Mountain fire risk assessment and holder monitoring method and system, medium and product

Abstract

The invention discloses a mountain fire risk assessment and holder monitoring method and system, a medium and a product, and relates to the field of mountain fire monitoring. According to the implementation of the method, firstly, a mountain fire monitoring system quantifies various risk factors (a historical fire factor, a weather influence factor, a vegetation flammable factor and a human activity factor), and calculates a fire probability index corresponding to each region of a mountain forest; meanwhile, the mountain fire monitoring system establishes a fire spreading model and evaluates fire influence indexes corresponding to all areas of the mountain forest; and finally, the forest fire monitoring system calculates a comprehensive risk value based on the fire probability index and the fire influence index, and then formulates a differentiated holder monitoring strategy. According to the differentiated monitoring scheme based on risk assessment, the problem of unreasonable distribution of monitoring resources caused by a traditional fixed time interval polling mode is avoided, it is ensured that a high-risk area is fully monitored and covered, waste of the monitoring resources in a low-risk area is avoided, and the accuracy and efficiency of mountain fire monitoring are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of mountain fire monitoring, and particularly relates to a mountain fire risk assessment and pan-tilt monitoring method and system, medium and product. BACKGROUND

[0002] In order to discover and deal with mountain fire hazards in time, a mountain fire monitoring system needs to be established to monitor mountains and forests in real time. In the process of mountain fire monitoring, how to reasonably allocate and use monitoring resources and improve monitoring efficiency has become an important problem to be solved.

[0003] At present, the mountain fire monitoring system mainly adopts a round-patrol mode for monitoring. Specifically, the mountain fire monitoring system scans different areas in turn by aiming the monitoring device at the areas according to a pre-set time interval. At the same time, the mountain fire monitoring system also performs smoke and fire identification analysis on the collected image data, and issues an alarm signal when a suspected fire is found, to notify relevant personnel to verify and deal with it.

[0004] However, due to the vast area of mountains and forests and the complex and changeable terrain, the round-patrol mode with a fixed time interval is likely to cause unreasonable allocation of monitoring resources, which may result in insufficient monitoring coverage of high-risk areas, and waste of monitoring resources on low-risk areas. SUMMARY

[0005] The present application provides a mountain fire risk assessment and pan-tilt monitoring method and system, medium and product, for improving the rationality of monitoring resource allocation.

[0006] In a first aspect, the present application provides a forest fire risk assessment and gimbal monitoring method applied to a forest fire monitoring system. The method comprises: obtaining historical fire data, meteorological environment data, vegetation distribution data, terrain facility data and human activity data corresponding to each region of the forest; determining historical fire factors according to the historical fire data, determining weather influence factors according to the meteorological environment data, determining vegetation flammability factors according to the vegetation distribution data, and determining human activity factors according to the human activity data. The historical fire factors are used to quantify the frequency of historical fires, the weather influence factors are used to quantify the probability of fire occurrence under current meteorological environment conditions, the vegetation flammability factors are used to quantify the probability of vegetation spontaneous combustion or ignition, and the human activity factors are used to represent the probability of human-induced fire; the historical fire factors, the weather influence factors, the vegetation flammability factors and the human activity factors are weighted and summed to obtain a fire probability index corresponding to each region of the forest; a fire spread model corresponding to each region of the forest is established by combining the meteorological environment data, the vegetation distribution data and the terrain facility data, to determine a fire impact index corresponding to each region of the forest. The fire spread model is used to represent the dynamic propagation behavior of the simulated fire, and the fire impact index is used to quantify the harmful influence range of the fire; the fire probability index and the fire impact index are substituted into a preset forest fire risk assessment function to obtain a comprehensive risk value corresponding to each region of the forest; based on the comprehensive risk value and a preset gimbal monitoring strategy, a gimbal monitoring strategy corresponding to each region of the forest is determined. The preset gimbal monitoring strategy includes different gimbal monitoring strategies corresponding to different comprehensive risk values, and the gimbal monitoring strategy includes gimbal control parameters.

[0007] By adopting the above technical solution, the forest fire monitoring system comprehensively considers multi-dimensional data such as historical fire data, meteorological environment data, vegetation distribution data, terrain facility data and human activity data to establish a comprehensive forest fire risk assessment system. First, the forest fire monitoring system quantifies various risk factors (historical fire factors, weather influence factors, vegetation flammability factors and human activity factors) to calculate a fire probability index corresponding to each region of the forest. At the same time, the forest fire monitoring system establishes a fire spread model to evaluate a fire impact index corresponding to each region of the forest. Finally, the forest fire monitoring system calculates a comprehensive risk value based on the fire probability index and the fire impact index, and then formulates a differentiated gimbal monitoring strategy. This differentiated monitoring scheme based on risk assessment avoids the problem of unreasonable allocation of monitoring resources caused by the traditional fixed time interval round patrol mode, which not only ensures that high-risk areas are fully monitored and covered, but also avoids wasting monitoring resources in low-risk areas, significantly improving the accuracy and efficiency of forest fire monitoring.

[0008] In some embodiments of the first aspect, in some embodiments, the historical fire factor is determined according to historical fire data, the weather influence factor is determined according to meteorological environment data, the vegetation flammability factor is determined according to vegetation distribution data, and the human activity factor is determined according to human activity data, specifically comprising: according to the historical fire data, the historical fire occurrence times corresponding to each region of the mountain forest are counted, and the historical fire occurrence times are normalized to obtain the historical fire factors corresponding to each region of the mountain forest; the meteorological environment data is input into a fire weather prediction model to obtain the weather influence factors corresponding to each region of the mountain forest, and the fire weather prediction model is used to construct a mapping relationship between meteorological environment conditions and fire occurrence probability; according to a preset mountain forest plant flammability table, the flammability of each type of plant is determined, and the vegetation flammability factors corresponding to each region of the mountain forest are determined in combination with the vegetation distribution data corresponding to each region of the mountain forest; according to the human activity data, the population density corresponding to each region of the mountain forest is determined to calculate the human activity factors corresponding to each region of the mountain forest.

[0009] By adopting the above technical solutions, the historical fire factor, the weather influence factor, the vegetation flammability factor, and the human activity factor are scientifically quantified, the randomness of subjective evaluation is avoided, and a reliable data basis is provided for subsequent determination of the fire probability index.

