Region site selection method, device and equipment for planned burn-off, and storage medium

Abstract

The invention provides a planned burn-off area site selection method, device and equipment and a storage medium, and is applied to the technical field of forest fire prevention and control. The method comprises the following steps: acquiring a fire risk data set of a target research area and a coexistence capability data set of the target research area; and utilizing the fire risk prediction model to predict a preset number of random points of the target research area to obtain a fire risk grid layer of the target research area. And based on the coexistence capability data set, constructing a coexistence capability grid layer of the target research area. And determining a fire management grid layer of the target research area according to the fire risk grid layer and the coexistence capability grid layer. And on the basis of the fire management grid map layer, determining the forest region of which the value is located before the priority ranking in the target research region as a priority burning-off region, so as to perform planned burning-off. The method can solve the problem that a traditional planned burning-off method cannot exert the fireproof and recovery functions of an ecological system.

Description

Technical Field

[0001] This application relates to the technical field of forest fire prevention and control, and particularly to a method, device, equipment and storage medium for site selection of prescribed burning areas. Background Art

[0002] Climate change has increased the frequency of abnormal and extremely destructive fire risk events. To effectively manage fires, prescribed burning, as an important technology for fire management and ecosystem restoration, has been widely applied. However, at the watershed scale, scientifically planning the areas and priorities of prescribed burning remains an important issue.

[0003] The scale and spatial pattern of prescribed burning treatments determine the extent to which its application can control the scope and severity of wildfires. Currently, prescribed burning methods mainly focus on single-dimensional risk assessment and geospatial analysis. Specifically, one is based on the static assessment of fuel load, for example, measuring the fuel load of shrub forests through stratified sampling or mechanical sampling methods, and designating areas with high fuel load as priority burning areas, mainly in areas with dense vegetation such as the wildland-urban interface (WUI) and the forest-grass interface. The other is based on the fire risk assessment of historical fire data and geospatial factors (vegetation type, terrain slope, meteorological conditions), and designating areas with high risk assessment results as priority burning areas.

[0004] However, this prescribed burning method mainly aims to reduce fire risks and ignores the ability of the ecosystem to coexist with fire itself, resulting in the inability to exert the fire prevention and restoration functions of the ecosystem itself. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, equipment and storage medium for site selection of prescribed burning areas, so as to solve the problem that the traditional prescribed burning method cannot exert the fire prevention and restoration functions of the ecosystem itself.

[0006] First aspect, an embodiment of the present application provides a method for selecting a location for planned burning, the method including: obtaining a fire risk data set of a target study area and a coexistence ability data set of the target study area. Using a fire risk prediction model to predict a preset number of random points in the target study area to obtain a fire risk raster layer of the target study area. Based on the coexistence ability data set, constructing a coexistence ability raster layer of the target study area. According to the fire risk raster layer and the coexistence ability raster layer, determining a fire management raster layer of the target study area. The fire management raster layer has a value for each pixel. A pixel is used to represent the smallest pixel in the fire management raster layer. Based on the fire management raster layer, determining the forest areas in the target study area with values ranked among the top priorities as the priority burning areas for planned burning. The value of a forest area is determined based on the values of at least one pixel corresponding to the forest area in the target study area.

[0007] The method for selecting a location for planned burning provided by the embodiment of the present application, in the regional planning of planned burning, by integrating multi-dimensional data of the fire risk data set and the coexistence ability data set of the target study area, can select the forest areas in the target study area with values ranked among the top priorities as the priority burning areas for planned burning, so as to carry out planned burning, realizing the scientific location selection of the planned burning space.

[0008] A possible implementation manner, the fire risk data set includes: a fire point data set and geospatial factors. The fire point data set includes historical fire points. The historical fire points are the points extracted from the historical fire records of the target study area. The geospatial factors include terrain data, land data, vegetation data, meteorological data, and human activity data. The terrain data is extracted from the digital elevation model image of the target study area. The terrain data includes elevation, slope, aspect, plane curvature, terrain position index, and terrain moisture index. The land data includes land cover and soil texture. The vegetation data includes the normalized difference vegetation index. The meteorological data includes potential evapotranspiration, drought index, wind speed, relative humidity, average temperature, and annual average rainfall. The human activity data is a multi-ring buffer generated based on the river network, roads, and built-up areas of the target study area according to a preset distance increment. The human activity data includes the distance to the river network, the distance to the road, and the distance to the built-up area.

[0009] A possible implementation manner, the determination process of the fire risk prediction model includes: constructing non-fire points in the target research area according to the historical fire points in the fire point data set in the fire risk data set. For each point among the historical fire points and non-fire points, extracting the corresponding geographical space factor value from the geographical space factors in the fire risk data set. The geographical space factor is a factor with a correlation less than the first threshold extracted from all factors in the target research area. Training and testing the historical fire points, non-fire points, the numerical values of the geographical space factors corresponding to the historical fire points, and the numerical values of the geographical space factors corresponding to the non-fire points based on a machine learning algorithm to obtain the fire risk prediction model.

[0010] A possible implementation manner, using the fire risk prediction model to predict a preset number of random points in the target research area to obtain the fire risk grid layer of the target research area, including: using the fire risk prediction model to predict the random points to obtain the point prediction result corresponding to each predicted point in the random points, and filling the point prediction result into the grid unit of the predicted point correspondingly. Using the interpolation method to fill the points outside the random points in the target research area to obtain the fire risk grid layer.

