dimensional point cloud data with the heat radiation distribution data to generate a target fire spread path includes: extracting wildfire terrain elevation features from the three-dimensional point cloud data; extracting temperature field gradient variation characteristics from the thermal radiation distribution data; Based on the wildfire terrain elevation characteristics and the temperature field gradient change characteristics, combined with the historical meteorological data of the wildfire area, a target fire spread path is generated; The wildfire terrain elevation features include: elevation value sequence, slope change rate set and slope angle set; the temperature field gradient change features include: temperature extreme area boundary, temperature gradient direction vector and temperature gradient intensity distribution; historical meteorological data include wind speed vector, humidity and vegetation type; The target fire spread path is generated based on the wildfire terrain elevation characteristics and the temperature field gradient change characteristics, combined with historical meteorological data of the wildfire area, including: Performing differential calculation on the elevation value sequence to obtain a distribution of terrain elevation change rates; Performing nonlinear mapping on the slope change rates and the temperature gradient direction vectors in the slope change rate set to generate a dynamic impact intensity parameter representing the characteristics of the terrain on the temperature field gradient change; Extracting topological structure features from the boundary of the temperature extreme value region, wherein the topological structure features include the number of connected domains and the boundary curvature radius; The slope angle set is partitioned into azimuth zones, a covariance coefficient between the slope angle and the temperature gradient intensity distribution in each zone is calculated, and a zone having a covariance coefficient greater than a preset coefficient threshold is defined as a fire spread sensitive zone; Calculating the thermal convection intensity of the fire front under different terrain conditions based on the terrain elevation change rate distribution, calculating the diffusion rate of the fire front based on the vegetation type, humidity, and wind speed vector, and correcting the direction of the fire front based on the dynamic impact intensity parameter to generate multiple candidate fire spread paths; According to the spatial coupling relationship between the distribution range of the fire spread sensitive area and the topological structure characteristics, a probability simulation is performed on each candidate fire spread path to screen out a target fire spread path that meets the preset diffusion conditions.

2. The method according to claim 1, characterized in that The nonlinear mapping of the slope change rate in the slope change rate set to the temperature gradient direction vector to generate a dynamic impact intensity parameter characterizing the terrain's influence on the temperature field gradient change characteristics includes: Normalizing all the slope change rates in the slope change rate set to obtain a standardized slope change rate vector; Decompose the temperature gradient direction vector into a horizontal component and a vertical component, and calculate the first cosine similarity between the horizontal component and the normalized slope change rate vector and the second cosine similarity between the vertical component and the normalized slope change rate vector respectively; performing exponential weighting on the first cosine similarity and the second cosine similarity to obtain a first weighted value and a second weighted value, and taking the sum of the first weighted value and the second weighted value as a directional consistency parameter; The directional consistency parameter is normalized to obtain the dynamic impact intensity parameter.

3. The method according to claim 1, characterized in that The method calculates the heat convection intensity of the fire front under different terrain conditions based on the terrain elevation change rate distribution, calculates the diffusion rate of the fire front based on the vegetation type, humidity and wind speed vector, and corrects the direction of the fire front based on the dynamic impact intensity parameter to generate multiple candidate fire spread paths, including: Calculating heat convection intensity correction coefficients for a first location and a second location within the volcanic region according to formulas with different parameter values, respectively; the first location being a location corresponding to a terrain elevation change rate greater than a preset elevation change rate threshold in the terrain elevation change rate distribution; and the second location being a location corresponding to a terrain elevation change rate less than or equal to the preset elevation change rate threshold in the terrain elevation change rate distribution; The thermal convection intensity at each location is obtained by multiplying the thermal convection intensity correction coefficient, the intensity reference value, and the correction coefficient corresponding to the vegetation type. The diffusion rate is calculated using a preset diffusion rate calculation formula according to the vegetation coefficient, rate reference value, humidity and wind speed vector corresponding to the vegetation type; Calculate the extension length of the path based on the heat convection intensity and diffusion rate at each location; determining a direction correction coefficient corresponding to the dynamic impact intensity parameter, and correcting the direction of the firing line based on the direction correction coefficient to obtain a corrected direction; The starting point of the path is initialized to the current coordinates of the fire front. Based on the corrected direction and the extension length of the path, combined with path screening conditions, multiple candidate fire spread paths are generated. The path screening conditions include: the overlapping area between the path and the fire spread sensitive area is greater than a preset overlapping area threshold, the diffusion rate of the path is greater than a preset rate threshold, and the diffusion duration is greater than a preset duration threshold.