[0010] In some embodiments of the first aspect, in some embodiments, a fire spread model corresponding to each region of the mountain forest is established in combination with the meteorological environment data, the vegetation distribution data, and the terrain facility data to determine the fire impact index corresponding to each region of the mountain forest, specifically comprising: extracting the wind speed, the wind direction, the temperature, and the humidity in the meteorological environment data as the kinetic parameters of the fire spread; extracting the vegetation type, the vegetation density, the vegetation water content, and the combustible load in the vegetation distribution data as the fuel parameters of the fire spread; extracting the elevation, the slope, the slope direction, the ridge line, and the valley line in the terrain facility data as the terrain parameters of the fire spread; based on the kinetic parameters, the fuel parameters, and the terrain parameters, the fire spread model corresponding to each region of the mountain forest is constructed; each region of the mountain forest is set as a virtual fire source point, and the fire spread model corresponding to each region of the mountain forest is run to simulate the spread process of the forest fire starting from the virtual fire source point; the maximum speed, the average speed, the coverage area, and the spread time of the forest fire spread are calculated to determine the fire impact index corresponding to each region of the mountain forest.

[0011] By employing the aforementioned technical solutions, the wildfire monitoring system constructs a sophisticated fire spread model to assess the fire impact index. The fire spread model comprehensively considers dynamic parameters such as wind speed, wind direction, temperature, and humidity; fuel parameters such as vegetation type, vegetation density, vegetation water content, and combustible material load; and topographic parameters such as elevation, slope, aspect, ridgeline, and valleyline. The wildfire monitoring system simulates fire spread by setting various areas of the forest as virtual fire sources, predicting the extent of the damage caused by fires in different areas of the forest. This impact assessment method based on a physical model is more scientific and accurate than simple empirical judgment, providing a more reliable basis for determining the fire impact index.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, the maximum speed, average speed, coverage area, and spread time of wildfire spread are calculated to determine the fire impact index corresponding to each area of ​​the forest. Specifically, this includes: periodically iterating the fire spread model at a fixed time step within a preset simulation period to obtain the spatial distribution data of wildfire spread at each fixed time step; based on the spatial distribution data, extracting the fire front position at each fixed time step and calculating the displacement vector of the fire front position between adjacent time steps; dividing the magnitude of the displacement vector by the fixed time step to obtain the instantaneous spread speed in a preset direction; selecting the maximum value from the instantaneous spread speed as the maximum speed; calculating the weighted average of the instantaneous spread speed as the average speed; statistically analyzing the total area covered by fire within the preset simulation period as the coverage area; recording the time required for the coverage area to reach a preset area threshold as the spread time; and using a nonlinear weighted fusion algorithm based on the maximum speed, average speed, coverage area, and spread time to calculate the fire impact index corresponding to each area of ​​the forest.

[0013] By adopting the above technical solution, the wildfire monitoring system calculates the maximum speed, average speed, coverage area, and spread time of wildfires based on a fire spread model. A nonlinear weighted fusion algorithm is then used to synthesize these indicators to determine the fire impact index. This refined analysis method based on time-series data can comprehensively and accurately assess the dynamic diffusion characteristics of fires, providing a more reliable quantitative basis for determining the fire impact index and effectively improving its scientific rigor and accuracy.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the preset wildfire risk assessment function is: R = αP + β +γ(PI); where R represents the comprehensive risk value, P represents the fire probability index, I represents the fire impact index, α, β and γ represent the weighting coefficients, n represents the exponential parameter of the fire impact index and n>1; the weighting coefficients satisfy β>α>0, γ>0 and α+β+γ=1.

[0015] By adopting the above technical solution, this nonlinear weighted preset wildfire risk assessment function not only considers the independent influence of the fire probability index and the fire impact index, but also reflects the coupling effect between the two by multiplying the cross-term fire probability index and the fire impact index. Setting the index parameter n>1 strengthens the weight of the fire impact index, reflecting the risk management principle that "high-impact, low-probability" events should receive more attention. This allows for a more accurate quantification of the comprehensive risk level in different regions, providing a reliable basis for formulating differentiated monitoring strategies.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after determining the corresponding PTZ monitoring strategy for each area of ​​the forest based on the comprehensive risk value and the preset PTZ monitoring strategy, the method further includes: receiving real-time image data corresponding to each area of ​​the forest; inputting the real-time image data into a smoke and fire recognition model to detect whether there is smoke and fire in the real-time image data; if there is smoke and fire, calling a fire spread model to predict the spread direction, spread speed and impact range of the smoke and fire in real time based on the geographical location information corresponding to the real-time image data; determining the optimal scheduling plan for fire prevention and control resources based on the prediction results, the optimal scheduling plan including the dispatch route of firefighters, the configuration plan of firefighting equipment and the evacuation plan; and sending the optimal scheduling plan to the terminal device to notify relevant personnel to carry out firefighting and evacuation work.

[0017] By adopting the above technical solutions, the forest fire monitoring system has established a complete fire monitoring and emergency response mechanism based on a comprehensive risk assessment of various areas of the forest, which significantly improves the timeliness and scientific nature of fire prevention and control, and can minimize the losses caused by fires.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, after sending the optimal scheduling plan to the terminal device to notify relevant personnel to carry out fire extinguishing and evacuation work, the method further includes: monitoring the wildfire spread status and fire extinguishing progress; if the wildfire spread status is inconsistent with the prediction results, updating the fire spread model based on the latest monitoring data to obtain a corrected fire spread prediction result; and adjusting the optimal scheduling plan based on the fire spread prediction result and / or fire extinguishing progress.

[0019] By adopting the above technical solutions, the wildfire monitoring system can monitor the spread of wildfires and the progress of firefighting in real time, and continuously optimize emergency response plans. This closed-loop feedback control mechanism significantly improves the adaptability of the wildfire monitoring system to complex fire situations, ensures that emergency response measures always meet actual needs, and greatly enhances the effectiveness of fire prevention and control.

[0020] In a second aspect, embodiments of this application provide a wildfire monitoring system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions, and the one or more processors call the computer instructions to cause the wildfire monitoring system to perform the methods described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a wildfire monitoring system, cause the wildfire monitoring system to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a wildfire monitoring system, cause the wildfire monitoring system to perform the method described in the first aspect and any possible implementation thereof.