[0011] A possible implementation manner, the coexistence ability data set includes: potential habitat distribution data of fire-resistant plants, potential habitat distribution data of fire-proof plants, potential habitat distribution data of regenerative plants, potential habitat distribution data of fire-dependent plants, potential habitat distribution data of animals reducing combustibles, potential habitat distribution data of animals constructing fire breaks, potential habitat distribution data of animals escaping or taking shelter, potential habitat distribution data of animals forming or maintaining wetland fire breaks, traditional ecological wisdom system data, and water system soil data. The traditional ecological wisdom system is used to represent the traditional knowledge ecosystem of the target research area. The water system soil data includes soil organic matter content, soil erodibility factor, depression density, and river network density. The potential habitat distribution data of fire-resistant plants, potential habitat distribution data of fire-proof plants, potential habitat distribution data of regenerative plants, and potential habitat distribution data of fire-dependent plants are determined based on the plant distribution points in the target research area, the fire response characteristics of plants, and the environmental variable data of the target research area. The potential habitat distribution data of animals reducing combustibles, potential habitat distribution data of animals constructing fire breaks, potential habitat distribution data of animals escaping or taking shelter, and potential habitat distribution data of animals forming or maintaining wetland fire breaks are determined based on the animal distribution points in the target research area, the fire response characteristics of animals, and the environmental variable data of the target research area. The traditional ecological wisdom system data is obtained by using an image segmentation network model to identify the image features of the traditional ecological wisdom system in the target research area, identifying the image features based on the image segmentation network model, and calculating the traditional ecological wisdom system data of the target research area through the grid method. The traditional ecological wisdom system data includes the density of the traditional irrigation system.

[0012] A possible implementation manner, based on the fire management grid layer, the forest areas in the target research area with values ranked among the top in the priority ranking are determined as the priority burning areas, including: determining the value of each pixel of the fire management grid layer based on the fire risk grid layer, the fire risk weight corresponding to the fire risk grid layer, the coexistence ability weight corresponding to the coexistence ability grid layer, the coexistence ability grid layer of each category in the coexistence grid ability layer, and the weight corresponding to the coexistence ability grid layer of each category. Based on the value of each pixel of the fire management grid layer, the forest areas in the target research area with values ranked among the top in the priority ranking are determined as the priority burning areas.

[0013] A possible implementation method is to determine the priority burning areas by the value of each pixel in the fire management grid layer, including: screening out at least one forest area from the target study area. A forest area corresponds to at least one pixel in the fire management grid layer. Based on the value of the pixels corresponding to the forest area in the fire management grid layer, determine the value of the forest area. Iteratively sort the values of the forest areas in the target study area, and determine the forest areas with values in the target study area before the priority sorting as the priority burning areas.

[0014] A possible implementation method is that the area selection method for planned burning provided by the embodiments of the present application further includes: for any target coexistence ability grid layer in the coexistence ability grid layer, determine the coexistence ability entropy value of the target coexistence ability grid layer according to the information amount of the target coexistence ability grid layer. According to the coexistence ability entropy value, determine the coexistence ability weight of the target coexistence ability grid layer.

[0015] In a second aspect, the embodiments of the present application provide a device for selecting an area for planned burning, which includes an acquisition module, a prediction module, a construction module, and a determination module.

[0016] Among them, the acquisition module is used to acquire the fire risk data set of the target study area and the coexistence ability data set of the target study area.

[0017] The prediction module is used to predict a preset number of random points in the target study area by using a fire risk prediction model to obtain the fire risk grid layer of the target study area.

[0018] The construction module is used to construct the coexistence ability grid layer of the target study area based on the coexistence ability data set.

[0019] The determination module is used to determine the fire management grid layer of the target study area according to the fire risk grid layer and the coexistence ability grid layer. Each pixel in the fire management grid layer has a value. A pixel is used to represent the smallest pixel in the fire management grid layer. Based on the fire management grid layer, determine the forest areas with values in the target study area before the priority sorting as the priority burning areas for planned burning. The value of the forest area is determined based on the values of at least one pixel corresponding to the forest area in the target study area.

[0020] In a third aspect, the embodiments of the present application provide a device for selecting an area for planned burning, and the device for selecting an area for planned burning has the function of implementing the area selection method for planned burning according to the first aspect or any possible implementation method above. This function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0021] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, which, when running on a computer, enable the computer to execute the method for selecting a location of a planned erasure region according to the first aspect or any possible implementation manner thereof.

[0022] In a fifth aspect, an embodiment of the present application provides a computer program product containing computer-executable instructions, which, when running on a computer, enable the computer to execute the method for selecting a location of a planned erasure region according to the first aspect or any possible implementation manner thereof.

[0023] The technical effects brought by any of the design manners in the second aspect to the fifth aspect can refer to the technical effects brought by different possible implementation manners in the first aspect, which will not be elaborated here. Description of the Drawings

[0024] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0025] Figure 1 FIG. is a flowchart of a method for selecting a location of a planned erasure region provided by an embodiment of the present application; Figure 2a FIG. is a specific example diagram of a method for selecting a location of a planned erasure region provided by an embodiment of the present application; Figure 2b FIG. is another specific example diagram of a method for selecting a location of a planned erasure region provided by an embodiment of the present application; Figure 2c FIG. is still another specific example diagram of a method for selecting a location of a planned erasure region provided by an embodiment of the present application; Figure 3 FIG. is a schematic structural diagram of an apparatus for selecting a location of a planned erasure region provided by an embodiment of the present application; Figure 4 FIG. is a schematic structural diagram of a system for selecting a location of a planned erasure region provided by an embodiment of the present application. Detailed Embodiments

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. Components of the embodiments of this application generally described and illustrated in the figures herein can be arranged and designed in a variety of different configurations.

[0027] Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of this application that is claimed, but is merely representative of selected embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of this application without creative efforts fall within the scope of protection of this application.

[0028] Currently, the planned burning method is mainly based on single-dimensional risk assessment and geospatial analysis, without considering the ability of the ecosystem to coexist with fire itself, resulting in the inability to exert the fire prevention and restoration functions of the ecosystem itself.