4. The method according to claim 1, wherein The embedded real-time processing chip deployed in the detection drone is used to process the thermal radiation distribution data in real time, identify temperature anomaly areas and remove interference areas, and generate fire point locations and classification results, including: Using the embedded real-time processing chip deployed in the detection drone, the thermal radiation distribution data is subjected to median filtering to eliminate environmental noise interference and generate a noise-reduced temperature field matrix; Processing the noise-reduced temperature field matrix to determine a target fire point area; Extracting a temperature distribution feature vector from the target fire point region, wherein the temperature distribution feature vector includes the average temperature of the region, the maximum temperature gradient, the standard deviation of the temperature distribution, and the ratio of the perimeter of the region to the area; Using the K-means clustering algorithm to perform dynamic threshold division on the temperature distribution feature vector, and generating fire point type labels and confidence parameters based on the Euclidean distance matching between the cluster center and the historical fire point samples, the fire point type labels are used as the classification results. The fire point type labels include surface fire, smoldering fire and re-ignition fire; Based on the spatial coordinates of the target fire point area corresponding to the three-dimensional point cloud data, the three-dimensional geographic coordinates of the fire point center in the UAV navigation coordinate system are calculated, and the three-dimensional geographic coordinates are used as the fire point position.

5. The method according to claim 4, characterized in that The processing of the temperature field matrix after noise reduction to determine the target fire point area includes: Binarizing and segmenting the denoised temperature field matrix based on a preset temperature threshold to extract candidate temperature anomaly regions, and merging adjacent candidate temperature anomaly regions using a region growing algorithm to generate an initial fire point region set; Deleting fire point regions whose areas are smaller than a preset area threshold from the initial fire point region set to obtain an intermediate fire point region set; Fire point areas with reflectivity lower than a preset reflectivity threshold are screened out from the set of intermediate fire point areas to obtain target fire point areas.

6. A forest fire monitoring system based on laser radar and thermal infrared, characterized in that: include: The acquisition module is used to control the flight altitude and heading angle of the detection drone, so that the laser radar and thermal infrared sensor deployed on the detection drone can respectively collect three-dimensional point cloud data of the vertical section of the wildfire area and thermal radiation distribution data of the wildfire area from multiple angles; A generation module, configured to combine the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path; An identification module is used to process the thermal radiation distribution data in real time using an embedded real-time processing chip deployed in the detection drone, identify temperature anomaly areas and remove interference areas, and generate fire point locations and classification results; An adjustment module is configured to send dynamic scheduling instructions to the fire-fighting drone cluster via a low-latency communication link based on the target fire spread path, the fire point location, and the classification result, so as to adjust the location, amount, and coverage of the fire-fighting agent released by each fire-fighting drone in the fire-fighting drone cluster; The step of combining the three-dimensional point cloud data with the heat radiation distribution data to generate a target fire spread path includes: extracting wildfire terrain elevation features from the three-dimensional point cloud data; extracting temperature field gradient variation characteristics from the thermal radiation distribution data; Based on the wildfire terrain elevation characteristics and the temperature field gradient change characteristics, combined with the historical meteorological data of the wildfire area, a target fire spread path is generated; The wildfire terrain elevation features include: elevation value sequence, slope change rate set and slope angle set; the temperature field gradient change features include: temperature extreme area boundary, temperature gradient direction vector and temperature gradient intensity distribution; historical meteorological data include wind speed vector, humidity and vegetation type; The target fire spread path is generated based on the wildfire terrain elevation characteristics and the temperature field gradient change characteristics, combined with historical meteorological data of the wildfire area, including: Performing differential calculation on the elevation value sequence to obtain a distribution of terrain elevation change rates; Performing nonlinear mapping on the slope change rates and the temperature gradient direction vectors in the slope change rate set to generate a dynamic impact intensity parameter representing the characteristics of the terrain on the temperature field gradient change; Extracting topological structure features from the boundary of the temperature extreme value region, wherein the topological structure features include the number of connected domains and the boundary curvature radius; The slope angle set is partitioned into azimuth zones, a covariance coefficient between the slope angle and the temperature gradient intensity distribution in each zone is calculated, and a zone having a covariance coefficient greater than a preset coefficient threshold is defined as a fire spread sensitive zone; Calculating the thermal convection intensity of the fire front under different terrain conditions based on the terrain elevation change rate distribution, calculating the diffusion rate of the fire front based on the vegetation type, humidity, and wind speed vector, and correcting the direction of the fire front based on the dynamic impact intensity parameter to generate multiple candidate fire spread paths; According to the spatial coupling relationship between the distribution range of the fire spread sensitive area and the topological structure characteristics, a probability simulation is performed on each candidate fire spread path to screen out a target fire spread path that meets the preset diffusion conditions.

7. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a wildfire monitoring method based on lidar and thermal infrared as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, a wildfire monitoring method based on lidar and thermal infrared is implemented as described in any one of claims 1 to 5.

Images