[0023] Understandably, the wildfire monitoring system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By adopting the above technical solution, the wildfire monitoring system comprehensively considers multi-dimensional data such as historical fire data, meteorological environment data, vegetation distribution data, terrain facility data, and human activity data to establish a comprehensive mountain fire risk assessment system. First, the system quantifies various risk factors (historical fire factors, weather impact factors, vegetation flammability factors, and human activity factors) and calculates the fire probability index for each area of ​​the forest. Simultaneously, the system establishes a fire spread model to assess the fire impact index for each area of ​​the forest. Finally, based on the fire probability index and the fire impact index, the system calculates a comprehensive risk value and then formulates differentiated PTZ monitoring strategies. This differentiated monitoring scheme based on risk assessment avoids the unreasonable allocation of monitoring resources caused by traditional fixed-time interval patrol methods. It ensures sufficient monitoring coverage of high-risk areas while avoiding wasting monitoring resources in low-risk areas, significantly improving the accuracy and efficiency of wildfire monitoring.

[0025] 2. By adopting the above technical solutions, the wildfire monitoring system constructs a refined fire spread model to assess the fire impact index. The fire spread model comprehensively considers dynamic parameters such as wind speed, wind direction, temperature, and humidity; fuel parameters such as vegetation type, vegetation density, vegetation water content, and combustible material load; and topographic parameters such as elevation, slope, aspect, ridgeline, and valleyline. The wildfire monitoring system sets various areas of the forest as virtual fire sources to simulate fire spread, predicting the hazard range of fires occurring in different areas of the forest. This impact assessment method based on a physical model is more scientific and accurate than simple empirical judgment, providing a more reliable basis for determining the fire impact index.

[0026] 3. By adopting the above technical solution, this nonlinear weighted preset wildfire risk assessment function not only considers the independent influence of the fire probability index and the fire impact index, but also reflects the coupling effect of the two by multiplying the cross-term fire probability index and the fire impact index. Setting the index parameter n>1 strengthens the weight of the fire impact index, reflecting the risk management principle that "high-impact, low-probability" events should receive more attention. This allows for a more accurate quantification of the comprehensive risk level in different regions, providing a reliable basis for formulating differentiated monitoring strategies. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating the wildfire risk assessment and PTZ monitoring method in this application embodiment; Figure 2 This is another flowchart illustrating the wildfire risk assessment and PTZ monitoring method in this application embodiment; Figure 3 This is a schematic diagram of the physical device structure of a wildfire monitoring system in the embodiments of this application. Detailed Implementation

[0028] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0030] The following describes the process of the method provided in this implementation. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating the wildfire risk assessment and PTZ monitoring method in this application.

[0031] S101. Obtain historical fire data, meteorological data, vegetation distribution data, terrain and facility data, and human activity data for each area of ​​the mountain forest; Among them, historical fire data represents recorded information on wildfire events that have occurred within a preset time period in the past, including the time, location, scale, duration, and extent of damage of the wildfires; meteorological environmental data refers to information on meteorological elements that affect the occurrence and spread of wildfires, such as temperature, humidity, wind speed, wind direction, and precipitation; vegetation distribution data is used to represent the spatial distribution characteristics of various plants in the mountains and forests, including vegetation type, coverage, and growth status; topographic facility data represents the spatial information of mountain topography and man-made facilities, including altitude, slope, aspect, roads, water systems, and firebreaks; and human activity data represents information on the characteristics of human activities in mountainous areas, including population density, distribution of tourist attractions, and agricultural activities.

[0032] Specifically, the wildfire monitoring system acquires the required data through multiple data collection channels: extracting historical fire data from historical databases; acquiring real-time meteorological and environmental data from meteorological stations or periodically from meteorological departments; acquiring vegetation distribution data based on remote sensing image interpretation or field surveys; extracting topographic facility data from geographic information systems or through surveying and mapping; and acquiring human activity data based on remote sensing images, tourist statistics, agricultural activity records, and other means.

[0033] S102. Determine historical fire factors based on historical fire data, weather influence factors based on meteorological environmental data, vegetation flammability factors based on vegetation distribution data, and human activity factors based on human activity data. Historical fire factors are used to quantify the frequency of historical fires, weather influence factors are used to quantify the probability of fires occurring under current meteorological environmental conditions, vegetation flammability factors are used to quantify the probability of vegetation spontaneous combustion or ignition, and human activity factors are used to represent the probability of human-caused fires. Specifically, firstly, the wildfire monitoring system constructs a Bayesian network describing the mechanism of fire occurrence, using historical fire events, meteorological conditions, vegetation characteristics, and human activities as nodes in the Bayesian network. Dependencies between nodes are determined through expert experience; for example, meteorological conditions affect vegetation water content, vegetation characteristics affect the probability of fire occurrence, and human activities directly affect fire occurrence. Secondly, the wildfire monitoring system learns the parameters of the Bayesian network based on historical data, obtaining a conditional probability table for each node. Finally, the wildfire monitoring system calculates historical fire factors, weather impact factors, vegetation flammability factors, and human activity factors through probabilistic inference: historical fire factors are obtained by marginalizing historical fire event nodes; weather impact factors are calculated using the conditional probability of meteorological condition nodes; vegetation flammability factors are inferred using the posterior probability of vegetation characteristic nodes; and human activity factors are determined using the probability distribution of human activity nodes.

[0034] For example: Suppose the Bayesian network calculation result for a certain area A in a mountain forest is as follows: Historical fire factor = 0.15 (the probability of a fire occurring in area A on average each year in the past was 15%). Weather impact factor = 0.3 (Current weather conditions in area A increase the probability of fire occurrence to 30%). Vegetation flammability factor = 0.25 (Vegetation conditions in area A result in a 25% probability of fire). Human activity factor = 0.2 (Human activity in Area A poses a 20% fire risk).

[0035] Optionally, under normal circumstances, determining historical fire factors based on historical fire data, weather influence factors based on meteorological data, vegetation flammability factors based on vegetation distribution data, and human activity factors based on human activity data can be achieved through the following methods, which are not limited here: Based on historical fire data, statistically analyze the number of historical fires in each area of ​​the forest, normalize the number of historical fires to obtain the historical fire factors for each area of ​​the forest; input meteorological data into the fire weather prediction model to obtain the weather influence factors for each area of ​​the forest, and use the fire weather prediction model to construct the mapping relationship between meteorological conditions and the probability of fire occurrence; determine the flammability of various plants based on a preset forest vegetation flammability table, and combine it with the vegetation distribution data for each area of ​​the forest to determine the vegetation flammability factors for each area of ​​the forest; determine the population density for each area of ​​the forest based on human activity data to calculate the human activity factors for each area of ​​the forest.