[0029] Based on this, the embodiments of this application provide a method, device, equipment, and storage medium for site selection of planned burning areas. The method includes obtaining a fire risk data set of a target research area and a coexistence ability data set of the target research area. Using a fire risk prediction model to predict a preset number of random points in the target research area to obtain a fire risk raster layer of the target research area. Based on the coexistence ability data set, a coexistence ability raster layer of the target research area is constructed. According to the fire risk raster layer and the coexistence ability raster layer, a fire management raster layer of the target research area is determined. The fire management raster layer has a value for each pixel. A pixel is used to represent the smallest pixel in the fire management raster layer. Based on the fire management raster layer, the forest areas in the target research area with values ranked among the top in Prioritization are determined as the priority burning areas for planned burning. The value of the forest area is determined based on the values of at least one pixel corresponding to the forest area in the target research area.

[0030] The method for site selection of planned burning areas provided by the embodiments of this application, in the regional planning of planned burning, by integrating multi-dimensional data of the fire risk data set and the coexistence ability data set of the target research area, can select the forest areas in the target research area with values ranked among the top as the priority burning areas for planned burning, so as to achieve scientific site selection for the planned burning space.

[0031] On the one hand, the embodiments of this application provide a method for site selection of planned burning areas, as Figure 1 shown, the method may include the following steps.

[0032] S101. Obtain the fire risk dataset of the target research area and the coexistence ability dataset of the target research area.

[0033] Exemplarily, first obtain the satellite remote sensing data of the target research area, import the satellite remote sensing data into a Geographic Information System (GIS) for processing to obtain a vector file of the target research area. The vector file stores the vector map data of the target research area, and the vector map data records the position and attribute information of each element. The vector file consists of a header file and entity information, and the entity information stores coordinate information.

[0034] Among them, the fire risk dataset includes a fire point dataset and geospatial factors.

[0035] A possible implementation, the fire risk dataset includes historical fire points. The historical fire points can be the points extracted from the historical fire records of the target research area.

[0036] For example, the historical fire points can be the historical fire records of the target research area extracted from the VIIRS satellite thermal anomaly data, and the points where fires occurred in the history of the target research area are extracted from the historical fire records.

[0037] After obtaining the historical fire points of the target research area, construct an equal number of random non-fire points as the historical fire points in the GIS, and jointly construct the fire point dataset with the fire points and non-fire points.

[0038] It should be noted that the target number of non-fire points can be the same as the number of historical fire points in the target research area, or different from the historical fire points. This application does not limit this.

[0039] A possible implementation, the geospatial factors include terrain data, land data, vegetation data, meteorological data, and human activity data.

[0040] Specifically, the determination process of the geospatial factors is to extract all geospatial factors related to geography in the target research area. The geospatial factors are geospatial data that affect the occurrence of fires in the target research area, or geospatial data related to the fire risk occurrence in the target research area. Perform Pearson correlation analysis on all geospatial factors, and eliminate the multi-collinear elements, that is, eliminate the factors with very high correlation among all geospatial factors to obtain the geospatial factors, namely terrain data, land data, vegetation data, meteorological data, and human activity data.

[0041] Among them, the terrain data is extracted from the digital elevation model image of the target study area. The terrain data includes elevation, slope, aspect, plan curvature, terrain position index, and terrain wetness index. The land data includes land cover and soil texture. The vegetation data includes the normalized difference vegetation index. The meteorological data includes potential evapotranspiration, drought index, wind speed, relative humidity, average temperature, and annual average rainfall. The human activity data is a multi-ring buffer generated based on the river network, roads, and built-up areas in the target study area according to a preset distance increment. The human activity data includes the distance to the river network, the distance to the road, and the distance to the built-up area.

[0042] Exemplarily, a digital elevation model image (Digital Elevation Model, DEM) with a 30m accuracy of the target study area is obtained. The terrain data of the target study area, such as elevation, slope, aspect, plan curvature, terrain position index, and terrain wetness index, is derived from the digital elevation model image. Based on GIS, multi-ring buffers of the river network, roads, and built-up areas in the target study area are generated at a preset distance of 1km respectively, and the distance to the river network, the distance to the road, and the distance to the built-up area in the human activity data are obtained.

[0043] By integrating the fire risk dataset and the coexistence ability dataset, this application breaks through the limitation of only focusing on fire prevention in traditional methods and takes both fire risk and adaptive coexistence into account.

[0044] A possible implementation manner is that the coexistence ability dataset may include: potential habitat distribution data of fire-resistant plants, potential habitat distribution data of fire-tolerant plants, potential habitat distribution data of regenerative plants, potential habitat distribution data of fire-dependent plants, potential habitat distribution data of animals reducing combustibles, potential habitat distribution data of animals constructing firebreaks, potential habitat distribution data of animals escaping or taking shelter, potential habitat distribution data of animals forming or maintaining wetland firebreaks, traditional ecological wisdom system (indigenous culture) data, and water system soil data.

[0045] Specifically, the determination process of the coexistence ability dataset is to extract all coexistence elements in the target study area. Perform Pearson correlation analysis on all coexistence elements, and remove any one of the coexistence elements whose correlation coefficients in all coexistence elements do not meet the preset range to obtain the coexistence ability dataset. For example, remove the coexistence elements with the correlation coefficient r in the range of |r|>0.95 and |r|<0.2.

[0046] Among them, the potential habitat distribution data of fire-resistant plants, fire-tolerant plants, fire-dependent plants, and regeneration plants are determined based on the plant distribution points, the fire response characteristics of plants, and the environmental variable data of the target study area.

[0047] The environmental variable data may include elevation, slope, aspect, vegetation index, annual mean temperature, mean diurnal range, temperature seasonality, maximum temperature of the warmest month, maximum temperature of the coldest month, annual temperature range, mean temperature of the wettest quarter, mean temperature of the driest quarter, mean temperature of the warmest quarter, mean temperature of the coldest quarter, annual precipitation, precipitation of the wettest month, driest and non-precipitation, precipitation seasonality, precipitation in the driest area, precipitation in the warmest quarter, precipitation in the coldest quarter, population density, night lights, etc.

