Method and device for determining ignition result of forest fire area and electronic equipment
By extracting the characteristics of wildfire data and calculating the fire index using the fire detection model, the problem of low accuracy of wildfire monitoring in the existing technology is solved, and more efficient and accurate wildfire monitoring is achieved.
Patent Information
- Application Number
- CN202411302634.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2025-03-04
AI Technical Summary
In the prior art, when monitoring wildfire events, the accuracy of the monitoring results is low, and this problem cannot be effectively solved.
By obtaining the target wildfire data of the target wildfire area, extracting its target wildfire characteristics, and determining the first wildfire characteristics and the second wildfire characteristics from it. The corresponding ignition index is calculated using the ignition detection model, and then determine whether the target wildfire area is on fire.
It improves the accuracy and efficiency of wildfire monitoring, and can more accurately determine whether the wildfire area is on fire, thereby improving fire management and emergency response.
Smart Images
Figure CN119557755A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wildfire monitoring, and in particular to a method, device and electronic equipment for determining a fire result in a wildfire area. Background Art
[0002] At present, wildfire is a natural disaster. After a wildfire occurs, the fire can easily spread to the vicinity of the overhead line corridor, causing the transmission lines to be forced to shut down. At the same time, due to the long duration of the wildfire, it may cause accidents such as line tripping and power outages, which seriously threaten the normal operation of the power company and the normal production of the national economy.
[0003] Therefore, in the related technology, when the related methods are used to monitor wildfire events, there is a technical problem that the accuracy of the monitoring results of wildfire events is low.
[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention
[0005] The embodiments of the present invention provide a method, device and electronic device for determining the ignition result of a wildfire area, so as to at least solve the technical problem in the related art that the accuracy of the monitoring results of wildfire events is low when monitoring wildfire events.
[0006] According to one aspect of an embodiment of the present invention, a method for determining a fire result in a wildfire area is provided, comprising: acquiring target wildfire data of a target wildfire area; extracting target wildfire features of the target wildfire data; determining a first wildfire feature and a second wildfire feature from the target wildfire features, wherein the first wildfire feature represents a wildfire feature whose corresponding fire feature index is greater than or equal to a first threshold, and the second wildfire feature represents a wildfire feature whose corresponding fire feature index is less than the first threshold; determining a first fire index corresponding to the first wildfire feature; when the first fire index is less than the second threshold, determining a second fire index corresponding to the second wildfire feature; and determining a fire result of whether the target wildfire area is on fire based on the second fire index.
[0007] Optionally, determining the first fire index corresponding to the first wildfire characteristic includes: inputting the first wildfire characteristic into a fire detection model to obtain the first fire index corresponding to the first wildfire characteristic, wherein the fire detection model is obtained by training an initial model with sample data, and the sample data includes: sample wildfire characteristics, and sample fire indices corresponding to the sample wildfire characteristics.
[0008] Optionally, the step of inputting the first wildfire feature into a fire detection model to obtain the first fire index corresponding to the first wildfire feature includes: when the target wildfire data includes image data, inputting the first wildfire feature into an identification frame module of the fire detection model to obtain multiple categories of initial feature maps, wherein the multiple categories of initial feature maps are feature maps identified by identification frames of different sizes; inputting the multiple categories of initial feature maps into a feature map processing module of the fire detection model, processing the feature map dimensions corresponding to the multiple categories of initial feature maps, respectively, to obtain multiple categories of target feature maps, wherein the multiple categories of target feature maps are feature maps with the same corresponding feature map dimensions; inputting the multiple categories of target feature maps into a fire index determination module of the fire detection model to obtain the first fire index.
[0009] Optionally, the method also includes: training the sample data in the following manner to obtain the fire detection model: when the target wildfire data includes image data, obtaining an initial model, and constructing a loss function for model training, wherein the loss function includes a first loss function term, a second loss function term, a third loss function term, a fourth loss function term and a fifth loss function term, the first loss function term represents the offset between the actual fire position coordinates and the predicted fire position coordinates, the second loss function term represents the offset between the size corresponding to the predicted identification box and the size corresponding to the real identification box, the third loss function term represents the offset between the confidence corresponding to the predicted identification box and the confidence corresponding to the real identification box, the fourth loss function term represents the offset between the confidence corresponding to the predicted background box and the confidence corresponding to the real background box, and the fifth loss function term represents the offset between the category probability corresponding to the predicted identification box and the category probability corresponding to the real identification box; based on the loss function, the sample data is used to train the initial model to obtain the fire detection model.
[0010] Optionally, determining the first wildfire characteristic and the second wildfire characteristic from the target wildfire characteristic includes: obtaining geographic location information of the target wildfire characteristic; determining a fire characteristic index corresponding to the target wildfire area based on the geographic location information; determining an indicator weight corresponding to the fire characteristic index; determining a fire characteristic index corresponding to the target wildfire characteristic based on the fire characteristic index and the corresponding indicator weight; and determining that a wildfire characteristic whose corresponding fire characteristic index is greater than the first threshold is the first wildfire characteristic, and a wildfire characteristic whose corresponding fire characteristic index is less than the first threshold is the second wildfire characteristic.
[0011] Optionally, after determining the first fire index corresponding to the first wildfire feature, the method further includes: determining a fire result of the target wildfire area when the first fire index is greater than or equal to the second threshold.