[0036] The construction of fire weather prediction models can be achieved in several ways, which are not limited here: A. Statistical analysis model (based on a large amount of historical data, analyzing the statistical relationship between fire occurrence and meteorological conditions): (1) Establish a basic indicator system: Temperature indicators: daily maximum temperature, average temperature; Humidity indicators: relative humidity, dew point difference; Wind force indicators: maximum wind speed, average wind speed; Precipitation indicators: precipitation amount, number of consecutive rainless days; (2) Determine the risk level of each indicator: Temperature: <20℃ (low risk), 20-30℃ (medium risk), >30℃ (high risk); Relative humidity: >70% (low risk), 40-70% (medium risk), <40% (high risk); Wind speed: <3m / s (low risk), 3-7m / s (medium risk), >7m / s (high risk); (3) Calculate the comprehensive index: Fire risk index = Σ (level value of each indicator × weight); (4) Establish the mapping relationship between the index and the probability: Fire risk probability = f (fire risk index); B. Machine Learning Model (Using machine learning algorithms to build a predictive model of weather conditions and fire risk probability): (1) Data preprocessing: Collect historical meteorological data and corresponding fire occurrence records, and process missing and outlier values; (2) Construct feature vectors: basic meteorological features (temperature, humidity, wind speed, etc.), time features (season, month, week, etc.), and combined features (temperature and humidity combination, dryness index, etc.). (3) Algorithm selection: Random Forest, Gradient Boosting Tree, Neural Network; (4) Model training and optimization: Train the model using the training set, adjust the parameters through cross-validation, and evaluate the performance on the validation set.

[0037] The following is a specific example to illustrate this: Suppose the data for a certain area A (1 square kilometer) in a mountain forest is as follows: 1. Calculation of historical fire factors: Historical fire records from the past 10 years show that there have been 8 fires in area A, and a total of 20 fires in the entire forest. The normalized historical fire factor for area A is calculated as 8 / 20 = 0.4. 2. Calculation of weather influencing factors: Current weather data: Temperature 32℃; Relative humidity 45%; Wind speed 5.5 m / s; 12 consecutive days without rain; Input the above current meteorological data into the fire weather prediction model for calculation, and output the weather impact factor for region A = 0.65; 3. Calculation of vegetation flammability factor: Vegetation distribution in Area A: 60% pine forest (flammability 0.8); 30% fir forest (flammability 0.6); 10% broadleaf forest (flammability 0.4). Weighted calculation: Flammability factor of vegetation in area A = 0.6 × 0.8 + 0.3 × 0.6 + 0.1 × 0.4 = 0.7; 4. Calculation of human activity factors: Human activity data for Area A: Average daily pedestrian flow: 200 people; Area: 1 square kilometer; Population density: 200 people / square kilometer; According to the preset standards: Low-risk density: <100 people / square kilometer (factor value 0.2); Medium-risk density: 100-300 people / square kilometer (factor value 0.5); High-risk density: >300 people / square kilometer (factor value 0.8); Area A has a medium risk density: human activity factor = 0.5; The final factor values ​​for region A are: Historical fire factor: 0.4; Weather impact factor: 0.65; Flammability factor of vegetation: 0.7; Human activity factor: 0.5.

[0038] S103. Weighted summation of historical fire factors, weather influence factors, vegetation flammability factors and human activity factors to obtain the fire probability index corresponding to each area of ​​the mountain forest. Weighted summation refers to a calculation method that assigns different weight coefficients to each factor according to its importance, and then sums all factors by multiplying them by their corresponding weight coefficients. The weight coefficients represent the relative importance of each factor in the fire probability assessment. The fire probability index is used to represent a quantitative indicator of the likelihood of a fire occurring after comprehensively considering various influencing factors.

[0039] Specifically, firstly, the wildfire monitoring system determines the weight coefficients of each factor using expert scoring or the analytic hierarchy process (AHP), ensuring that the sum of the weight coefficients is 1. Then, the system multiplies each factor by its corresponding weight coefficient. Finally, the system sums all the weighted factor values ​​to obtain a fire probability index reflecting the likelihood of fires occurring in different areas of the forest. To ensure the comparability of the assessment results, the system normalizes the final fire probability index, unifying its value range to the [0, 1] interval.

[0040] S104. By combining meteorological environmental data, vegetation distribution data, and terrain facility data, establish fire spread models for different areas of the mountain forest to determine the fire impact index for each area of ​​the mountain forest. The fire spread model is used to represent the dynamic propagation behavior of simulated fires, and the fire impact index is used to quantify the scope of the harm caused by a fire. Among them, the fire spread model refers to the mathematical and physical model used to simulate the spread of wildfires, which can predict the direction, speed and range of fire development; dynamic spread behavior represents the spatial diffusion characteristics of fire over time, including the speed of fire line advance and fire intensity; the fire impact index is used to represent the degree of harm that a fire may cause, and is obtained by assessing factors such as the range, speed and duration of fire spread.

[0041] Optionally, under normal circumstances, combining meteorological environmental data, vegetation distribution data, and topographic data to establish fire spread models for different areas of the forest to determine the corresponding fire impact index for each area can be achieved in the following ways, without limitation: extracting wind speed, wind direction, temperature, and humidity from meteorological environmental data as dynamic parameters for fire spread; extracting vegetation type, vegetation density, vegetation water content, and combustible material load from vegetation distribution data as fuel parameters for fire spread; extracting elevation, slope, aspect, ridgeline, and valleyline from topographic data as topographic parameters for fire spread; constructing fire spread models for different areas of the forest based on dynamic parameters, fuel parameters, and topographic parameters; setting each area of ​​the forest as a virtual fire source and running the corresponding fire spread models for each area to simulate the spread of wildfire from the virtual fire source; calculating the maximum speed, average speed, coverage area, and spread time of wildfire spread to determine the corresponding fire impact index for each area of ​​the forest.

[0042] Among them, dynamic parameters refer to meteorological elements that affect the spread of fire, such as wind speed, wind direction, temperature and humidity; fuel parameters are used to represent the characteristics of combustibles, including vegetation type, vegetation density, vegetation water content and combustible load; topographic parameters refer to topographic features that affect the spread of fire, including elevation, slope, aspect, ridgeline and valley line, etc.