[0048] Exemplarily, as Figure 2a shown, first, plants are classified into fire-resistant plants, fire-tolerant plants, fire-dependent plants, and regeneration plants based on the fire response characteristics of plants in the target study area. Obtain the distribution points of fire-resistant plants, fire-tolerant plants, fire-dependent plants, and regeneration plants in the target study area. Use the environmental variable data of the target study area to determine the correlation between the distribution points of fire-resistant plants, fire-tolerant plants, fire-dependent plants, and regeneration plants and the environmental variable data, and use the MaxEnt model for prediction to obtain the potential habitat distribution data of fire-resistant plants, fire-tolerant plants, fire-dependent plants, and regeneration plants.

[0049] The potential habitat distribution data of animals reducing combustibles, animals constructing firebreaks, animals escaping or taking shelter, and animals forming or maintaining wetland firebreaks are determined based on the animal distribution points, the fire response characteristics of animals, and the environmental variable data of the target study area.

[0050] Exemplarily, first, animals are classified into combustible - reducing animals, fire - belt - constructing animals, escaping or avoiding animals, and wetland - fire - belt - forming or maintaining animals based on the characteristics of animals' responses to fires in the target research area. Distribution points of combustible - reducing animals, fire - belt - constructing animals, escaping or avoiding animals, and wetland - fire - belt - forming or maintaining animals in the target research area are obtained. Using the environmental variable data of the target research area, the correlations between the distribution points of combustible - reducing animals, fire - belt - constructing animals, escaping or avoiding animals, and wetland - fire - belt - forming or maintaining animals in the target research area and the environmental variable data are determined, and the MaxEnt model is used for prediction to obtain the potential habitat distribution data of animals for reducing combustibles, the potential habitat distribution data of animals for constructing fire belts, the potential habitat distribution data of animals for escaping or avoiding, and the potential habitat distribution data of animals for forming or maintaining wetland fire belts.

[0051] The traditional ecological wisdom system data is used to represent the traditional knowledge ecosystem in the target research area. This traditional ecological wisdom system data can be obtained by using the U - Net image segmentation technology of satellite remote sensing or aerial photography images. The image segmentation network model is used to identify the traditional fire management traces in the remote sensing images of the target research area, such as traditional irrigation systems, periodic burning patches, or land - use boundaries, etc., and the traditional ecological wisdom system data is calculated by the grid method. For example, the traditional ecological wisdom system data can be the density of traditional irrigation systems.

[0052] In this application, through the traditional ecological wisdom system data, the traditional fire management knowledge is transformed into quantifiable spatial data, which can increase the cultural adaptability and ecological rationality of the traditional ecological wisdom system.

[0053] The water - system soil data includes the content of soil erodibility factor, the content of soil organic matter, river network density, and depression (Sink) density.

[0054] Among them, the content of soil erodibility factor is used to represent the soil's anti - erosion ability. The content of organic matter is used to represent the nutrient release and soil microbial activity after burning. The river network density is used to reflect the density of the surface runoff network. The depression density is used to represent the density of natural water storage areas.

[0055] It should be noted that the content of soil erodibility factor and the content of soil organic matter can be obtained by monitoring the target research area or directly downloaded from relevant websites. This application does not limit this.

[0056] Among them, the river network density and the depression density can be obtained by performing hydrological analysis based on the topographic data of the target research area. For example, the grid method is used to calculate the river network density and the depression density of the target research area in the DEM.

[0057] S102. Use the fire risk prediction model to predict a preset number of random points in the target research area, and obtain the fire risk raster layer of the target research area.

[0058] Among them, the fire risk prediction model is obtained by training and testing the fire point dataset and geospatial factors based on machine learning algorithms.

[0059] A possible implementation method is to construct non-fire points in the target research area according to the historical fire points in the fire point dataset. For each point in the historical fire points and non-fire points, extract the corresponding geospatial factor values from the geospatial factors. Based on machine learning algorithms, train and test the historical fire points, non-fire points, the numerical values of the geospatial factors corresponding to the historical fire points, and the numerical values of the geospatial factors corresponding to the non-fire points to obtain the fire risk prediction model.

[0060] Exemplarily, for each point in the historical fire points and non-fire points, extract the corresponding geospatial factor values from the geospatial factors. Divide the fire point dataset into a training set and a test set according to a preset ratio, such as 7:3 or 8:2. Among them, the ratio of fire points and non-fire points in the subset remains the same. Use the Python algorithm Sklearn package in PyCharm to construct a Gradient Boosting Decision Tree (GBDT) model, use the training set model to train the GBDT model, and use methods such as grid search or Bayesian optimization to systematically tune the parameters of the GBDT model.

[0061] A possible implementation method is to use the fire risk prediction model to predict random points, obtain the point prediction results corresponding to each predicted point in the random points, and fill the point prediction results into the grid cells of the predicted points correspondingly. Use the interpolation method to fill the points outside the random points in the target research area to obtain the fire risk raster layer.

[0062] Exemplarily, create uniform random points covering the target research area in GIS. Among them, the density of the random points can be set according to requirements. Extract the environmental variable values of each random point, input them into the trained GBDT model, predict the fire occurrence probability of each random point, and attach the predicted probability to the grid cell of the random point to obtain the fire risk table of the random points and the fire occurrence probability of the random points in the target research area.

[0063] Then, use the Inverse Distance Weighting (IDW) tool in GIS to fill in the points outside the random points in the target study area, and convert the fire risk table of the random points in the target study area into a continuous fire risk raster layer as shown in Figure 2b The continuous fire risk raster layer is divided into five levels: extremely high, high, medium, low, and extremely low.

[0064] It should be noted that the model in this application can be a GBDT model, or a Random Forests (RF) or Support Vector Machines (SVM). This application does not limit this.

