Fire danger grade prediction method and device, storage medium and program product

Abstract

The invention discloses a fire danger grade prediction method and device, a storage medium and a program product. Relates to the field of image processing, and the method comprises the steps that remote sensing data and ground point location data of a target area are collected, a first data set is obtained, the ground point location data comprise data collected by a ground monitoring station in the target area, and the data types of the remote sensing data and the ground point location data comprise meteorological data, vegetation data and topographic data; data screening is carried out on the first data set to obtain a second data set, and the second data set is the screened first data set; performing data fusion on the remote sensing data and the ground point location data in the second data set to obtain a target data set; and determining a fire danger level of the target area based on the target data set. Through the method and the device, the problem of inaccurate fire danger level prediction result caused by adopting single-scale data to predict the fire danger level in the related technology is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image processing, in particular to a fire risk level prediction method and device, a storage medium and a program product. BACKGROUND

[0002] Fire risk level prediction technology plays a crucial role in forest fire prevention and management. The fire risk level prediction method in the related art mainly relies on single-scale data, which leads to limitations in prediction accuracy and real-time performance at different spatial scales. For example, in large-scale prediction, it is difficult to achieve comprehensive control of fire risk factors due to the lack of fine data for local areas. In small-scale prediction, although more accurate data can be provided, it is often limited by the real-time performance and update frequency of the data, making it difficult to meet the demand for rapid response. Therefore, the fire risk level prediction result is not accurate when using single-scale data in the related art.

[0003] To address the issue of inaccurate fire risk level prediction results caused by using single-scale data in the related art, no effective solutions have been proposed so far. SUMMARY

[0004] The main purpose of the present application is to provide a fire risk level prediction method, device, storage medium and program product to solve the problem of inaccurate fire risk level prediction results caused by using single-scale data in the related art.

[0005] To achieve the above-mentioned purpose, according to one aspect of the present application, a fire risk level prediction method is provided. The method comprises: collecting remote sensing data and ground point data of a target area to obtain a first data set, wherein the ground point data includes data collected by ground monitoring stations in the target area, and the data types of the remote sensing data and the ground point data include meteorological data, vegetation data and terrain data; performing data filtering on the first data set to obtain a second data set, wherein the second data set is the first data set after filtering; performing data fusion on the remote sensing data and the ground point data in the second data set to obtain a target data set; and determining the fire risk level of the target area based on the target data set.

[0006] Further, the data fusion of the remote sensing data and the ground point data in the second data set to obtain the target data set comprises: determining a data fusion algorithm, wherein the data fusion algorithm comprises a Bayesian fusion algorithm; and performing data fusion on the remote sensing data and the ground point data in the second data set based on the data fusion algorithm to obtain the target data set.

[0007] Further, the first data set is subjected to data screening to obtain a second data set, including: the first data set is subjected to data screening based on statistical parameters of the first data set, and / or the first data set is subjected to data screening based on a clustering analysis strategy to obtain the second data set, wherein the statistical parameters include at least one of the following: the average of the remote sensing data, the standard deviation of the remote sensing data.

[0008] Further, the first data set is subjected to data screening based on statistical parameters of the first data set, including: obtaining the spatial resolution of the remote sensing data; determining a target value range based on the spatial resolution of the remote sensing data, the average of the remote sensing data and the standard deviation of the remote sensing data, wherein the target value range is used to identify abnormal data in the point data; filtering the ground point data in the first data set based on the target value range.

[0009] Further, the first data set is subjected to data screening based on a clustering analysis strategy, including: interpolating the ground point data in the first data set to obtain a surface data set, wherein the surface data set records point data in multiple dimensions; determining a target distance range based on the surface data set and the spatial resolution of the remote sensing data, wherein the target distance range is used to identify abnormal data in the surface data set; determining the distance between each data in the surface data set and the center position of the surface data set to obtain a target distance set; and filtering the surface data set based on the target distance set and the target distance range.

[0010] Further, the target data set includes: target weather data, target vegetation data and target terrain data, wherein the target weather data is obtained by fusing weather data in the remote sensing data and weather data in the ground point data, the target vegetation data is obtained by fusing vegetation data in the remote sensing data and vegetation data in the ground point data, and the target terrain data is obtained by fusing terrain data in the remote sensing data and terrain data in the ground point data; based on the target data set, determining the fire risk level of the target area, including: obtaining preset weight data, and performing weighted calculation on the target weather data, the target vegetation data and the target terrain data based on the preset weight data to obtain target data, wherein the target data is used to indicate the probability of fire in the target area; and determining the fire risk level of the target area based on the target data.

[0011] Further, before data screening on the first data set to obtain a second data set, further comprising: converting remote sensing data and ground point data in the first data set into a unified format, and / or filling missing values in the first data set by using a target filling strategy, wherein the target filling strategy includes interpolation method and extrapolation method.