[0012] According to one aspect of an embodiment of the present invention, a device for determining a fire result in a wildfire area is provided, comprising: an acquisition module for acquiring target wildfire data of a target wildfire area; an extraction module for extracting target wildfire features of the target wildfire data; a first determination module for determining a first wildfire feature and a second wildfire feature from the target wildfire features, wherein the first wildfire feature represents a wildfire feature whose corresponding ignition feature index is greater than or equal to a first threshold, and the second wildfire feature represents a wildfire feature whose corresponding ignition feature index is less than the first threshold; a second determination module for determining a first ignition index corresponding to the first wildfire feature; a third determination module for determining a second ignition index corresponding to the second wildfire feature when the first ignition index is less than a second threshold; and a fourth determination module for determining a ignition result of whether the target wildfire area is on fire based on the second ignition index.
[0013] According to one aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory for storing requests executable by the processor; wherein the processor is configured to execute the request to implement a method for determining a fire result in a wildfire area as described in any one of the above.
[0014] According to one aspect of an embodiment of the present invention, a computer-readable storage medium is provided, comprising: when a request in the computer-readable storage medium is executed by a processor of an electronic device, the electronic device is enabled to execute a method for determining a fire result in a wildfire area as described in any one of the above.
[0015] In an embodiment of the present invention, target wildfire data of a target wildfire area is obtained, target wildfire features of the target wildfire data are extracted, and a first wildfire feature and a second wildfire feature are determined from the target wildfire features, wherein the first wildfire feature represents a wildfire feature whose corresponding ignition feature index is greater than or equal to a first threshold value, and the second wildfire feature represents a wildfire feature whose corresponding ignition feature index is less than the first threshold value. The first ignition index corresponding to the first wildfire feature is determined, and when the first ignition index is less than the second threshold value, the second ignition index corresponding to the second wildfire feature is determined, so as to achieve the purpose of determining whether the target wildfire area is on fire based on the second ignition index. Since the target wildfire features can be extracted through the wildfire monitoring model, they can help in target detection and analysis, thereby improving the efficiency and accuracy of wildfire monitoring and prevention and control, and further solving the technical problem of low accuracy of monitoring results of wildfire events when monitoring wildfire events in related technologies. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0017] Figure 1 is a flow chart of a method for determining a fire result in a wildfire area according to an embodiment of the present invention;
[0018] Figure 2 4 is a structural block diagram of a device for determining a fire result in a wildfire area according to an embodiment of the present invention. DETAILED DESCRIPTION
[0019] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0021] Example 1
[0022] According to an embodiment of the present invention, an embodiment of a method for determining the ignition result of a wildfire area is provided. It should be noted that the steps shown in the flowchart of 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 can be executed in an order different from that shown here.
[0023] Figure 1 FIG. 4 is a flow chart of a method for determining a fire result in a wildfire area according to an embodiment of the present invention. Figure 1As shown, the method comprises the following steps:
[0024] Step S102, obtaining target wildfire data of the target wildfire area;
[0025] In step S102 provided in the present application, the target wildfire area may be a wildfire event within a specific geographical location, which may be a specific forest, nature reserve, mountain range, village or other area. The target wildfire area can be determined based on factors such as the specific geographical location, time range, wildfire scale, and severity.
[0026] The above-mentioned target wildfire data can be various data related to the target wildfire area, and the wildfire data of the target wildfire area can be obtained through a suitable data source or sensor, that is, collecting wildfire data in the target wildfire area includes: setting up a number of collection sites in the environmental monitoring area, and setting one or more sensors at each collection site to obtain wildfire data in the environmental monitoring area, specifically: obtaining the monitoring video collected by the video monitoring equipment within the regional monitoring range and the equipment status of the video sensor, and sending the acquired monitoring video and equipment status to the data acquisition interaction platform to process the equipment status information to prevent inaccurate monitoring results due to sensor damage, collect environmental perception parameters within the target monitoring range, and obtain wildfire data in the environmental monitoring area through sensors; wherein, environmental parameter data includes environmental parameters and location data.
[0027] It should be noted that the above-mentioned target wildfire data can be used to monitor the development of wildfires in real time, including the spread rate, fire intensity, etc. The target wildfire data provides an important reference for fire-fighting decisions.
[0028] Step S104, extracting target wildfire features of target wildfire data;
[0029] In step S104 provided in the present application, the target wildfire characteristics may include characteristics with obvious flames and characteristics with unclear flames. The target wildfire characteristics may be extracted and analyzed through image processing, machine learning and other technologies. The method for extracting the target wildfire characteristics is not limited here and may be customized according to specific scenarios and applications.
[0030] It should be noted that the above steps can be used to extract the target wildfire features of the target wildfire data through the wildfire resolution model. The above feature extraction method can help monitor and predict the development trend and degree of harm of wildfires, so as to further analyze and understand wildfire activities.
[0031] Step S106, determining a first wildfire feature and a second wildfire feature from the target wildfire features, wherein the first wildfire feature represents a wildfire feature whose corresponding ignition feature index is greater than or equal to a first threshold, and the second wildfire feature represents a wildfire feature whose corresponding ignition feature index is less than the first threshold;
[0032] In step S106 provided in the present application, the above-mentioned fire index can be an indicator for evaluating the potential risk of fire, which is usually calculated based on a series of fire-related factors. The specific fire index may vary according to different wildfire areas and is not limited here. It can be customized according to specific scenarios and applications.