[0065] This application constructs a prediction model based on historical fire points and geospatial data to predict the fire probability of random points in the target study area, generates a fire risk raster layer, and can quantify the ignition probability and spread risk of different points in the target study area, avoiding judgment biases of subjective experience.

[0066] S103. Based on the coexistence ability dataset, construct a coexistence ability raster layer for the target study area.

[0067] Among them, the coexistence ability raster layer includes the potential habitat distribution layer of fire-resistant plants, the potential habitat distribution layer of fire-proof plants, the potential habitat distribution layer of regenerative plants, the potential habitat distribution layer of fire-dependent plants, the potential habitat distribution layer of animals reducing combustibles, the potential habitat distribution layer of animals constructing fire breaks, the potential habitat distribution layer of animals escaping or taking shelter, the potential habitat distribution layer of animals forming or maintaining wetland fire breaks, the traditional ecological wisdom system layer, the soil organic matter content layer, the soil erodibility factor layer, the depression density layer, and the river network density layer.

[0068] A possible implementation method is to construct layers based on the fire response characteristics of plants and animals respectively. Based on the distribution data of the potential habitats of fire-resistant plants, construct a layer of the distribution of potential habitats of fire-resistant plants. Based on the distribution data of the potential habitats of fire-proof plants, construct a layer of the distribution of potential habitats of fire-proof plants. Based on the distribution data of the potential habitats of regenerative plants, construct a layer of the distribution of potential habitats of regenerative plants. Based on the distribution data of the potential habitats of fire-dependent plants, construct a layer of the distribution of potential habitats of fire-dependent plants. Based on the distribution data of the potential habitats of animals reducing combustibles, construct a layer of the distribution of potential habitats of animals reducing combustibles. Based on the distribution data of the potential habitats of animals constructing fire breaks, construct a layer of the distribution of potential habitats of animals constructing fire breaks. Based on the distribution data of the potential habitats of animals escaping or taking shelter, construct a layer of the distribution of potential habitats of animals escaping or taking shelter. Based on the distribution data of the potential habitats of animals forming or maintaining wetland fire breaks, construct a layer of the distribution of potential habitats of animals forming or maintaining wetland fire breaks.

[0069] Then, based on the data of the traditional ecological wisdom system, construct a layer of the traditional ecological wisdom system. Based on the content of soil organic matter in the water system soil data, construct a layer of soil organic matter content. Based on the content of soil erodibility factors in the water system soil data, construct a layer of soil erodibility factors. Based on the depression density in the water system soil data, construct a layer of depression density. Based on the river network density in the water system soil data, construct a layer of river network density.

[0070] S104. According to the fire risk raster layer and the coexistence ability raster layer, determine the fire management raster layer of the target study area.

[0071] A possible implementation method is to integrate the fire risk raster layer and the coexistence ability raster layer through a fusion tool to obtain the fire management raster layer of the target study area.

[0072] Exemplarily, as Figure 2c shown, use Zonation5 to unify the formats and attributes of the fire risk raster layer and the coexistence ability raster layer, and spatially align the pixels of the fire risk raster layer and the pixels of the coexistence ability raster layer. Among them, the spatial alignment includes one or more of the following: resolution alignment, range alignment, and coordinate system alignment. Then, integrate the fire risk raster layer and the coexistence ability raster layer to obtain the fire management raster layer of the target study area.

[0073] S105. Based on the fire management raster layer, determine the forest areas in the target study area with values ranked among the top priorities as the priority burning areas for planned burning.

[0074] A possible implementation method is to determine the coexistence ability entropy value of any target coexistence ability grid layer in the coexistence ability grid layer according to the information volume of the target coexistence ability grid layer. According to the coexistence ability entropy value, determine the coexistence ability weight of the target coexistence ability grid layer.

[0075] Specifically, for the anti-fire plant potential habitat distribution layer, fire-resistant plant potential habitat distribution layer, regenerative plant potential habitat distribution layer, fire-dependent plant potential habitat distribution layer, animal-reducing combustible potential habitat distribution layer, animal-constructing firebreak potential habitat distribution layer, animal-escape or -evasion potential habitat distribution layer, animal-forming or -maintaining wetland firebreak potential habitat distribution layer, traditional ecological wisdom system layer, soil organic matter content layer, soil erodibility factor layer, depression density layer, and river network density layer in the coexistence ability grid layer, determine their coexistence ability entropy values according to their information volumes. Furthermore, according to their coexistence ability entropy values, the anti-fire plant potential habitat distribution weight, fire-resistant plant potential habitat distribution weight, regenerative plant potential habitat distribution weight, fire-dependent plant potential habitat distribution weight, animal-reducing combustible potential habitat distribution weight, animal-constructing firebreak potential habitat distribution weight, animal-escape or -evasion potential habitat distribution weight, animal-forming or -maintaining wetland firebreak potential habitat distribution weight, traditional ecological wisdom system weight, soil organic matter content weight, soil erodibility factor weight, depression density weight, and river network density weight can be obtained.

[0076] A possible implementation method is to determine the value of each pixel in the fire management grid layer according to the fire risk grid layer, the fire risk weight corresponding to the fire risk grid layer, the coexistence ability weight corresponding to the coexistence ability grid layer, the coexistence ability grid layer, and the weight corresponding to the coexistence ability grid layer.

[0077] Specifically, based on the fire risk raster layer, fire risk weight, coexistence ability weight, potential habitat distribution layer of fire-resistant plants, potential habitat distribution weight of fire-resistant plants, potential habitat distribution layer of fire-retardant plants, potential habitat distribution weight of fire-retardant plants, potential habitat distribution layer of regenerative plants, potential habitat distribution weight of regenerative plants, potential habitat distribution layer of fire-dependent plants, potential habitat distribution weight of fire-dependent plants, potential habitat distribution layer of animals reducing combustibles, potential habitat distribution weight of animals reducing combustibles, potential habitat distribution layer of animals constructing fire breaks, potential habitat distribution weight of animals constructing fire breaks, potential habitat distribution layer of animals escaping or taking shelter, potential habitat distribution weight of animals escaping or taking shelter, potential habitat distribution layer of animals forming or maintaining wetland fire breaks, potential habitat distribution weight of animals forming or maintaining wetland fire breaks, traditional ecological wisdom system layer, traditional ecological wisdom system weight, soil organic matter content layer, soil organic matter content weight, soil erodibility factor layer, soil erodibility factor weight, depression density layer, depression density weight, river network density layer, and river network density weight, determine the value of each pixel in the fire management raster layer.