[0012] To achieve the above object, according to another aspect of the present application, a fire risk level prediction device is provided. The device comprises: an acquisition unit configured to acquire remote sensing data and ground point data of a target region to obtain a first data set, wherein the ground point data comprises data collected by ground monitoring stations in the target region, and data types of the remote sensing data and the ground point data include meteorological data, vegetation data and terrain data; a screening unit configured to perform data screening on the first data set to obtain a second data set, wherein the second data set is the first data set after screening; a fusion unit configured to perform data fusion on remote sensing data and ground point data in the second data set to obtain a target data set; and a determination unit configured to determine a fire risk level of the target region based on the target data set.

[0013] Further, the fusion unit comprises: a first determination subunit configured to determine a data fusion algorithm, wherein the data fusion algorithm includes a Bayesian fusion algorithm; and a fusion subunit configured to perform data fusion on remote sensing data and ground point data in the second data set based on the data fusion algorithm to obtain the target data set.

[0014] Further, the screening unit comprises: a screening subunit configured to perform data screening on the first data set based on statistical parameters of the first data set, and / or perform data screening on the first data set based on a clustering analysis strategy to obtain the second data set, wherein the statistical parameters include at least one of the following: mean value of the remote sensing data and standard deviation of the remote sensing data.

[0015] Further, the screening subunit comprises: an acquisition module configured to acquire spatial resolution of the remote sensing data; a first determination module configured to determine a target value range based on the spatial resolution of the remote sensing data, the mean value of the remote sensing data and the standard deviation of the remote sensing data, wherein the target value range is used to identify abnormal data in the point data; and a first screening module configured to screen ground point data in the first data set based on the target value range.

[0016] Further, the filtering subunit includes: an interpolation module for interpolating the ground point data in the first dataset to obtain a surface dataset, wherein the surface dataset records point data in multiple dimensions; a second determination module for determining a target distance range based on the surface dataset and the spatial resolution of the remote sensing data, wherein the target distance range is used to identify abnormal data in the surface dataset; a third determination module for determining the distance between each data point in the surface dataset and the center position of the surface dataset to obtain a target distance set; and a second filtering module for filtering the surface dataset based on the target distance set and the target distance range.

[0017] Furthermore, the target dataset includes: target meteorological data, target vegetation data, and target terrain data, wherein the target meteorological data is obtained by fusing the meteorological data in the remote sensing data and the meteorological data in the ground point data; the target vegetation data is obtained by fusing the vegetation data in the remote sensing data and the vegetation data in the ground point data; and the target terrain data is obtained by fusing the terrain data in the remote sensing data and the terrain data in the ground point data.

[0018] Further, the determining unit includes: a processing subunit, used to acquire preset weight data, and perform weighted calculations on the target meteorological data, the target vegetation data, and the target terrain data based on the preset weight data to obtain target data, wherein the target data is used to indicate the probability of a fire occurring in the target area; and a second determining subunit, used to determine the fire risk level of the target area based on the target data.

[0019] Furthermore, the fire risk level prediction device also includes: a processing unit, used to convert the remote sensing data and ground point data in the first dataset into a unified format before filtering the first dataset to obtain the second dataset, and / or to fill the missing values ​​in the first dataset using a target filling strategy, wherein the target filling strategy includes: interpolation method and extrapolation method.

[0020] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is executed, it controls the device where the computer-readable storage medium is located to perform the fire hazard level prediction method.

[0021] According to another aspect of this application, an electronic device is provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the fire risk level prediction method when it runs.

[0022] According to another aspect of this application, a computer program product is provided, including computer instructions that, when executed by a processor, implement the steps of the fire hazard level prediction method.

[0023] In this application, remote sensing data and ground point data of the target area are collected to obtain a first dataset. The ground point data includes data collected by ground monitoring stations within the target area. The data types of the remote sensing data and ground point data include meteorological data, vegetation data, and topographic data. The first dataset is then filtered to obtain a second dataset, which is the filtered version of the first dataset. The remote sensing data and ground point data in the second dataset are then fused to obtain a target dataset. Based on the target dataset, the fire risk level of the target area is determined, thus solving the technical problem in related technologies where using single-scale data to predict fire risk levels leads to inaccurate predictions. In this application, by filtering and fusing small-scale ground point data and large-scale remote sensing data, the fire risk level of the target area is determined based on the fused data. This avoids the inaccurate predictions that occur in related technologies where fire risk levels are predicted based solely on single-scale data, thereby improving the accuracy of fire risk level predictions. Attached Figure Description

[0024] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0025] Figure 1 A hardware block diagram of a computer terminal for implementing a fire hazard level prediction method is shown.

[0026] Figure 2 This is a flowchart of a fire risk level prediction method provided according to an embodiment of this application;

[0027] Figure 3 This is a schematic diagram of the fire risk level prediction process provided according to the embodiments of this application;

[0028] Figure 4 This is a schematic diagram of a fire risk level prediction device provided according to an embodiment of this application;

[0029] Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0032] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse. For example, interfaces are set up between this system and relevant users or organizations, providing users with corresponding access points to choose to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0033] Example 1

[0034] According to an embodiment of this application, a method embodiment for predicting fire risk level is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0035] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1A hardware block diagram of a computer terminal (or mobile device) for implementing a fire hazard level prediction method is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102b, ..., 102n in the figure) (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPG, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0036] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0037] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the fire risk level prediction method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned fire risk level prediction method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0038] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0039] The display can be, for example, a touchscreen liquid crystal display (LCD), which allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0040] Under the aforementioned operating environment, this application provides the following: Figure 2 The method for predicting fire hazard levels is shown. Figure 2 This is a flowchart of a fire risk level prediction method according to Embodiment 1 of this application.