[0043] Specifically, firstly, the wildfire monitoring system extracts the necessary parameters for modeling from meteorological environmental data, vegetation distribution data, and terrain data: dynamic parameters such as wind speed, wind direction, temperature, and humidity are extracted from meteorological environmental data; fuel parameters such as vegetation type, vegetation density, vegetation water content, and combustible load are extracted from vegetation distribution data; and terrain parameters such as elevation, slope, aspect, ridgeline, and valleyline are extracted from terrain data. Then, based on these parameters, the wildfire monitoring system constructs a fire spread model that considers heat transfer, mass conservation, and momentum conservation. The system sequentially sets each area as a virtual fire source point, runs the fire spread model to simulate the fire development process, calculates the maximum spread rate, average spread rate, coverage area, and spread time of the wildfire, and synthesizes these features using a nonlinear weighted fusion algorithm to obtain the fire impact index.

[0044] Optionally, under normal circumstances, the calculation of the maximum speed, average speed, coverage area, and spread time of wildfire to determine the fire impact index corresponding to each area of ​​the forest can be achieved in the following ways, without limitation: Within a preset simulation period, the fire spread model is periodically iterated at a fixed time step to obtain the spatial distribution data of wildfire spread at each fixed time step; based on the spatial distribution data, the position of the fire front at each fixed time step is extracted, and the displacement vector of the fire front position between adjacent time steps is calculated; the magnitude of the displacement vector is divided by the fixed time step to obtain the instantaneous spread speed in the preset direction; the maximum value is selected from the instantaneous spread speed as the maximum speed; the weighted average of the instantaneous spread speeds is calculated as the average speed; the total area covered by the fire within the preset simulation period is counted as the coverage area; the time required for the coverage area to reach the preset area threshold is recorded as the spread time; based on the maximum speed, average speed, coverage area, and spread time, a nonlinear weighted fusion algorithm is used to calculate the fire impact index corresponding to each area of ​​the forest.

[0045] Here is a specific example of calculating the fire impact index: Assuming the parameters for the mountainous / forested area are set as follows: Area scope: 1000m × 1000m mountainous terrain; Preset simulation period: 13:00-17:00 (4 hours); Fixed time step: 10 minutes; Preset area threshold: 20 hectares (200,000 square meters). Initial environmental conditions: Location of fire: 120°15'30"E, 30°25'15"N; Wind speed: 5 m / s; Wind direction: Northeast (45 degrees); Relative humidity: 30%; Temperature: 35℃; Terrain: 15-degree slope, southeast facing; Combustible materials: mainly pine forests; Example of calculation process: Spatial distribution of fire spread (fire boundary coordinates are recorded every 10 minutes to form a fire spread outline) For example: 0 minutes: A circular area with a radius of 50 meters around the ignition point; 10 minutes: Irregular ellipse, with the main axis extending in the northeast direction; 20 minutes: It will expand further to the northeast, then shift southeast due to the influence of the terrain; Fire front displacement calculation (taking the northeast direction as an example): 0-10 minutes: Displacement 18 meters; 10-20 minutes: Displacement 22 meters; 20-30 minutes: Displacement of 25 meters (the amount of displacement varies in different directions depending on wind direction and terrain); Spread rate calculation (taking the northeast direction as an example): 0-10 minutes: 0.03 m / s; 10-20 minutes: 0.037 m / s; 20-30 minutes: 0.042 m / s; Statistical results: Maximum speed: 0.08 m / s (occurred at 15 minutes and 2 hours); Average speed: 0.045 m / s; Coverage area: 30 minutes: 5 hectares; 1 hour: 8 hectares; 2 hours: 15 hectares; 3 hours: 25 hectares; 4 hours: 28 hectares; Total coverage area: 28 hectares; Time to spread: It took 180 minutes for the fire to cover 20 hectares (preset area threshold); Fire Impact Index Calculation (using a nonlinear weighted fusion method): Maximum speed weight: 0.3; Average speed weight: 0.2; Coverage area weight: 0.3; Spread time weight: 0.2; After normalization of each indicator: Maximum speed: 0.8; Average speed: 0.6; Coverage area: 0.7; Spread time: 0.65; The final fire impact index is 0.75 (range 0-1).

[0046] S105. Substitute the fire probability index and fire impact index into the preset wildfire risk assessment function to obtain the comprehensive risk value corresponding to each area of ​​the forest. The preset wildfire risk assessment function represents the mathematical model used to calculate the comprehensive risk value. The preset wildfire risk assessment function is R=αP+β. +γ(PI), where R represents the comprehensive risk value, P represents the fire probability index, I represents the fire impact index, α, β and γ represent the weighting coefficients, n represents the exponential parameter of the fire impact index and n>1; the weighting coefficients satisfy β>α>0, γ>0 and α+β+γ=1.

[0047] The comprehensive risk value is a quantitative risk indicator obtained by comprehensively considering the fire probability index and the fire impact index; the weighting coefficients α, β and γ represent the relative importance of each item in the risk assessment; the index parameter n is used to represent the degree of focus on high-impact events, and takes a value greater than 1; the cross term PI represents the coupling effect between the fire probability index and the fire impact index.

[0048] Specifically, firstly, the wildfire monitoring system determines the weighting coefficients α, β, and γ in the preset wildfire risk assessment function through expert evaluation or historical data analysis, satisfying the constraints that β>α>0, γ>0, and α+β+γ=1. Then, the system selects an appropriate value for the index parameter n (typically between 1.5 and 3). Finally, the system substitutes the fire probability index P and fire impact index I of each forest area into the preset wildfire risk assessment function to calculate the comprehensive risk value R, reflecting the overall fire risk level of the area. The system normalizes the comprehensive risk values ​​for all areas to facilitate risk level classification and comparison.

[0049] S106. Based on the comprehensive risk value and the preset PTZ monitoring strategy, determine the corresponding PTZ monitoring strategy for each area of ​​the mountain forest. The preset PTZ monitoring strategy includes different PTZ monitoring strategies corresponding to different comprehensive risk values, and the PTZ inspection strategy includes PTZ control parameters.

[0050] Among them, the PTZ monitoring strategy refers to the set of rules for controlling the rotation of the monitoring equipment, including monitoring frequency, dwell time, and scanning range; PTZ control parameters refer to the specific parameters required to implement the PTZ monitoring strategy, such as horizontal rotation angle, vertical pitch angle, rotation speed, and image magnification; monitoring frequency refers to the time interval for image acquisition of a specific area; dwell time refers to the duration for which the PTZ stays at a certain monitoring position; and scanning range is used to represent the angular range of the PTZ rotation.