[0078] For example, the fire risk weight can be 50%, and the coexistence ability weight can be 50%.

[0079] Furthermore, based on the value of each pixel in the fire management raster layer, determine the forest areas in the target study area with values ranked among the top priorities as the priority burn areas.

[0080] Specifically, screen out at least one forest area from the target study area. Among them, one forest area corresponds to at least one pixel in the fire management raster layer. Based on the value of the pixels corresponding to the forest area in the fire management raster layer, determine the value of the forest area. Iteratively sort the values of the forest areas in the target study area, and determine the forest areas in the target study area with values ranked among the top priorities as the priority burn areas.

[0081] Exemplarily, through the Zonation software and the iterative conditional sort algorithm, gradually eliminate the low-value areas in the target study area. Substitute the values of the forest areas into GIS for visualization processing, and retain the forest areas in the target study area with values ranked in the top 5% as the priority burn areas, so as to preferentially select the areas with higher fire risks and significant ecological benefits for planned burning.

[0082] The above mainly introduced the solution provided by the embodiments of the present application from the perspective of the working principle of the device. It can be understood that in order to implement the above functions, the device for selecting the location of the area to be burned includes the corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should easily realize that, in combination with the algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the way of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0083] The embodiments of the present application can divide the functional modules of the device for selecting the location of the area to be burned according to the above method examples. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module.

[0084] It should be noted that the division of modules in the embodiments of the present application is illustrative, only a logical function division, and there can be other division methods in actual implementation. In the case of dividing each functional module corresponding to each function, Figure 3 shows a possible schematic composition diagram of the device for selecting the location of the area to be burned involved in the above and embodiments. As Figure 3 shown, the device 300 for selecting the location of the area to be burned may include: an acquisition module 301, a prediction module 302, a construction module 303, and a determination module 304.

[0085] Among them, the acquisition module 301 is used to support the device 300 for selecting the location of the area to be burned to execute Figure 1 S101 in the method for selecting the location of the area to be burned shown schematically.

[0086] The prediction module 302 is used to support the device 300 for selecting the location of the area to be burned to execute Figure 1 S102 in the method for selecting the location of the area to be burned shown schematically.

[0087] The construction module 303 is used to support the device 300 for selecting the location of the area to be burned to execute Figure 1 S103 in the method for selecting the location of the area to be burned shown schematically.

[0088] The determination module 304 is used to support the device 300 for selecting the location of the area to be burned to execute Figure 1 S104 and S105 in the method for selecting the location of the area to be burned shown schematically.

[0089] A possible implementation is that the fire risk dataset includes: a fire point dataset and geospatial factors. The fire point dataset includes historical fire points. The historical fire points are the points extracted from the historical fire records of the target research area. The geospatial factors include terrain data, land data, vegetation data, meteorological data, and human activity data. The terrain data is extracted from the digital elevation model image of the target research area. The terrain data includes elevation, slope, aspect, plane curvature, terrain position index, and terrain moisture index. The land data includes land cover and soil texture. The vegetation data includes the normalized difference vegetation index. The meteorological data includes potential evapotranspiration, drought index, wind speed, relative humidity, average temperature, and annual average rainfall. The human activity data is a multi-ring buffer generated based on the river network, roads, and built-up areas of the target research area in increasing order of a preset distance. The human activity data includes the river network, roads, and built-up areas.

[0090] A possible implementation is that the device can also be used to construct non-fire points in the target research area according to the historical fire points of the target research area recorded in the fire point dataset. For each point in the historical fire points and non-fire points, the geospatial factor values corresponding to the point are extracted from the geospatial factors. The geospatial factors are the factors with a correlation less than the first threshold extracted from all factors in the target research area. Based on the machine learning algorithm, the historical fire points, non-fire points, the numerical values of the geospatial factors corresponding to the historical fire points, and the numerical values of the geospatial factors corresponding to the non-fire points are trained and tested to obtain a fire risk prediction model.

[0091] A possible implementation is that the device can also be used to predict random points using the fire risk prediction model, obtain the point prediction results corresponding to each predicted point in the random points, and fill the point prediction results into the grid cells of the predicted points. The interpolation method is used to fill the points outside the random points in the target research area to obtain a fire risk grid layer.

[0092] A possible implementation, the coexistence ability dataset includes: potential habitat distribution data of fire-resistant plants, potential habitat distribution data of fire-tolerant plants, potential habitat distribution data of regenerative plants, potential habitat distribution data of fire-dependent plants, potential habitat distribution data of animals reducing combustibles, potential habitat distribution data of animals constructing firebreaks, potential habitat distribution data of animals escaping or taking shelter, potential habitat distribution data of animals forming or maintaining wetland firebreaks, traditional ecological wisdom system data, and water system soil data. The traditional ecological wisdom system is used to represent the traditional knowledge ecosystem of the target research area. The water system soil data includes soil organic matter content, soil erodibility factor, depression density, and river network density. The potential habitat distribution data of fire-resistant plants, potential habitat distribution data of fire-tolerant plants, potential habitat distribution data of regenerative plants, and potential habitat distribution data of fire-dependent plants are determined based on the plant distribution points in the target research area, the fire response characteristics of plants, and the environmental variable data of the target research area. The potential habitat distribution data of animals reducing combustibles, potential habitat distribution data of animals constructing firebreaks, potential habitat distribution data of animals escaping or taking shelter, and potential habitat distribution data of animals forming or maintaining wetland firebreaks are determined based on the animal distribution points in the target research area, the fire response characteristics of animals, and the environmental variable data of the target research area. The traditional ecological wisdom system data is obtained by using an image segmentation network model to identify the image features of the traditional ecological wisdom system in the target research area, identifying the image features based on the image segmentation network model, and calculating the traditional ecological wisdom system data of the target research area through the grid method. The traditional ecological wisdom system data includes the density of traditional irrigation systems.