[0041] Step S201: Collect remote sensing data and ground point data of the target area to obtain the first dataset. The ground point data includes data collected by ground monitoring stations in the target area. The data types of remote sensing data and ground point data include meteorological data, vegetation data and terrain data.

[0042] The first dataset mentioned above includes remote sensing data and ground point data of the target area. When collecting remote sensing data and ground point data of the target area, topographic data such as slope, digital elevation, and aspect (i.e., topographic factors) can be collected. Key parameters affecting the fire probability of the target area can also be collected, such as forest type, forest canopy density, tree height, biomass, combustible material type, combustible material moisture content, temperature, vegetation drought index, etc. (i.e., vegetation factors). Meteorological data such as temperature, humidity, wind speed, and wind direction (i.e., meteorological factors) can also be collected to ensure that fire risk assessment can be based on the most accurate and up-to-date information. Among them, all meteorological factors and vegetation factors such as temperature and vegetation drought index data can be obtained from ground monitoring station points and remote sensing methods. It should be noted that the data obtained from ground monitoring station points can be called ground point data (or simply point data), and the data obtained from remote sensing methods can be called remote sensing data.

[0043] Step S202: Filter the first dataset to obtain the second dataset, where the second dataset is the first dataset after filtering.

[0044] For example, data screening criteria can be defined, such as data timeliness, spatial scale, spatial resolution, and scope of application, to ensure that the selected data reflects the latest forest underlying surface and meteorological conditions, enabling the prediction model to respond promptly to real-time changes. High-resolution data that meets the requirements can be selected to capture subtle changes in the above-mentioned characteristics, especially in small-scale areas prone to fire. Validated data sources can be chosen to ensure data reliability and reduce prediction errors. In this embodiment, invalid or low-quality data that does not meet the criteria can also be removed. For example, statistical parameters in the first dataset, such as mean, median, and standard deviation, can be determined first. Then, thresholds are set based on these statistical parameters to identify possible outliers, which are then removed from the first dataset. Cluster analysis can also be used to remove outliers from the dataset. By screening the first dataset, the purpose of optimizing the first dataset is achieved, and the screened first dataset can be called the second dataset.

[0045] Step S203: Perform data fusion on the remote sensing data and ground location data in the second dataset to obtain the target dataset.

[0046] In this embodiment, a suitable fusion algorithm can be selected according to the characteristics of the data and the fusion objective. For example, the Bayesian method can be used to fuse remote sensing data and ground point data in the second dataset to obtain the target dataset, thus achieving the purpose of data fusion of large-scale remote sensing data and small-scale ground point data.

[0047] Step S204: Determine the fire risk level of the target area based on the target dataset.

[0048] The aforementioned target dataset includes: target meteorological data, target vegetation data, and target terrain data. The target meteorological data is obtained by fusing meteorological data from remote sensing data and meteorological data from ground point data. The target vegetation data is obtained by fusing vegetation data from remote sensing data and vegetation data from ground point data. The target terrain data is obtained by fusing terrain data from remote sensing data and terrain data from ground point data. The fire risk prediction probability can be calculated based on the proportion parameters (e.g., weights) of the target meteorological data, target vegetation data, and target terrain data. Then, the fire risk level of the target area can be determined based on the fire risk prediction probability.

[0049] Through the above steps, in this embodiment, by filtering and fusing small-scale ground point data and large-scale remote sensing data, the fire risk level of the target area is determined based on the fused data. This avoids the inaccurate prediction results caused by relying solely on data of a single scale in related technologies, thus improving the accuracy of fire risk level prediction. Furthermore, it solves the technical problem of inaccurate fire risk level predictions caused by using data of a single scale in related technologies.

[0050] Optionally, in the fire risk level prediction method provided in this application embodiment, data fusion is performed on the remote sensing data and ground location data in the second dataset to obtain the target dataset, including: determining a data fusion algorithm, wherein the data fusion algorithm includes: a Bayesian fusion algorithm; and based on the data fusion algorithm, data fusion is performed on the remote sensing data and ground location data in the second dataset to obtain the target dataset.

[0051] Among various data fusion algorithms, the Bayesian fusion algorithm is widely used because it can handle the uncertainty in data. Based on Bayes' theorem, the Bayesian fusion algorithm can combine prior knowledge (such as the macro fire risk trend reflected by remote sensing data) with newly observed information (such as the local detailed conditions provided by point data) to calculate the posterior probability, so as to determine the probability of the fire risk level of a certain area after considering all information.