[0033] The first threshold may be a limit value set when performing target wildfire feature division, that is, in the wildfire feature division scenario, the first threshold may be a feature index value, which is used to determine whether the feature is greater than or equal to the threshold. If the feature index is greater than or equal to the first threshold, the feature is classified as a first wildfire feature; if the feature index is less than the first threshold, the feature is classified as a second wildfire feature. The setting of the first threshold is not limited here and can be adjusted and optimized according to actual conditions.
[0034] It should be noted that the above-mentioned setting of the fire index and the first threshold can provide more refined division and analysis, which is helpful for applications in risk assessment, monitoring and early warning, data analysis and decision support. By reasonably setting the first threshold, the fire index can be better understood and applied, and the effectiveness of fire management and emergency response can be improved.
[0035] Step S108, determining a first fire index corresponding to the first wildfire characteristic;
[0036] In step S108 provided in the present application, the above-mentioned first wildfire characteristic can be determined by target wildfire data, which can represent the characteristics of a wildfire with obvious flames. The above-mentioned first ignition index can be a series of characteristic indicators used to judge the fire transmission, which can be the average value, maximum value or other statistical indicators of the ignition index corresponding to the first wildfire characteristic. It is not limited here and can be customized according to specific scenarios and applications.
[0037] It should be noted that, for the above steps, the best first ignition index can be determined by testing and analyzing known wildfire data, and adjusted and optimized according to actual conditions to obtain better division results and application effects.
[0038] Step S110, when the first ignition index is less than a second threshold, determining a second ignition index corresponding to a second wildfire feature;
[0039] In step S110 provided in the present application, the second threshold can be used to determine whether the first ignition index is less than the second threshold, which is the threshold that makes the first ignition index less than the second threshold. It is not limited here and can be customized according to specific scenarios and applications.
[0040] The second ignition index may be a series of characteristic indicators used to determine fire severity, and may be an average value, maximum value or other statistical indicator of the ignition index corresponding to the second wildfire characteristic. It is not limited here and may be customized according to specific scenarios and applications.
[0041] It should be noted that for the above steps, when the above-mentioned first ignition index is less than the second threshold value, the corresponding ignition index is calculated for the samples corresponding to the second wildfire characteristics according to the characteristics of the second wildfire characteristics, and the determination of the second ignition index also requires sufficient experiments and verification to ensure its accuracy and reliability.
[0042] Step S112, determining whether the target wildfire area is on fire according to the second fire index.
[0043] In step S112 provided in the present application, whether the target wildfire area is on fire is determined by comparing the index value with a preset threshold value based on the second fire index.
[0044] It should be noted that, for the above steps, a fire judgment threshold can be determined according to the specific range and standard of the fire index. This threshold can be set according to historical data. Then, according to the second fire characteristic determined previously and the corresponding second fire index data, the second fire index in the target fire area is obtained, and the second fire index of the target fire area is compared with the fire judgment threshold. If the second fire index is greater than or equal to the fire judgment threshold, it indicates that the target fire area is on fire. If the second fire index is less than the fire judgment threshold, it indicates that the target fire area is not on fire. According to the comparison result, the fire result of the target fire area is determined.
[0045] It should also be noted that judging whether the target wildfire area is on fire based on the size of the second fire index is helpful to understand the actual situation of the wildfire, guide emergency response and take corresponding measures to ensure the control and handling of the fire. Please set the fire judgment threshold according to the specific situation and data characteristics, and fully consider the influence of other factors to obtain accurate and reliable fire results.
[0046] Through the above steps S102-S112, the target wildfire data of the target wildfire area is obtained, the target wildfire features of the target wildfire data are extracted, and the first wildfire feature and the second wildfire feature are determined from the target wildfire features, wherein the first wildfire feature represents a wildfire feature whose corresponding ignition feature index is greater than or equal to the first threshold, and the second wildfire feature represents a wildfire feature whose corresponding ignition feature index is less than the first threshold. The first ignition index corresponding to the first wildfire feature is determined, and when the first ignition index is less than the second threshold, the second ignition index corresponding to the second wildfire feature is determined, so as to achieve the purpose of determining whether the target wildfire area is on fire based on the second ignition index. Since the target wildfire features can be extracted through the wildfire monitoring model, it can help target detection and analysis, thereby improving the efficiency and accuracy of wildfire monitoring and prevention and control, and thus solving the technical problem of low accuracy of the monitoring results of wildfire events when monitoring wildfire events in the related technology.
[0047] The above method of this embodiment is further introduced below.
[0048] As an optional embodiment, determining a first ignition index corresponding to a first wildfire characteristic includes: inputting the first wildfire characteristic into a ignition detection model to obtain a first ignition index corresponding to the first wildfire characteristic, wherein the ignition detection model is obtained by training an initial model with sample data, and the sample data includes: a sample wildfire characteristic and a sample ignition index corresponding to the sample wildfire characteristic.
[0049] In this embodiment, the process of determining the first fire index corresponding to the first wildfire feature can be implemented by using a fire detection model, where the model is trained using sample data.
[0050] The above-mentioned fire detection model can be a model for identifying and predicting wildfires. It can be based on machine learning and deep learning technology, through training algorithms and a large number of data samples, so as to predict and detect potential fire situations. The specific process is: collect a sample data set including sample wildfire features and a sample fire index corresponding to the sample wildfire features, input this sample data set into the initial model for training, obtain a fire detection model, and then input the first wildfire feature as input into the trained fire detection model. The model calculates and outputs the corresponding first fire index based on the input first wildfire feature.