[0093] A possible implementation, the device can also be used to determine the value of each pixel of the fire management grid layer based on the fire risk grid layer, the fire risk weight corresponding to the fire risk grid layer, the coexistence ability weight corresponding to the coexistence ability grid layer, the coexistence ability grid layer of each category in the coexistence grid ability layer, and the weight corresponding to the coexistence ability grid layer of each category. Based on the value of each pixel of the fire management grid layer, the forest area in the target research area with a value ranked among the top in the priority ranking is determined as the priority burning area.

[0094] A possible implementation, the device can also be used to screen out at least one forest area from the target research area. One forest area corresponds to at least one pixel of the fire management grid layer. Based on the value of the pixel corresponding to the forest area in the fire management grid layer, the value of the forest area is determined. The values of the forest areas in the target research area are iteratively ranked, and the forest area in the target research area with a value ranked among the top in the priority ranking is determined as the priority burning area.

[0095] In a possible implementation, the device can also be used to determine the coexistence ability entropy value of any target coexistence ability grid layer in the coexistence ability grid layer according to the information amount of the target coexistence ability grid layer. According to the coexistence ability entropy value, determine the coexistence ability weight of the target coexistence ability grid layer.

[0096] It should be noted that all relevant contents of each step involved in the above method embodiments can be cited in the function descriptions of the corresponding functional modules, and will not be elaborated here.

[0097] The area selection device 300 for planned burning provided in the embodiments of the present application is used to execute the above Figure 1 shown area selection method for planned burning, so it can achieve the same effect as the above area selection method for planned burning.

[0098] The embodiments of the present application also provide an area selection device for planned burning. The area selection device for planned burning can execute the area selection method and related steps in the above method embodiments.

[0099] The embodiments of the present application also provide a computer-readable storage medium, on which instructions are stored. When the instructions are executed, the area selection method and related steps in the above method embodiments are executed.

[0100] The embodiments of the present application also provide a computer program product. When the computer program product runs on a computer, the computer is caused to execute the area selection method and related steps in the above method embodiments.

[0101] In some embodiments, the method shown in the present application can be implemented as computer program instructions encoded in a computer-readable storage medium in a machine-readable format or encoded in other non-transitory media or articles.

[0102] The embodiments of the present application also provide an area selection system 400 for planned burning, as Figure 4 shown. The area selection system 400 for planned burning includes at least one processor 401 and at least one interface circuit 402.

[0103] As an example, when the area selection system 400 for planned burning includes one processor and one interface circuit, the one processor can be Figure 4 the processor 401 shown in the solid line box (or the processor 401 shown in the dotted line box), and the one interface circuit can be Figure 4 the interface circuit 402 shown in the solid line box (or the interface circuit 402 shown in the dotted line box). When the area selection system 400 for planned burning includes two processors and two interface circuits, the two processors includeFigure 4 The processor 401 shown by the solid line box and the processor 401 shown by the dashed line box, the two interface circuits include Figure 4 the interface circuit 402 shown by the solid line box and the interface circuit 402 shown by the dashed line box. There is no limitation on this.

[0104] The processor 401 and the interface circuit 402 can be interconnected by lines. For example, the interface circuit 402 can be used to receive signals. Also, for example, the interface circuit 402 can be used to send signals to other devices (such as the processor 401). By way of example, the interface circuit 402 can read the computer instructions stored in the memory and send the computer instructions to the processor 401. The processor 401 executes the instruction and, in combination with the input / output device, implements each step in the above embodiments, for example, implements Figure 1 each step performed in the method embodiment shown. Of course, the planned burn area location system may also include other discrete devices, and there is no specific limitation on this in the embodiments of the present application.

[0105] Through the description of the above embodiments, those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.

[0106] In several embodiments provided in the present application, it should be understood that the disclosed device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point, the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.

[0107] The units described as separate components may or may not be physically separated. The components shown as units can be one physical unit or multiple physical units, that is, they can be located in one place, or they can be distributed to multiple different places. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0108] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0109] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present application, in essence, or the part that makes a contribution, or all or part of the technical solution, can be embodied in the form of a software product. The software product is stored in a storage medium and includes several instructions for causing a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0110] The above content is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for selecting a site for planned burning, characterized in that: The method comprises: Acquire a fire risk dataset of a target study area and a coexistence capacity dataset of the target study area; Using a fire risk prediction model to predict a preset number of random points in the target study area, to obtain a fire risk grid layer of the target study area; Based on the coexistence capability dataset, construct a coexistence capability grid layer of the target research area; Determine a fire management raster layer of the target study area according to the fire risk raster layer and the coexistence capacity raster layer; the fire management raster layer carries the value of each pixel; the pixel is used to represent the minimum pixel in the fire management raster layer; Based on the fire management raster layer, the forest area in the target study area whose value is higher in the priority ranking is determined as the priority burning area for planned burning; the value of the forest area is determined based on the value of at least one pixel corresponding to the forest area in the target study area.