[0052] For prior probabilities in Bayesian fusion algorithms: Remote sensing data typically provides estimates of the fire risk level of a study area, forming the basis of prior probabilities. For example, remote sensing imagery might show low vegetation moisture content and dry vegetation type in a certain area, indicating a high fire risk level. For likelihood probabilities in Bayesian fusion algorithms: Ground point data provides detailed information for specific locations, such as real-time temperature, humidity, and wind speed. Within the Bayesian framework, the probability of observing these specific point conditions given a fire risk level can be calculated, i.e., the likelihood probability. For marginal probabilities in Bayesian fusion algorithms: Marginal probabilities refer to the probability of the joint occurrence of remote sensing data and point data, typically used to normalize posterior probabilities to ensure that the sum of the probabilities of all possible fire risk levels is 1. For posterior probabilities in Bayesian fusion algorithms: By combining the prior probabilities of remote sensing data, the likelihood probabilities of point data, and marginal probabilities, the posterior probability of the fused data for the fire risk level can be calculated using Bayes' theorem, reflecting the updated estimate of the fire risk level of a certain area after comprehensively considering all information.

[0053] In this embodiment, after fusing the remote sensing data and ground location data in the second dataset using the aforementioned Bayesian fusion algorithm, a comprehensive fire risk level dataset, namely the target dataset, is obtained. This target dataset is more reliable and accurate than the original remote sensing data or location data because it comprehensively considers the data characteristics of large-scale remote sensing data and small-scale ground location data, thus overcoming the limitations of a single data source.

[0054] In this embodiment, a fusion strategy (e.g., a data fusion algorithm) can be applied to combine remote sensing data with ground location data, ensuring that the fused data reflects the fire risk level characteristics at different scales. During the fusion process, the characteristics of data at different scales can be considered to ensure that the fused data can capture both large-scale macro trends and small-scale micro details.

[0055] Based on the characteristics of the data and the fusion objective, a suitable fusion algorithm can be selected, such as the Bayesian method (i.e., the Bayesian fusion algorithm). The dataset obtained by interpolating remote sensing data and ground point data into a surface is expressed as follows: (Where Bm can represent remote sensing data and ground point data of a certain ground location), the fusion probability of the target expression can be calculated by Bayesian data fusion method, as shown in formula (1).

[0056] (1)

[0057] in, Interpolate an extended dataset from known self-remote sensing data and surrounding data from different points. Features in the case (Representing the posterior probability of the occurrence of the pixel value after fusing remote sensing image data and ground point data) Known In the event of The likelihood probability of occurrence for The prior probability of occurrence, for The marginal probability of occurrence.

[0058] In this embodiment, a Bayesian fusion algorithm is used to fuse small-scale ground point data and large-scale remote sensing data, eliminating the edge differences between different data sources and making the fused data transition smoothly in spatial scale, so as to improve the accuracy of fire risk level prediction results.

[0059] Optionally, in the fire risk level prediction method provided in this application embodiment, the first dataset is filtered to obtain the second dataset, including: filtering the first dataset based on statistical parameters of the first dataset, and / or filtering the first dataset based on a clustering analysis strategy to obtain the second dataset, wherein the statistical parameters include at least one of the following: the mean of remote sensing data, the standard deviation of remote sensing data.

[0060] In this embodiment, outliers in the first dataset can be removed based on a standard deviation elimination method. For example, if a data point deviates from the mean by more than a set multiple of the standard deviation, it is considered an outlier and removed. In an optional example, outliers in the first dataset can also be removed based on a clustering analysis strategy. For example, ground point data obtained from a certain ground monitoring station can be interpolated to form surface data. Then, outliers obtained after interpolation can be identified and removed using a clustering algorithm (i.e., a clustering analysis strategy), avoiding the introduction of new inconsistencies due to outliers. This achieves the goal of optimizing the data filtering process, facilitating high-quality fusion of remote sensing "surface" data and station "point" data, thereby significantly improving the accuracy of fire risk level prediction.

[0061] Optionally, in the fire risk level prediction method provided in this application embodiment, data filtering of the first dataset is performed based on the statistical parameters of the first dataset, including: obtaining the spatial resolution of the remote sensing data; determining a target numerical range based on the spatial resolution of the remote sensing data, the average value of the remote sensing data, and the standard deviation of the remote sensing data, wherein the target numerical range is used to identify abnormal data in the point data; and filtering the ground point data in the first dataset based on the target numerical range.

[0062] In this embodiment, if the deviation of a data point from the mean exceeds a set multiple of the standard deviation (i.e., it is not within the target value range), it is considered an outlier and is removed. In this embodiment, outliers can be identified using formulas (2), (3), and (4).

[0063] (2)

[0064] (3)

[0065] (4)

[0066] in, For the spatial resolution of remote sensing data, These are the parameters obtained from calculating the spatial resolution. Data for locations to be analyzed. This represents the average value of the remote sensing data. This represents the standard deviation of the remote sensing data.

[0067] If the data of the points to be analyzed and If the data of the point to be analyzed is within the target value range, it is considered a normal value; otherwise, it is considered an outlier and will be removed.