[0051] It should be noted that the above-mentioned fire detection model is obtained by training the initial model with sample data. Through the fire detection model, not only the first fire index corresponding to the first wildfire characteristic can be predicted more accurately, but also early warning and rapid response can be provided, which has a positive effect on reducing the hazards and losses of fire, thereby helping to judge the severity of wildfires and the development trend of fires.
[0052] As an optional embodiment, a first wildfire feature is input into a fire detection model to obtain a first fire index corresponding to the first wildfire feature, including: when the target wildfire data includes image data, the first wildfire feature is input into an identification frame module of the fire detection model to obtain multiple categories of initial feature maps, wherein the multiple categories of initial feature maps are feature maps identified by identification frames of different sizes; the multiple categories of initial feature maps are input into a feature map processing module of the fire detection model, and the feature map dimensions corresponding to the multiple categories of initial feature maps are processed to obtain multiple categories of target feature maps, wherein the multiple categories of target feature maps are feature maps with the same dimensions as the corresponding feature maps; the multiple categories of target feature maps are input into a fire index determination module of the fire detection model to obtain a first fire index.
[0053] In this embodiment, the first wildfire feature is input into the fire detection model, and the following steps can be followed to obtain the first fire index corresponding to the first wildfire feature, specifically: confirm that the target wildfire data contains image data, which can be satellite images, drone images or other available image data; input the first wildfire feature into the identification frame module of the fire detection model. The purpose of this module is to locate and identify the flame area in the image. It can use a target detection algorithm (such as the target detection model YOLO, etc.) to generate multiple identification frames of different sizes; feature map processing module: input multiple categories of initial feature maps into the feature map processing module of the fire detection model. The goal of this module is to process multiple categories of initial feature maps to extract more advanced feature representations. Common methods include feature extraction and dimensionality reduction of convolutional neural networks (CNNs), etc., and input multiple categories of target feature maps into the fire index determination module of the fire detection model. The task of this module is to determine the fire index based on the input feature map. The specific method can be selected according to the task requirements. For example, a classification model, a regression model or other applicable methods can be used. Finally, based on the input multi-class target feature map, the model will calculate the first ignition index corresponding to the first wildfire feature, that is, the result obtained based on the training and prediction of the model.
[0054] It should be noted that through the above steps, the image data and the fire detection model can be used to input the first wildfire feature into the fire detection model, perform identification frame positioning, feature map processing and fire index determination, and finally obtain the first fire index, which helps to more accurately assess the extent and hazards of fire in wildfire monitoring and early warning.
[0055] As an optional embodiment, the sample data is trained in the following manner to obtain a fire detection model: when the target wildfire data includes image data, an initial model is obtained, and a loss function for model training is constructed, wherein the loss function includes a first loss function term, a second loss function term, a third loss function term, a fourth loss function term and a fifth loss function term, the first loss function term represents the offset between the actual fire position coordinates and the predicted fire position coordinates, the second loss function term represents the offset between the size corresponding to the predicted identification box and the size corresponding to the actual identification box, the third loss function term represents the offset between the confidence corresponding to the predicted identification box and the confidence corresponding to the actual identification box, the fourth loss function term represents the offset between the confidence corresponding to the predicted background box and the confidence corresponding to the actual background box, and the fifth loss function term represents the offset between the category probability corresponding to the predicted identification box and the category probability corresponding to the actual identification box; based on the loss function, the sample data is used to train the initial model to obtain a fire detection model.
[0056] In this embodiment, the sample data can be trained in the following manner to obtain a fire detection model, and the specific process is: collect image data and corresponding target wildfire data sets, and the target wildfire data set should include the real coordinates of the fire position in the image, the real size and confidence of the identification box, and the category probability corresponding to the identification box. According to specific needs and available resources, a suitable pre-trained model can be selected as the initial model, and the loss function is divided into five parts for construction according to needs. The first loss function term is used to calculate the offset between the real fire position coordinates and the predicted fire position coordinates; the second loss function term calculates the offset between the predicted identification box size and the real identification box size; the third loss function term calculates the offset between the predicted identification box confidence and the real identification box confidence; the fourth loss function term calculates the offset between the predicted background box confidence and the real background box confidence; the fifth loss function term calculates the offset between the predicted identification box category probability and the real identification box category probability. Based on the constructed loss function, the initial model is trained using sample data, and the training process can tune the initial model to obtain the final fire detection model.
[0057] It should be noted that by constructing these loss function items with different functions and combining them together, we can comprehensively consider information from multiple aspects and improve the comprehensive performance of the model in the fire detection task. The loss function constructed in this step can measure the difference between the predicted output of the model and the true value, which can enable the fire model to more accurately predict the position, size, confidence and category of the target, thereby improving the effect of the fire detection model.
[0058] As an optional embodiment, determining the first wildfire characteristic and the second wildfire characteristic from the target wildfire characteristic includes: obtaining geographic location information of the target wildfire characteristic; determining a fire characteristic index corresponding to the target wildfire area based on the geographic location information; determining an indicator weight corresponding to the fire characteristic index; determining a fire characteristic index corresponding to the target wildfire characteristic based on the fire characteristic index and the corresponding indicator weight; determining a wildfire characteristic whose corresponding fire characteristic index is greater than a first threshold as the first wildfire characteristic, and determining a wildfire characteristic whose corresponding fire characteristic index is less than the first threshold as the second wildfire characteristic.