[0051] Specifically, the risk level of each area in the forest is determined based on its comprehensive risk value range, and different pan-tilt-zoom (PTZ) monitoring strategies are set according to the risk level: for high-risk areas, a high-frequency, long-duration intensive monitoring strategy is adopted, with increased scanning angles to ensure no blind spots; for medium-risk areas, a medium-frequency routine monitoring strategy is used; and for low-risk areas, a lower-frequency general monitoring strategy is employed. Each PTZ monitoring strategy specifies detailed motion parameters for the PTZ, including the range of horizontal rotation angle, vertical pitch angle, rotation speed, acceleration, and image scaling. The forest fire monitoring system optimizes and adjusts the PTZ monitoring strategies based on the actual terrain conditions and performance parameters of the monitoring equipment in each area to ensure monitoring effectiveness. Simultaneously, the forest fire monitoring system also considers the coordination between adjacent PTZs to avoid overlapping monitoring ranges or blind spots.

[0052] By adopting the above technical solution, the wildfire monitoring system comprehensively considers multi-dimensional data, including historical fire data, meteorological data, vegetation distribution data, terrain data, and human activity data, to establish a comprehensive mountain fire risk assessment system. First, the system quantifies various risk factors (historical fire factors, weather impact factors, vegetation flammability factors, and human activity factors) and calculates the fire probability index for each area of ​​the forest. Simultaneously, it establishes a fire spread model to assess the fire impact index for each area. Finally, based on the fire probability index and the fire impact index, the system calculates a comprehensive risk value and then formulates differentiated PTZ monitoring strategies. This risk-assessment-based differentiated monitoring scheme avoids the unreasonable allocation of monitoring resources caused by traditional fixed-time interval patrol methods. It ensures sufficient monitoring coverage for high-risk areas while avoiding wasting monitoring resources in low-risk areas, significantly improving the accuracy and efficiency of wildfire monitoring.

[0053] The following provides a more detailed description of the process of the method provided in this implementation. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating the wildfire risk assessment and PTZ monitoring method in this application embodiment.

[0054] The following steps may or may not be performed after step S106; this is not limited here: S201, Receive real-time image data corresponding to various areas of the mountain forest.

[0055] Real-time image data refers to video or image information collected by monitoring equipment at the current moment, including visible light images and thermal imaging images.

[0056] Specifically, the wildfire monitoring system collects real-time image data of various areas in the forest through monitoring equipment. Each monitoring device includes a visible light camera and a thermal imaging camera. The visible light camera operates during the day, providing clear color images; the thermal imaging camera operates around the clock to capture temperature anomalies. The collected real-time image data is accompanied by metadata information such as timestamps, GPS coordinates, and camera orientation.

[0057] S202. Input the real-time image data into the smoke and fire recognition model to detect whether there is smoke and fire in the real-time image data.

[0058] Among them, the smoke and fire recognition model refers to the image recognition algorithm model developed based on deep learning technology, which is used to automatically detect smoke and fire features in images; smoke features represent the visual representation of smoke in an image, including color, shape, texture and motion state; fire features represent the typical features of fire in an image, including brightness, color, flicker frequency, etc.

[0059] Specifically, the wildfire monitoring system employs a two-stage identification strategy: First, real-time image data is rapidly pre-screened, using color and motion features to initially identify suspicious areas; then, a deep convolutional neural network is used for precise identification of these suspicious areas. For visible light images, the system focuses on extracting the gray-white and rising motion features of smoke, as well as the orange-red and flickering features of flames; for infrared images, it primarily analyzes the size, shape, and evolution characteristics of areas with abnormal temperatures. The smoke and fire identification model outputs detailed information such as the location, area, and confidence level of detected smoke and fire.

[0060] S203. If there is smoke or fire, the fire spread model is invoked to predict the direction, speed, and extent of the fire spread in real time, based on the geographical location information corresponding to the real-time image data.

[0061] Among them, the geographical location information represents the spatial coordinates and surrounding information of the location where the fire occurred; the direction of spread refers to the main path and trend of the fire spread, which is affected by factors such as wind direction and terrain; the spread speed represents the speed at which the fire front advances, including speed components in different directions; the scope of influence is used to represent the area that the fire may affect, which needs to take into account the topography and distribution of combustibles; and real-time prediction refers to the dynamic updating of fire development forecasts based on current monitoring data.

[0062] Specifically, firstly, the wildfire monitoring system determines the precise geographical location of the smoke and fire by spatial coordinate transformation based on the installation location and rotation angle of the monitoring equipment and the pixel coordinates of the smoke and fire in real-time image data. Then, the system calls upon a previously established fire spread model, inputting current meteorological environmental data (such as wind speed, wind direction, temperature, and humidity), terrain data (such as slope and aspect), and vegetation distribution data (such as combustible material type and density). The fire spread model uses numerical simulation to calculate the fire spread range at different future time points, predicting the main direction and speed of spread. The wildfire monitoring system generates visualized results of the fire spread prediction, including isochronous fireline maps, velocity vector field maps, and impact range prediction maps, providing a basis for emergency response decisions.

[0063] S204. Based on the prediction results, determine the optimal scheduling plan for fire prevention and control resources. The optimal scheduling plan includes the dispatch routes of firefighters, the configuration plan for firefighting equipment, and the evacuation plan.

[0064] Among them, fire prevention and control resources refer to all kinds of resources used for fighting fires and protecting personnel safety, including firefighters, firefighting equipment, and rescue vehicles; the optimal dispatch plan represents the best organizational plan for resource allocation and use under the current circumstances; the dispatch route for firefighters refers to the optimal travel path for firefighters and firefighting equipment from the assembly point to the fire scene, taking into account factors such as road conditions, terrain, and safety; the configuration plan for firefighting equipment is used to represent the layout and usage strategies of different types of firefighting equipment; the evacuation plan represents the specific arrangements for the safe evacuation of personnel from dangerous areas, including evacuation routes, assembly points, and refuge areas; the assembly point refers to the location where rescue forces are concentrated and on standby.

[0065] Specifically, firstly, the wildfire monitoring system divides the firefield into operational zones based on prediction results, categorizing the fire area into key firefighting zones, fire control zones, and warning zones. Then, based on the location information and capability characteristics of various rescue resources, combined with road network data from the geographic information system, the system employs a multi-objective optimization algorithm to calculate the optimal dispatch plan: For firefighters, the system assigns task areas to different teams based on their professional expertise and physical condition, and plans the shortest safe route from the assembly point to the task area; for firefighting equipment, the system determines the optimal deployment location and usage strategy based on equipment performance characteristics and firefield terrain conditions, such as the parking location of fire trucks and the setting of water pump relay points; for evacuation operations, the system analyzes personnel distribution and terrain conditions, plans multiple evacuation routes, identifies temporary refuge areas, and calculates the personnel capacity and evacuation time for each evacuation route.