[0068] Optionally, in the fire risk level prediction method provided in this application embodiment, data filtering of the first dataset based on a clustering analysis strategy includes: interpolating the ground point data in the first dataset to obtain a surface dataset, wherein the surface dataset records point data in multiple dimensions; determining a target distance range based on the spatial resolution of the surface dataset and remote sensing data, wherein the target distance range is used to identify abnormal data in the surface dataset; determining the distance between each data point in the surface dataset and the center position of the surface dataset to obtain a target distance set; and filtering the surface dataset based on the target distance set and the target distance range.

[0069] The aforementioned surface dataset may include two-dimensional point data. In this embodiment, the point data obtained by the ground monitoring station is interpolated to form a surface dataset. Then, based on the elimination method of cluster analysis, the outliers obtained after interpolation are identified and eliminated by the clustering algorithm to avoid introducing new inconsistencies due to outliers.

[0070] (5)

[0071] (6)

[0072] (7)

[0073] Where scale represents the spatial resolution of the remote sensing data. and These represent the location of the data points to be analyzed and the center location of the area dataset, respectively. This is the average distance of multidimensional data (e.g., the average location of all points to be analyzed in a surface dataset). and Distances in different single dimensions (e.g., distances in a single dimension of two-dimensional data). This is the average value. The data of the points to be analyzed and the location of the center point. If the data of a certain point to be analyzed satisfies the formula (7) (corresponding to the target distance range), then the data of the point to be analyzed is a normal value and is retained; otherwise, it is an abnormal value and is removed. This avoids the situation in related technologies where data is filtered by directly setting thresholds, resulting in inaccurate filtering results.

[0074] Optionally, in the fire risk level prediction method provided in this application embodiment, the target dataset includes: target meteorological data, target vegetation data, and target terrain data. The target meteorological data is obtained by fusing meteorological data from remote sensing data and meteorological data from ground point data. The target vegetation data is obtained by fusing vegetation data from remote sensing data and vegetation data from ground point data. The target terrain data is obtained by fusing terrain data from remote sensing data and terrain data from ground point data. Based on the target dataset, determining the fire risk level of the target area includes: obtaining preset weight data; performing weighted calculations on the target meteorological data, target vegetation data, and target terrain data based on the preset weight data to obtain target data, wherein the target data is used to indicate the probability of a fire occurring in the target area; and determining the fire risk level of the target area based on the target data.

[0075] Table 1 records the specific parameter information of an optional meteorological factor (corresponding to meteorological data), vegetation factor (corresponding to vegetation data), and topographic factor (corresponding to topographic data). The process of determining the fire risk level of the target area is explained below with reference to Table 1.

[0076] Table 1

[0077]

[0078] First, a fire risk prediction model (hereinafter referred to as the prediction model) can be constructed based on the processed factors (i.e., target meteorological data, target vegetation data and target terrain data) and the obtained factor proportions (corresponding to the preset weight data), as shown in formula (8).

[0079] (8)

[0080] Wherein, FRID is the fire risk prediction probability reference value (corresponding to the target data), A is the meteorological factor, B is the vegetation factor, C is the topographic factor, and a, b, and c are the proportion parameters of each factor.

[0081] in:

[0082] (9)

[0083] (10)

[0084] (11)

[0085] In this example, adding the suffix "level" after the corresponding parameter of the input parameter indicates the parameter value after fusion. For example, Txlevel represents the daily maximum temperature after fusing the daily maximum temperature in remote sensing data and the daily maximum temperature in ground point data. Input parameters without the suffix "level" represent the original parameter values. In this embodiment, the prediction model (FRID) and forest fire weather level can be combined to classify the fire risk level of different regions based on the prediction results.

[0086] (12)

[0087] In this embodiment, the fire risk prediction probability reference value (FRID) level can be defined for different regions based on the fire risk prediction probability level of the forest fire weather level. For example, based on factors such as terrain, climate conditions and historical fire distribution, the forest fire weather level can be divided into five levels from low to high: low fire risk (level 1), relatively low fire risk (level 2), relatively high fire risk (level 3), high fire risk (level 4), and extremely high fire risk (level 5). By combining topographic factors, meteorological factors and vegetation factors, the fire risk level can be predicted, thus achieving the goal of accurately predicting the fire risk level.

[0088] In one alternative example, a fire risk level distribution map can also be constructed based on the predicted fire risk levels for different regions.

[0089] Optionally, in the fire risk level prediction method provided in the embodiments of this application, before filtering the first dataset to obtain the second dataset, the method further includes: converting the remote sensing data and ground point data in the first dataset into a unified format, and / or filling the missing values ​​in the first dataset with a target filling strategy, wherein the target filling strategy includes: interpolation method and extrapolation method.

[0090] In this embodiment, the original data (i.e., the data in the first dataset) can also be normalized to handle missing values. Data from different sources and types are converted into a uniform format, and missing values ​​are handled by interpolation or extrapolation, thereby providing a set of accurate, complete, and consistent input parameters for the fire risk rating model. In an optional example, the data can also be classified and evaluated; the data can be classified according to its source (satellite remote sensing and ground point monitoring) and type (meteorological data, vegetation data, and topographic information).