2. The method according to claim 1, characterized in that The fire risk dataset includes: a fire point dataset and a geographic space factor; The fire point data set includes historical fire points; the historical fire points are points extracted from historical fire records of the target research area; The geospatial factors include terrain data, land data, vegetation data, meteorological data and human activity data; The terrain data is extracted from the digital elevation model image of the target research area; the terrain data includes elevation, slope, slope aspect, plane curvature, terrain position index and terrain moisture index; The land data includes land cover and soil texture; The vegetation data includes a normalized vegetation difference index; The meteorological data include potential evapotranspiration, drought index, wind speed, relative humidity, average temperature and average annual rainfall; The human activity data is a multi-ring buffer zone generated incrementally according to a preset distance based on the river network, roads and built-up areas in the target study area; the human activity data includes the distance to the river network, the distance to the road and the distance to the built-up area.

3. The method according to claim 1 or 2, characterized in that: The determination process of the fire risk prediction model includes: Constructing non-fire points in the target study area according to the historical fire points in the target study area recorded in the fire point data set in the fire risk data set; For each of the historical fire points and the non-fire points, extracting a geospatial factor value corresponding to the point from the geospatial factors in the fire risk dataset; the geospatial factor is a factor extracted from all factors in the target study area whose correlation is less than a first threshold; The fire risk prediction model is obtained by training and testing the historical fire points, the non-fire points, the values ​​of the geographic spatial factors corresponding to the historical fire points, and the values ​​of the geographic spatial factors corresponding to the non-fire points based on a machine learning algorithm.

4. The method according to claim 1, characterized in that The method of using the fire risk prediction model to predict a preset number of random points in the target study area to obtain a fire risk grid layer of the target study area includes: Using the fire risk prediction model to predict the random points, obtain a point prediction result corresponding to each predicted point in the random points, and fill the point prediction result into the grid unit of the predicted point; The points other than the random points in the target study area are filled by using the interpolation method to obtain the fire risk grid layer.

5. The method according to claim 1, characterized in that The coexistence capacity dataset includes: distribution data of potential habitats of fire-resistant plants, distribution data of potential habitats of fire-tolerant plants, distribution data of potential habitats of regenerating plants, distribution data of potential habitats of fire-dependent plants, distribution data of potential habitats of animal-reduced fuels, distribution data of potential habitats of animal-built firebreaks, distribution data of potential habitats of animal-escaped or -avoided, distribution data of potential habitats of animal-formed or -maintained wetland firebreaks, traditional ecological wisdom system data and water system soil data; the traditional ecological wisdom system is used to represent the traditional knowledge ecosystem of the target research area; the water system soil data includes soil organic matter content, soil erodibility factor, depression density and river network density; The fire-resistant plant potential habitat distribution data, the fire-tolerant plant potential habitat distribution data, the regeneration plant potential habitat distribution data and the fire-dependent plant potential habitat distribution data are determined based on the plant distribution points in the target study area, the plant's fire response characteristics and the environmental variable data of the target study area; The distribution data of potential habitats for reducing combustibles by animals, the distribution data of potential habitats for building firebreaks by animals, the distribution data of potential habitats for escaping or hiding by animals, and the distribution data of potential habitats for forming or maintaining wetland firebreaks by animals are determined based on the animal distribution points in the target study area, the fire response characteristics of the animals, and the environmental variable data of the target study area; The traditional ecological wisdom system data is obtained by using an image segmentation network model to identify the image features of the traditional ecological wisdom system in the target research area, identifying the image features based on the image segmentation network model, and calculating the traditional ecological wisdom system data of the target research area through a grid method; the traditional ecological wisdom system data includes the density of the traditional irrigation system.

6. The method according to claim 1, characterized in that The method of determining the forest areas in the target study area with values ​​higher than the priority ranking as priority burning areas based on the fire management grid layer includes: Determine the value of each pixel of the fire management grid layer based on the fire risk grid layer, the fire risk weight corresponding to the fire risk grid layer, the coexistence capacity weight corresponding to the coexistence capacity grid layer, the coexistence capacity grid layer of each category in the coexistence capacity grid layer, and the weight corresponding to the coexistence capacity grid layer of each category; Based on the value of each pixel of the fire management raster layer, the forest areas in the target study area whose values ​​are at the top of the priority ranking are determined as priority burning areas.

7. The method according to claim 6, characterized in that The method of determining the forest area whose value in the target study area is higher than the priority ranking as the priority burning area based on the value of each pixel of the fire management grid layer comprises: Screening out at least one forest area from the target study area; one forest area corresponds to at least one pixel of the fire management raster layer; Determining the value of the forest area based on the value of the pixel corresponding to the forest area in the fire management raster layer; The values ​​of the forest areas in the target study area are iteratively ranked, and the forest areas in the target study area whose values ​​are at the top of the priority ranking are determined as priority burning areas.

8. The method according to claim 6, characterized in that The method further comprises: For any target coexistence capability grid layer in the coexistence capability grid layers, determining a coexistence capability entropy value of the target coexistence capability grid layer according to the information amount of the target coexistence capability grid layer; The coexistence capability weight of the target coexistence capability grid layer is determined according to the coexistence capability entropy value.

9. A device for selecting a location for planned burning, characterized in that: The device comprises: An acquisition module, used to acquire a fire risk dataset of a target study area and a coexistence capability dataset of the target study area; A prediction module, used to predict a preset number of random points in the target study area using a fire risk prediction model to obtain a fire risk grid layer of the target study area; A construction module, used to construct a coexistence capability grid layer of the target research area based on the coexistence capability dataset; A determination module is used to determine the fire management raster layer of the target study area based on the fire risk raster layer and the coexistence capacity raster layer; the fire management raster layer carries the value of each pixel; the pixel is used to represent the minimum pixel in the fire management raster layer; based on the fire management raster layer, the forest area in the target study area whose value is at the top of the priority ranking is determined as the priority burning area for planned burning; the value of the forest area is determined based on the value of at least one of the pixels corresponding to the forest area in the target study area.

10. A device for selecting a location for planned burning, characterized in that: The area site selection device for planned burning includes a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the area site selection method for planned burning according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the regional site selection method for planned burning according to any one of claims 1 to 8.

Images