[0059] In this embodiment, the geographic location information of the above-mentioned target wildfire characteristics can be obtained through methods such as satellites, images, geographic information system (GIS) data, etc.; the above-mentioned fire characteristic indicators can include fire size, burning speed, flame height, smoke density, etc.; the above-mentioned indicator weights can be determined according to the importance of the fire characteristic indicators, that is, they can be the indicator weights corresponding to the wildfire characteristic indicators, which are not limited here and can be determined according to specific scenarios and applications.
[0060] It should be noted that for the above steps, the first wildfire characteristic and the second wildfire characteristic can be determined from the target wildfire characteristic: the specific process is: through the acquired geographic location information of the target wildfire characteristic, and then according to the geographic location information of the target wildfire area, determine the corresponding fire characteristic index related to the target wildfire area, and determine the corresponding weight for each indicator according to the importance and influence of the fire characteristic index. This index reflects the degree of fire or the severity of the fire in the wildfire area. By setting a first threshold, the severity of the wildfire is judged according to the fire characteristic index. If the fire characteristic index is greater than the first threshold, the wildfire characteristic is defined as the first wildfire characteristic; if the fire characteristic index is less than the first threshold, the wildfire characteristic is defined as the second wildfire characteristic.
[0061] It should also be noted that the fire characteristic index corresponding to the target wildfire characteristic can be obtained based on the geographic location information of the target wildfire characteristic and the corresponding fire characteristic index, as well as the index weight, and compared with the threshold value to determine the first wildfire characteristic and the second wildfire characteristic. The above-mentioned fire characteristic index and the corresponding weights of the index can help identify and analyze the potential danger level of wildfires to a certain extent, and can also reflect the importance of different fire characteristic indicators, so as to more accurately assess the severity of wildfires.
[0062] As an optional embodiment, after determining the fire result of whether the target wildfire area is on fire based on the second fire index, it also includes: determining the target wildfire data collection frequency of the target wildfire area based on the second fire index; and updating the wildfire data collection frequency of the target wildfire area to the target wildfire data collection frequency.
[0063] In this embodiment, according to the result of the second ignition index, the target wildfire data collection frequency of the target wildfire area can be determined according to the severity of the fire and the warning needs. Wildfire areas with small fires or those that have been controlled may not need frequent data collection, while wildfire areas with large fires or potential dangers may require more frequent data collection, as well as real-time monitoring and early warning; according to the determined target wildfire data collection frequency, the data collection of the target wildfire area is updated, which may involve the collection of satellite images, drone images, meteorological data, sensor data, etc. Based on the collected data, the wildfire situation can be analyzed in a timely manner, and relevant emergency processing and decision-making can be carried out as needed. The determination and update of the target wildfire data collection frequency in this step are not limited here and need to be adjusted according to specific scenarios and needs.
[0064] It should be noted that through the above steps, the ignition situation of the target wildfire area can be determined according to the second ignition index, and the frequency of collecting target wildfire data can be further determined, so as to obtain relevant data in a timely manner and conduct effective monitoring and response. By setting the collection frequency, on the one hand, it can ensure that relevant data of the target wildfire area is obtained in a timely manner to help monitor wildfires and improve the response speed and accuracy to wildfires. On the other hand, the setting of the collection frequency can ensure the continuity and accuracy of the data. By collecting data regularly, data at multiple time points can be obtained, so as to better understand the changing trends and evolution of wildfires, and provide a data basis for formulating response strategies.
[0065] As an optional embodiment, after determining the first fire index corresponding to the first wildfire feature, the method further includes: when the first fire index is greater than or equal to a second threshold, determining a fire result of the target wildfire area being on fire.
[0066] In this embodiment, for the above steps, it is first necessary to determine the threshold of the first ignition index, that is, the boundary for judging whether a wildfire has occurred. The setting of the second threshold is not limited here, and can be determined based on historical data and empirical judgment, and then the first ignition index is compared with the second threshold. If the first ignition index is greater than or equal to the second threshold, it means that the target wildfire area is on fire; when it is determined that the target wildfire area is on fire, the specific content of the ignition result can be determined according to specific needs and the characteristics of the target wildfire area, which may include information such as the location of the fire, the speed of the spread of the fire, and the generation of smoke. Relevant data can be obtained through data acquisition equipment, sensors, etc., and analyzed and processed to determine the ignition result of the target wildfire area. By determining the ignition result of the target wildfire area, it is possible to make effective predictions of the wildfire area in a targeted manner based on the detected ignition results to a certain extent, providing important support for the formulation of wildfire prevention decisions.
[0067] Based on the above embodiments and optional embodiments, an optional implementation is provided, which is described in detail below.
[0068] In the related art, when monitoring wildfire events, there is a technical problem that the accuracy of the monitoring results of wildfire events is low.
[0069] In view of this, a method for determining the ignition result of a wildfire area is provided in an optional implementation manner of the present invention, and the optional implementation manner of the present application is described in detail below.