[0091] Figure 3 This is a schematic diagram of the fire risk level prediction process provided in the embodiments of this application, such as... Figure 3As shown, the process includes: data input (i.e., inputting meteorological factors, vegetation factors, and topographic factors). Meteorological factors include: daily maximum temperature, consecutive days without precipitation, daily maximum wind speed, and daily minimum relative humidity. Vegetation factors include: forest type, forest canopy density, tree height, biomass, and temperature-vegetation drought index. Topographic factors can include: slope, aspect, and digital elevation. Data preprocessing processes such as data filtering, data removal, and data fusion can then be performed. After data preprocessing, the fire risk prediction probability value can be predicted, followed by modeling and validation to obtain a prediction model. Finally, the prediction model can be used to predict the fire risk probability value and determine the fire risk level.

[0092] In this embodiment, optimized data filtering and preprocessing steps ensure high-quality fusion input of remote sensing "area" data and station "point" data, thereby significantly improving the accuracy of fire risk level prediction. The data fusion algorithm enables rapid data processing and integration, allowing the fire risk prediction model to respond in real-time to changes in fire risk levels, providing timely decision support for forest fire prevention and control. This data fusion method also eliminates edge differences between different data sources, resulting in a smooth transition in spatial scale of the fused data and providing a more intuitive and easily understandable fire risk level distribution map. In this embodiment, by providing high-quality data fusion results and an intuitive fire risk level distribution map, the user experience is improved, enabling monitoring personnel to understand and assess fire risk levels more clearly and accurately.

[0093] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0094] Example 2

[0095] This application also provides a fire hazard level prediction device. It should be noted that the fire hazard level prediction device of this application can be used to execute the fire hazard level prediction method provided in this application. The following describes the fire hazard level prediction device provided in this application.

[0096] According to an embodiment of this application, an apparatus for implementing the above-described fire hazard level prediction method is also provided, such as... Figure 4 As shown, the device includes: a data acquisition unit 41, a filtering unit 42, a fusion unit 43, and a determination unit 44.

[0097] The acquisition unit 41 is used to acquire remote sensing data and ground point data of the target area to obtain the first dataset. The ground point data includes data collected by ground monitoring stations within the target area. The data types of the remote sensing data and ground point data include meteorological data, vegetation data and topographic data.

[0098] The filtering unit 42 is used to filter the first dataset to obtain the second dataset, wherein the second dataset is the first dataset after filtering.

[0099] Fusion unit 43 is used to fuse remote sensing data and ground point data in the second dataset to obtain the target dataset;

[0100] Unit 44 is used to determine the fire risk level of the target area based on the target dataset.

[0101] In the fire risk level prediction device provided in this embodiment of the invention, a data acquisition unit 41 is used to acquire remote sensing data and ground point data of the target area to obtain a first dataset. The ground point data includes data collected by ground monitoring stations within the target area. The data types of the remote sensing data and ground point data include meteorological data, vegetation data, and topographic data. A filtering unit 42 filters the first dataset to obtain a second dataset, which is the filtered first dataset. A fusion unit 43 fuses the remote sensing data and ground point data in the second dataset to obtain a target dataset. A determination unit 44 determines the fire risk level of the target area based on the target dataset. This solves the technical problem in related technologies where using single-scale data to predict fire risk levels leads to inaccurate prediction results. In this embodiment, by filtering and fusing small-scale ground point data and large-scale remote sensing data, the fire risk level of the target area is determined based on the fused data. This avoids the inaccurate prediction results caused by relying solely on single-scale data in related technologies, thereby improving the accuracy of fire risk level prediction.

[0102] Optionally, in the fire risk level prediction device provided in this application embodiment, the fusion unit includes: a first determining subunit, used to determine a data fusion algorithm, wherein the data fusion algorithm includes: a Bayesian fusion algorithm; and a fusion subunit, used to perform data fusion on the remote sensing data and ground point data in the second dataset based on the data fusion algorithm to obtain a target dataset.

[0103] Optionally, in the fire risk level prediction device provided in the embodiments of this application, the screening unit includes: a screening subunit, used to screen the first dataset based on statistical parameters of the first dataset, and / or to screen the first dataset based on a clustering analysis strategy to obtain a second dataset, wherein the statistical parameters include at least one of the following: the mean value of remote sensing data, the standard deviation of remote sensing data.

[0104] Optionally, in the fire risk level prediction device provided in this application embodiment, the screening subunit includes: an acquisition module for acquiring the spatial resolution of remote sensing data; a first determination module for determining a target numerical range based on the spatial resolution of the remote sensing data, the average value of the remote sensing data, and the standard deviation of the remote sensing data, wherein the target numerical range is used to identify abnormal data in the point data; and a first screening module for screening the ground point data in the first dataset based on the target numerical range.

[0105] Optionally, in the fire risk level prediction device provided in this application embodiment, the screening subunit includes: an interpolation module, used to interpolate the ground point data in the first dataset to obtain a surface dataset, wherein the surface dataset records point data in multiple dimensions; a second determination module, used to determine a target distance range based on the spatial resolution of the surface dataset and remote sensing data, wherein the target distance range is used to identify abnormal data in the surface dataset; a third determination module, used to determine the distance between each data in the surface dataset and the center position of the surface dataset to obtain a target distance set; and a second screening module, used to screen the surface dataset based on the target distance set and the target distance range.