[0066] S205. Send the optimal scheduling plan to the terminal equipment to notify relevant personnel to carry out firefighting and evacuation work.

[0067] Among them, terminal equipment refers to various mobile devices used to receive and display optimal dispatch plans, including handheld walkie-talkies, tablets, smartphones, etc.; relevant personnel refers to various personnel involved in firefighting and evacuation work, including firefighters, emergency rescue personnel, forest rangers, security personnel, etc.

[0068] Specifically, the wildfire monitoring system employs a multi-layered information dissemination strategy: First, it sends a complete dispatch plan, including overall information such as the fire situation, resource distribution, and operational deployment, to commanders at all levels through the emergency command system. Then, it pushes task execution instructions to the mobile terminals of frontline rescue personnel, including task descriptions with both text and images, real-time updated navigation routes, and key information such as the location of surrounding resources. Simultaneously, it sends information such as evacuation zone delineation, evacuation routes, and refuge locations to personnel responsible for evacuation. The wildfire monitoring system tracks the information reception status of terminal devices to ensure timely delivery of instructions. For communication blind spots, the system activates backup communication solutions, such as satellite communication or relay stations, to ensure the reliability of information transmission. The system also establishes an information feedback mechanism, enabling on-site personnel to promptly report task execution status and any problems encountered.

[0069] S206. Monitor the spread of wildfires and the progress of firefighting efforts.

[0070] Among them, the wildfire spread status indicates the real-time situation of fire development, including dynamic characteristics such as fire area, fire line location, and fire intensity; the fire fighting progress status refers to the execution status of fire fighting and rescue work, including personnel arrival, equipment usage effectiveness, and degree of fire control; the fire area indicates the area affected by the fire; the fire line location refers to the leading boundary of fire activity; and the fire intensity is used to indicate the intensity of combustion.

[0071] Specifically, the wildfire monitoring system employs a multi-source data fusion monitoring strategy: it continuously collects images of the fire scene through fixed monitoring equipment and analyzes changes in fire intensity in real time; it dispatches drones to conduct aerial reconnaissance along preset routes to obtain the overall fire situation; it receives text, images, and videos uploaded by on-site rescue personnel via mobile terminals; and it collects real-time data from meteorological stations to monitor changes in weather conditions. The system processes and analyzes this data in real time, calculating key indicators such as the rate of change of fire area, the speed of fire line advance, and the temperature distribution at the fire scene to assess the effectiveness of firefighting efforts. Simultaneously, the system tracks the location and operational status of rescue forces, compiles statistics on water consumption and equipment usage, and generates firefighting progress reports.

[0072] S207. If the wildfire spread status is inconsistent with the prediction results, the fire spread model shall be updated based on the latest monitoring data to obtain the corrected fire spread prediction results.

[0073] Among them, the inconsistency in prediction results refers to a significant deviation between the actual observed fire development and the fire development predicted by the fire spread model; the latest monitoring data refers to various real-time data collected at the current moment, including information such as fire situation, meteorological conditions, and combustion characteristics; updating the fire spread model refers to dynamically correcting the model parameters and prediction results based on actual observation data; the corrected fire spread prediction results are used to represent the new fire development prediction obtained after adjustment.

[0074] Specifically, firstly, the wildfire monitoring system calculates the deviation between the wildfire spread status and the predicted results, including deviations in fire area, spread speed, and direction. When the deviation exceeds a preset deviation threshold, the wildfire monitoring system initiates a model update process: collecting the latest monitoring data, including fire images, meteorological data, and surface temperature; analyzing possible causes of the prediction deviation, such as sudden changes in meteorological conditions or changes in fuel characteristics; adjusting model parameters based on the analysis results, such as wind speed influence coefficient, terrain influence factor, and fuel parameters; and recalculating the fire development prediction using the corrected fire spread model to generate updated fire spread prediction results, including the spatiotemporal evolution of the fire spread range, speed, and direction.

[0075] S208. Adjust the optimal dispatch plan based on the fire spread prediction results and / or fire extinguishing progress.

[0076] Among them, scheduling means optimizing and improving the original deployment based on the actual situation.

[0077] Specifically, the wildfire monitoring system adopts corresponding adjustment strategies based on different situations: if the fire develops beyond expectations, the system calculates the required reinforcements and determines the type, quantity, and deployment plan of the reinforcements; if the fire suppression effect in certain areas of the forest is not ideal, the system reassesses the terrain conditions and fire characteristics of these areas and adjusts the fire suppression strategies and equipment configuration; if new threats are discovered, the system promptly adjusts the warning area and evacuation plan. During the adjustment of the optimal dispatch plan, the system fully considers the allocation capacity and mobility of existing resources to ensure the operability of the adjusted plan. The adjusted plan is immediately distributed to all implementing units through the command system to ensure the continuity and coordination of the rescue work.

[0078] The wildfire monitoring system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference]. Figure 3 This is a schematic diagram of the physical device structure of a wildfire monitoring system in this application embodiment.

[0079] It should be noted that, Figure 3 The structure of the wildfire monitoring system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0080] like Figure 3 As shown, the wildfire monitoring system includes a CPU 301, which can perform various appropriate actions and processes based on a program stored in the read-only memory ROM 302 or a program loaded from the storage section 308 into the random access memory RAM 303, such as executing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An I / O interface 305 is also connected to the bus 304.

[0081] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0082] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by CPU 301, it performs the various functions defined in the present invention.

[0083] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0084] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0085] Specifically, the wildfire monitoring system in this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the wildfire risk assessment and PTZ monitoring method provided in the above embodiment.

[0086] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the wildfire monitoring system described in the above embodiments; or it may exist independently and not be assembled into the wildfire monitoring system. The storage medium carries one or more computer programs, which, when executed by a processor of the wildfire monitoring system, enable the wildfire monitoring system to implement the wildfire risk assessment and PTZ monitoring method provided in the above embodiments.