[0070] S1. Obtain wildfire data in a wildfire area, and extract wildfire features from the wildfire data, wherein the wildfire features include a first wildfire feature and a second wildfire feature;
[0071] It should be noted that the wildfire data in the above-mentioned wildfire area can be the monitoring video collected by the video monitoring equipment within the regional monitoring range and the equipment status of the video sensor, and the acquired monitoring video and equipment status are sent to the data acquisition interaction platform to process the equipment status information to prevent inaccurate monitoring results due to sensor damage, collect environmental perception parameters within the target monitoring range, and obtain wildfire data in the environmental monitoring area through sensors; wherein, the environmental parameter data includes environmental parameters and location data, and the environmental parameter data can be collected and processed by the environmental information collection module and the environmental information processing module, specifically including: the environmental information collection module includes: an environmental information collection sub-module; the environmental information collection sub-module includes a number of sensors, and a number of collection sites are set in the environmental monitoring area, and a number of sensors are set at each collection site for monitoring the wildfire data in the environmental monitoring area. The environmental information processing module includes: an information input submodule, a wildfire resolution model establishment submodule and an information output submodule; the information input submodule is used to input wildfire data into the wildfire resolution model establishment submodule; the wildfire resolution model establishment submodule is used to establish a wildfire resolution model based on wildfire indicator data, and obtain characteristic information by inputting wildfire data into the wildfire resolution model; the information output submodule is used to output characteristic information, and the characteristic information includes: wildfires with obvious characteristics and wildfires with unclear characteristics; the information input submodule, the wildfire resolution model establishment submodule and the information output submodule are connected in sequence.
[0072] The above-mentioned wildfire characteristics extracted from wildfire data, wherein wildfires with obvious characteristics include: information data with obvious flames in the collected information, and wildfires with unclear characteristics include: information data without obvious flames in the collected information, that is, the above-mentioned first wildfire characteristic is the characteristic of obvious flames in the collected information, and the above-mentioned second wildfire characteristic is the characteristic of no obvious flames in the collected information.
[0073] It should also be noted that when extracting the wildfire characteristics of the wildfire data as mentioned above, the wildfire data can be input into the wildfire identification model to obtain the wildfire characteristics, wherein the wildfire data are all collected from the Internet and the data pre-stored in the substation, and the characteristics of the wildfire are analyzed, and the wildfire index data are established. The wildfire identification model is established based on the wildfire index data.
[0074] S2. Inputting the first mountain fire feature into a fire detection model to obtain a first fire index corresponding to the first mountain fire feature, wherein the fire detection model is obtained by training an initial model with sample data, and the sample data includes: a sample mountain fire feature and a sample fire index corresponding to the sample mountain fire feature;
[0075] It should be noted that the above-mentioned fire detection model is obtained by training the initial model based on sample data. The sample data includes sample wildfire characteristics and sample fire indexes corresponding to the sample wildfire characteristics. The sample wildfire characteristics can be extracted based on historical wildfire data. The accuracy and recall rate are used to reflect the indicators of model performance. The mean clustering algorithm is used to select the initial value (K means++) method to cluster the historical wildfire data; spatial pyramid pooling (SPP) is introduced to enhance the receptive field, and the spatial pyramid pooling structure is improved by dilated convolution; the prediction results of the optimized model are improved, and the channel attention mechanism and the incentive process are improved according to the differences in sample wildfire characteristics to improve the monitoring effect.
[0076] The above-mentioned fire detection model can be a wildfire monitoring model. It treats the classification problem as a regression problem and divides the image into several grid points for judgment in turn. In addition to outputting the position parameters of 4 objects and 1 confidence level of whether there is a target, the deep network also includes the probabilities corresponding to each category. The core of this one-stage algorithm lies in its loss function that combines multiple losses; anchor frame-based monitoring model; extracting the depth of the network and adding a 1×1 convolution between each 3×3 convolution, using fewer parameters to adjust the number of channels, thereby speeding up the monitoring speed; using multi-scale prediction, feature pyramid (FPN) structure, and convolution to improve monitoring performance.
[0077] The above-mentioned fire detection model can adopt a single-stage target detection algorithm (Yolov3) model, and its specific principles are as follows: The monitoring model based on the anchor frame includes the following steps: The Yolov3 model obtains 9 more reasonable initial frames by clustering on the data set; These 9 prior frames of different sizes are placed with the center point of each grid as the center as the initial prediction position; The prior frame is adjusted by regression to obtain the final prediction frame, and even if there are a large number of targets with extreme length-to-width ratios in the data set, they can be quickly regressed and monitored; The depth of the network is extracted, and a 1×1 convolution is added between each 3×3 convolution, and the number of channels is adjusted with fewer parameters, thereby speeding up the detection speed; In order to prevent the gradient vanishing problem, the residual edge structure is introduced, and the input of the convolution layer is jump-connected to the output end, and the input and output are added before being passed to the next convolution layer. The LeakyReLU function is used as a replacement. When the input is negative, there is a very small gradient, which solves the problem of neuron death. Multi-scale prediction, FPN structure, and convolution are used to improve detection performance. In addition to the last layer of feature maps, Yolov3 also outputs two shallower feature maps, which divide the image into different numbers of grid points and give each grid point of different sizes 3 prior frames of appropriate sizes. The last layer outputs a 13×13 feature layer with a large receptive field, which is used to detect large objects. The middle layer outputs a 26×26 feature map to detect medium-sized targets, and the shallow layer outputs a 52×52 feature map to detect small targets. Multi-scale prediction enables the network to detect more small targets, but due to the network size of small target detection, the network size of the network is not large enough. The network depth is shallow and it is unable to extract deep detailed features, so the accuracy is not ideal. Therefore, Yolov3 introduces the FPN structure based on multi-scale output. First, the more abstract deep features are increased to the same size as the shallow features through upsampling, and then stacked with the shallow output in the channel dimension, which improves the accuracy of medium and small target classification. In order to further improve the performance, Yolov3's downsampling abandons the use of pooling operations and uses convolutions with a step size of 2 whose parameters can be trained instead. Learnable parameters are used to more reasonably decide which information to discard to reduce the size of feature maps. By detecting whether there is a wildfire on the feature map with obvious features, the detection result is output if there is; if not, it further detects whether there is a wildfire with unclear features and outputs the detection result.