[0106] Optionally, in the fire risk level prediction device provided in this application embodiment, the target dataset includes: target meteorological data, target vegetation data, and target terrain data. The target meteorological data is obtained by fusing meteorological data from remote sensing data and meteorological data from ground point data. The target vegetation data is obtained by fusing vegetation data from remote sensing data and vegetation data from ground point data. The target terrain data is obtained by fusing terrain data from remote sensing data and terrain data from ground point data. The determining unit includes: a processing subunit, used to acquire preset weight data, and perform weighted calculations on the target meteorological data, target vegetation data, and target terrain data based on the preset weight data to obtain target data, wherein the target data is used to indicate the probability of a fire occurring in the target area; and a second determining subunit, used to determine the fire risk level of the target area based on the target data.

[0107] Optionally, in the fire risk level prediction device provided in the embodiments of this application, the fire risk level prediction device further includes: a processing unit, used to convert the remote sensing data and ground point data in the first dataset into a unified format before filtering the first dataset to obtain the second dataset, and / or to fill the missing values ​​in the first dataset using a target filling strategy, wherein the target filling strategy includes: interpolation method and extrapolation method.

[0108] It should be noted that the acquisition unit 41, filtering unit 42, fusion unit 43, and determination unit 44 mentioned above correspond to steps S201 to S204 in Embodiment 1. Each unit and its corresponding step implements the same instance and application scenario, but is not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processor 102 fire hazard level prediction, 102b, ..., 102n). The above modules can also be part of the device and run in the computer terminal 10 provided in Embodiment 1.

[0109] Example 3

[0110] Embodiments of this application may provide an electronic device. Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 5 As shown, the electronic device may include: one or more ( Figure 5 (Only one is shown) processor 502, memory 504, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0111] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0112] The processor can access information and applications stored in the memory via a transmission device to perform the following steps: Collect remote sensing data and ground point data of the target area to obtain a first dataset, wherein the ground point data includes data collected by ground monitoring stations within the target area, and the data types of the remote sensing data and ground point data include meteorological data, vegetation data, and topographic data; filter the first dataset to obtain a second dataset, wherein the second dataset is the filtered first dataset; fuse the remote sensing data and ground point data in the second dataset to obtain a target dataset; and determine the fire risk level of the target area based on the target dataset.

[0113] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: fusing remote sensing data and ground point data in the second dataset to obtain a target dataset, including: determining a data fusion algorithm, wherein the data fusion algorithm includes: a Bayesian fusion algorithm; and fusing remote sensing data and ground point data in the second dataset based on the data fusion algorithm to obtain a target dataset.

[0114] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: filtering the first dataset to obtain the second dataset, including: filtering the first dataset based on the statistical parameters of the first dataset, and / or filtering the first dataset based on the clustering analysis strategy to obtain the second dataset, wherein the statistical parameters include at least one of the following: the mean of the remote sensing data, the standard deviation of the remote sensing data.

[0115] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: filtering the first dataset based on the statistical parameters of the first dataset, including: obtaining the spatial resolution of the remote sensing data; determining the target numerical range based on the spatial resolution of the remote sensing data, the average value of the remote sensing data, and the standard deviation of the remote sensing data, wherein the target numerical range is used to identify abnormal data in the point data; and filtering the ground point data in the first dataset based on the target numerical range.

[0116] The processor can also access information and applications stored in the memory via a transmission device to perform the following steps: filtering data in the first dataset based on a clustering analysis strategy, including: interpolating the ground point data in the first dataset to obtain a surface dataset, wherein the surface dataset records point data in multiple dimensions; determining a target distance range based on the spatial resolution of the surface dataset and remote sensing data, wherein the target distance range is used to identify abnormal data in the surface dataset; determining the distance between each data point in the surface dataset and the center position of the surface dataset to obtain a target distance set; and filtering data in the surface dataset based on the target distance set and the target distance range.

[0117] The processor can also invoke information and applications stored in the memory via a transmission device to execute the following steps: The target dataset includes: target meteorological data, target vegetation data, and target terrain data, wherein the target meteorological data is obtained by fusing meteorological data from remote sensing data and meteorological data from ground point data; the target vegetation data is obtained by fusing vegetation data from remote sensing data and vegetation data from ground point data; and the target terrain data is obtained by fusing terrain data from remote sensing data and terrain data from ground point data. Based on the target dataset, the fire risk level of the target area is determined, including: acquiring preset weight data; performing weighted calculations on the target meteorological data, target vegetation data, and target terrain data based on the preset weight data to obtain target data, wherein the target data is used to indicate the probability of a fire occurring in the target area; and determining the fire risk level of the target area based on the target data.

[0118] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: before filtering the first dataset to obtain the second dataset, it also includes: converting the remote sensing data and ground point data in the first dataset into a unified format, and / or filling the missing values ​​in the first dataset with a target filling strategy, wherein the target filling strategy includes: interpolation method and extrapolation method.