[0087] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0088] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0089] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for wildfire risk assessment and PTZ monitoring, characterized in that, The method, applied to a wildfire monitoring system, includes: Obtain historical fire data, meteorological data, vegetation distribution data, terrain and facility data, and human activity data for various areas of the mountains and forests; Based on the historical fire data, historical fire factors are determined; based on the meteorological environment data, weather influence factors are determined; based on the vegetation distribution data, vegetation flammability factors are determined; and based on the human activity data, human activity factors are determined. The historical fire factors are used to quantify the frequency of historical fires; the weather influence factors are used to quantify the probability of fires occurring under current meteorological conditions; the vegetation flammability factors are used to quantify the probability of vegetation spontaneous combustion or ignition; and the human activity factors are used to represent the probability of human-caused fires. The fire probability index corresponding to each area of ​​the mountain forest is obtained by weighted summation of the historical fire factor, the weather influence factor, the vegetation flammability factor, and the human activity factor. By combining the meteorological environment data, the vegetation distribution data, and the terrain facility data, a fire spread model corresponding to each area of ​​the mountain forest is established to determine the fire impact index corresponding to each area of ​​the mountain forest. The fire spread model is used to represent the dynamic propagation behavior of simulated fires, and the fire impact index is used to quantify the scope of the harm caused by a fire. Substituting the fire probability index and the fire impact index into the preset wildfire risk assessment function, the comprehensive risk value corresponding to each area of ​​the mountain forest is obtained. Based on the comprehensive risk value and the preset PTZ monitoring strategy, the corresponding PTZ monitoring strategy for each area of ​​the mountain forest is determined. The preset PTZ monitoring strategy includes different PTZ monitoring strategies corresponding to different comprehensive risk values, and the PTZ inspection strategy includes PTZ control parameters.

2. The method according to claim 1, characterized in that, The process of determining historical fire factors based on historical fire data, weather influence factors based on meteorological data, vegetation flammability factors based on vegetation distribution data, and human activity factors based on human activity data specifically includes: Based on the historical fire data, the number of historical fires in each area of ​​the mountain forest is counted, and the number of historical fires is normalized to obtain the historical fire factor for each area of ​​the mountain forest. The meteorological environmental data is input into the fire weather prediction model to obtain the weather influencing factors corresponding to each area of ​​the mountain forest. The fire weather prediction model is used to construct the mapping relationship between meteorological environmental conditions and the probability of fire occurrence. The flammability of various plants is determined based on a pre-set flammability table of forest plants. Combined with the vegetation distribution data corresponding to each area of ​​the forest, the vegetation flammability factor corresponding to each area of ​​the forest is determined. Based on the human activity data, the population density corresponding to each area of ​​the mountain forest is determined, so as to calculate the human activity factor corresponding to each area of ​​the mountain forest.

3. The method according to claim 1, characterized in that, The process of combining meteorological environmental data, vegetation distribution data, and terrain data to establish fire spread models for different areas of the mountain forest, in order to determine the fire impact index for each area of ​​the mountain forest, specifically includes: Wind speed, wind direction, temperature, and humidity were extracted from the meteorological environmental data and used as dynamic parameters for fire spread. The vegetation type, vegetation density, vegetation water content, and combustible load in the vegetation distribution data are extracted as fuel parameters for fire spread. The elevation, slope, aspect, ridgeline, and valleyline of the terrain facilities data are extracted as terrain parameters for fire spread. Based on the dynamic parameters, fuel parameters, and terrain parameters, a fire spread model corresponding to each area of ​​the mountain forest is constructed. Each area of ​​the mountain forest was set as a virtual fire source, and the corresponding fire spread model for each area of ​​the mountain forest was run to simulate the spread of the wildfire from the virtual fire source. The maximum speed, average speed, coverage area, and spread time of wildfires are calculated to determine the fire impact index for each area of ​​the forest.

4. The method according to claim 3, characterized in that, The calculation of the maximum speed, average speed, coverage area, and spread time of wildfires to determine the fire impact index for each area of ​​the forest specifically includes: Within a preset simulation period, the fire spread model is periodically iterated at a fixed time step to obtain the spatial distribution data of the wildfire spread at each fixed time step. Based on the spatial distribution data, the position of the fire front at each fixed time step is extracted, and the displacement vector of the fire front position between adjacent time steps is calculated; the magnitude of the displacement vector is divided by the fixed time step to obtain the instantaneous spread velocity in the preset direction. The maximum value among the instantaneous spread velocities is selected as the maximum velocity; The weighted average of the instantaneous spread velocity is calculated as the average velocity; The total area covered by fire during the preset simulation period is defined as the covered area. The time required for the coverage area to reach a preset area threshold is recorded as the spread time. Based on the maximum speed, the average speed, the coverage area, and the spread time, a nonlinear weighted fusion algorithm is used to calculate the fire impact index corresponding to each area of ​​the mountain forest.

5. The method according to claim 1, characterized in that, The preset wildfire risk assessment function is: R = αP + β +γ(PI) Wherein, R represents the comprehensive risk value, P represents the fire probability index, I represents the fire impact index, α, β and γ represent weighting coefficients, n represents the exponential parameter of the fire impact index and n>1; the weighting coefficients satisfy β>α>0, γ>0 and α+β+γ=1.

6. The method according to claim 1, characterized in that, After the step of determining the corresponding PTZ monitoring strategy for each area of ​​the mountain forest based on the comprehensive risk value and the preset PTZ monitoring strategy, the method further includes: Receive real-time image data corresponding to each area of ​​the mountain forest; The real-time image data is input into the smoke and fire recognition model to detect whether smoke and fire are present in the real-time image data; If smoke or fire is present, the fire spread model is invoked to predict the spread direction, spread speed, and impact range of the smoke or fire in real time, based on the geographical location information corresponding to the real-time image data. Based on the prediction results, the optimal scheduling plan for fire prevention and control resources is determined. The optimal scheduling plan includes the dispatch route of firefighters, the configuration plan of firefighting equipment, and the evacuation plan. The optimal scheduling scheme is sent to the terminal device to notify relevant personnel to carry out firefighting and evacuation work.

7. The method according to claim 6, characterized in that, After the step of sending the optimal scheduling plan to the terminal device to notify relevant personnel to carry out firefighting and evacuation work, the method further includes: Monitor the spread of wildfires and the progress of firefighting efforts; If the wildfire spread status is inconsistent with the prediction result, the fire spread model is updated based on the latest monitoring data to obtain the corrected fire spread prediction result. Based on the fire spread prediction results and / or the fire extinguishing progress, the optimal dispatching scheme is adjusted.

8. A wildfire monitoring system, characterized in that, The wildfire monitoring system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the wildfire monitoring system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the wildfire monitoring system, the wildfire monitoring system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the wildfire monitoring system, the wildfire monitoring system performs the method as described in any one of claims 1-7.

Images