[0078] In the process of model training, it is also necessary to construct a loss function for model training. The loss function of Yolov3 is as follows:
[0079]
[0080] Among them, S is the grid size and the grid size, B is the preset fixed value box, and contains two functions that are irrelevant to the prediction results. Indicates whether the jth anchor box of the i-th grid is responsible for predicting a target object. If the rectangular box is responsible for predicting a target, its size is 1, otherwise it is 0. Indicates that the jth anchor box of the i-th grid is not responsible for the target. If the rectangular box is not responsible for predicting a target, its size is 1, otherwise it is 0. Regardless of the prediction result given by the model, as long as the position of the prediction box is actually without a target, the category loss will not be calculated. x and y represent the offset of the coordinates of the center of the real wildfire compared to the coordinates of the upper left corner of the current grid. To predict the offset of the real wildfire center coordinates, ω and h are the width and height of the real frame, and is the width and height of the prediction box, where the letters with subscript i or superscript i represent the data of the i-th corresponding box. i is the confidence of the anchor box of the i-th grid, The confidence of the anchor box of the i-th grid of the prediction box, P i (c) represents the category probability of the bounding box generated by the anchor box of the i-th grid, P i (c) is the category probability of the true box, ceclasses represents the number of categories of the cross entropy loss function in the loss function, is the category probability of the predicted box, λ coord and λ noobj are two hyperparameters, λ coord It is used to coordinate the inconsistent contributions of rectangular boxes of different sizes to the error function. On the contrary, when the rectangular box is larger, the coefficient is smaller, λ noobj It indicates the weight of the confidence error in the loss function when the prediction box has no target, in order to balance the impact of positive and negative samples on the loss function.
[0081] When the grid point contains the true wildfire label, the former is equal to 1 and the latter is equal to 0. When the grid point does not contain the true wildfire label, the former is equal to 0 and the latter is equal to 1.
[0082] The loss function of the Yolov3 model consists of five parts. The first line of the above formula is the loss function of the center of the wildfire. x and y represent the offset of the coordinates of the center of the wildfire compared to the coordinates of the upper left corner of the current grid. The second line is the loss of the width w and height h of the predicted box compared to the real box. The network calculates the location of the wildfire through the four values of x, y, w, and h. The third line is the loss of the confidence of the predicted box that actually has a wildfire. The fourth line is the loss of the confidence of the predicted box that is actually the background. The fifth line is the core of realizing the regression prediction category, which represents the loss of all categories. So no matter what the prediction result given by the model is, as long as the location of the predicted box is actually without an object, the category loss will not be calculated. coord and λ noobj The purpose of these two hyperparameters is to balance the impact of positive and negative samples on the loss function.
[0083] S3. When the first ignition index is less than a second threshold, determining a second ignition index corresponding to a second wildfire characteristic;
[0084] It should be noted that before this step, after determining the first wildfire index corresponding to the first wildfire characteristic, it is necessary to determine the ignition result of the wildfire area when the first ignition index is greater than or equal to the second ignition index, and then determine the second ignition index corresponding to the second wildfire characteristic when the first wildfire index corresponding to the first wildfire characteristic is less than the second threshold.
[0085] It should be noted that the second fire index is determined by inputting the second wildfire feature into a fire detection model for detection, and the fire result of whether the wildfire area is on fire is determined based on the second fire index.
[0086] S4. Determine whether the wildfire area is on fire based on the second fire index.
[0087] The above-mentioned fire detection model is included in the environmental information monitoring module, that is, the fire detection model establishment submodule and the feedback submodule; the fire detection model establishment submodule is used to establish a fire detection model based on deep learning and historical wildfire data, input the fire detection model, monitor whether there is a wildfire with obvious characteristics, and output the monitoring results if it does; if it does not exist, input the wildfire with unclear characteristics into the fire detection model, further monitor whether there is a wildfire with unclear characteristics, and output the monitoring results; the feedback submodule is used to calculate the monitoring results according to the wildfire indicator data, obtain the wildfire risk level, and adjust the collection frequency based on the wildfire risk level; the wildfire monitoring model establishment submodule and the feedback submodule are connected.
[0088] According to the fire result, the collection frequency can be further adjusted, including: calculating the fire result according to the wildfire index data, obtaining the wildfire risk level, and adjusting the collection frequency based on the wildfire risk level; wherein the collection frequency increases with the increase of the wildfire risk level, and decreases with the decrease of the wildfire risk level, specifically: analyzing the wildfire risk level, the probability of the wildfire risk level and the corresponding location data according to the environmental parameter data and historical wildfire data, and adjusting the collection frequency based on the wildfire risk level; wherein the collection frequency increases with the increase of the wildfire risk level, and decreases with the decrease of the wildfire risk level. At the same time, the alarm level and the corresponding original environmental data will be classified and saved in the data storage module, and the above information will be sent to the information push and alarm module for real-time SMS alarm and email alarm, thereby realizing real-time storage, real-time alarm and real-time governance.