[0119] By filtering and fusing small-scale ground point data and large-scale remote sensing data, the fire risk level of the target area is determined based on the fused data. This avoids the inaccurate prediction results caused by relying solely on data of a single scale in related technologies, thus improving the accuracy of fire risk level prediction. Furthermore, the embodiments of this application solve the technical problem of inaccurate fire risk level predictions caused by using data of a single scale in related technologies.

[0120] Those skilled in the art will understand that Figure 5The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 5 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 5 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 5 The different configurations shown.

[0121] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0122] Example 4

[0123] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the fire hazard level prediction method provided in Embodiment 1.

[0124] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0125] This application also provides a computer program product that, when executed on a data processing device, is suitable for performing steps of a fire hazard rating prediction method.

[0126] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0127] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0128] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0130] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0131] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0132] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for predicting fire risk levels, characterized in that, include: Remote sensing data and ground point data of the target area are collected to obtain a first dataset, wherein the ground point data includes data collected by ground monitoring stations in the target area, and the data types of the remote sensing data and the ground point data include meteorological data, vegetation data and topographic data; The first dataset is filtered to obtain a second dataset, wherein the second dataset is the first dataset after filtering; The remote sensing data and ground location data in the second dataset are fused to obtain the target dataset; Based on the target dataset, the fire risk level of the target area is determined.

2. The fire risk level prediction method according to claim 1, characterized in that, The remote sensing data and ground location data in the second dataset are fused to obtain the target dataset, which includes: A data fusion algorithm is determined, wherein the data fusion algorithm includes: a Bayesian fusion algorithm; Based on the data fusion algorithm, the remote sensing data and ground location data in the second dataset are fused to obtain the target dataset.

3. The fire risk level prediction method according to claim 1, characterized in that, The first dataset is filtered to obtain the second dataset, which includes: The first dataset is filtered based on statistical parameters of the first dataset, and / or the first dataset is filtered based on a clustering analysis strategy to obtain the second dataset, wherein the statistical parameters include at least one of the following: the mean of the remote sensing data, the standard deviation of the remote sensing data.

4. The fire risk level prediction method according to claim 3, characterized in that, Data filtering of the first dataset based on statistical parameters of the first dataset includes: Obtain the spatial resolution of the remote sensing data; Based on the spatial resolution of the remote sensing data, the average value of the remote sensing data, and the standard deviation of the remote sensing data, a target numerical range is determined, wherein the target numerical range is used to identify abnormal data in the point data; The ground location data in the first dataset are filtered based on the target numerical range.

5. The fire risk level prediction method according to claim 3, characterized in that, The first dataset is filtered based on a clustering analysis strategy, including: Interpolate the ground point data in the first dataset to obtain a surface dataset, wherein the surface dataset records point data in multiple dimensions; Based on the spatial resolution of the area dataset and the remote sensing data, a target distance range is determined, wherein the target distance range is used to identify abnormal data in the area dataset; Determine the distance between each data point in the polygon dataset and the center position of the polygon dataset to obtain the target distance set; Based on the target distance set and the target distance range, the surface dataset is filtered.

6. The fire risk level prediction method according to claim 1, characterized in that, The target dataset includes: target meteorological data, target vegetation data, and target terrain data. The target meteorological data is obtained by fusing meteorological data from the remote sensing data and meteorological data from the ground location data. The target vegetation data is obtained by fusing vegetation data from the remote sensing data and vegetation data from the ground location data. The target terrain data is obtained by fusing terrain data from the remote sensing data and terrain data from the ground location data. Based on the target dataset, the fire risk level of the target area is determined, including: Obtain preset weight data, and perform weighted calculations on the target meteorological data, the target vegetation data, and the target terrain data based on the preset weight data to obtain target data, wherein the target data is used to indicate the probability of a fire occurring in the target area; Based on the target data, the fire risk level of the target area is determined.

7. The fire risk level prediction method according to claim 1, characterized in that, Before filtering the first dataset to obtain the second dataset, the process also includes: The remote sensing data and ground point data in the first dataset are converted into a unified format, and / or, the missing values ​​in the first dataset are filled using a target filling strategy, wherein the target filling strategy includes: interpolation method and extrapolation method.

8. A fire hazard level prediction device, characterized in that, include: The acquisition unit is used to acquire remote sensing data and ground point data of the target area to obtain a first dataset. The ground point data includes data collected by ground monitoring stations in the target area. The data types of the remote sensing data and the ground point data include meteorological data, vegetation data and terrain data. A filtering unit is used to filter the first dataset to obtain a second dataset, wherein the second dataset is the first dataset after filtering; The fusion unit is used to fuse the remote sensing data and ground point data in the second dataset to obtain the target dataset. The determining unit is used to determine the fire risk level of the target area based on the target dataset.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device containing the computer-readable storage medium to perform the fire hazard level prediction method according to any one of claims 1 to 7.

10. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the fire risk level prediction method according to any one of claims 1 to 7.

Images