[0089] Through the above optional implementation, at least the following beneficial effects can be achieved:
[0090] (1) The target wildfire features are determined based on the target wildfire data of the target wildfire area through the corresponding feature extraction method. That is, the target wildfire features can help with target detection and analysis, while improving the efficiency and accuracy of wildfire monitoring and prevention;
[0091] (2) Since the fire detection model is obtained by training the initial model with sample data, that is, training the fire detection model with sample data can not only optimize the initial model, but also improve the automatic detection capability, real-time monitoring and early warning capability, and quickly locate the fire location. To a certain extent, it can improve the accuracy of fire identification, reduce reaction time and loss, and provide data support for decision-making;
[0092] (3) Since the loss function is constructed during the initial model training process, that is, the loss function can measure the difference between the model's predicted output and the true value, the fire model can more accurately predict the location, size, confidence, and category of the target, thereby improving the effect of the fire detection model.
[0093] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.
[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.
[0095] Example 2
[0096] According to an embodiment of the present invention, a device for implementing the above-mentioned method for determining the ignition result of a wildfire area is also provided. Figure 2 FIG. 4 is a structural block diagram of a device for determining a fire result in a wildfire area according to an embodiment of the present invention. Figure 2 As shown, the device includes: an acquisition module 202, an extraction module 204, a first determination module 206, a second determination module 208, a third determination module 210 and a fourth determination module 212. The device is described in detail below.
[0097] An acquisition module 202 is used to acquire target wildfire data of a target wildfire area; an extraction module 204 is connected to the acquisition module 202 and is used to extract target wildfire features of the target wildfire data; a first determination module 206 is connected to the extraction module 204 and is used to determine a first wildfire feature and a second wildfire feature from the target wildfire features, wherein the first wildfire feature represents a wildfire feature whose corresponding ignition feature index is greater than or equal to a first threshold, and the second wildfire feature represents a wildfire feature whose corresponding ignition feature index is less than the first threshold; a second determination module 208 is connected to the first determination module 206 and is used to determine a first ignition index corresponding to the first wildfire feature; a third determination module 210 is connected to the second determination module 208 and is used to determine a second ignition index corresponding to the second wildfire feature when the first ignition index is less than a second threshold; a fourth determination module 212 is connected to the third determination module 210 and is used to determine a fire result of whether the target wildfire area is on fire based on the second ignition index.
[0098] It should be noted here that the above-mentioned acquisition module 202, extraction module 204, first determination module 206, second determination module 208, third determination module 210 and fourth determination module 212 correspond to steps S102 to S112 in the method for determining the fire results in the wildfire area, and the instances and application scenarios implemented by the multiple modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment 1.
[0099] Example 3
[0100] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor, wherein the processor is configured to execute instructions to implement any of the above-mentioned methods for determining the ignition result of a wildfire area.
[0101] Example 4
[0102] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute any of the above-mentioned methods for determining the fire results of a wildfire area.
[0103] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.
[0104] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0105] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.
[0106] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0107] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0108] If the 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 computer-readable storage medium. Based on this understanding, the technical solution of the present invention, 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, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.
[0109] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A method for determining the ignition result of a wildfire area, characterized in that: include: Acquire target wildfire data of target wildfire area; Extracting target wildfire features of the target wildfire data; Determine a first wildfire feature and a second wildfire feature from the target wildfire feature, wherein the first wildfire feature represents a wildfire feature whose corresponding ignition feature index is greater than or equal to a first threshold, and the second wildfire feature represents a wildfire feature whose corresponding ignition feature index is less than the first threshold; determining a first fire index corresponding to the first wildfire characteristic; When the first fire index is less than a second threshold, determining a second fire index corresponding to the second wildfire characteristic; A fire result of whether the target wildfire area is on fire is determined according to the second fire index.
2. The method according to claim 1, characterized in that The determining of a first fire index corresponding to the first wildfire characteristic comprises: The first wildfire feature is input into a fire detection model to obtain the first fire index corresponding to the first wildfire feature, wherein the fire detection model is obtained by training an initial model with sample data, and the sample data includes: sample wildfire features and sample fire indexes corresponding to the sample wildfire features.
3. The method according to claim 2, characterized in that The step of inputting the first wildfire feature into a fire detection model to obtain the first fire index corresponding to the first wildfire feature includes: In the case where the target wildfire data includes image data, the first wildfire feature is input into an identification frame module of a fire detection model to obtain multiple types of initial feature maps, wherein the multiple types of initial feature maps are feature maps identified by identification frames of different sizes; Inputting the multiple types of initial feature maps into the feature map processing module of the fire detection model, processing the feature map dimensions corresponding to the multiple types of initial feature maps respectively, and obtaining multiple types of target feature maps, wherein the multiple types of target feature maps are feature maps with the same corresponding feature map dimensions; The multi-class target feature map is input into the ignition index determination module of the ignition detection model to obtain the first ignition index.
4. The method according to claim 2, characterized in that: The method further includes: training the sample data in the following manner to obtain the fire detection model: In the case where the target wildfire data includes image data, an initial model is obtained, and a loss function for model training is constructed, wherein the loss function includes a first loss function term, a second loss function term, a third loss function term, a fourth loss function term, and a fifth loss function term, wherein the first loss function term represents the offset between the coordinates of the actual fire position and the coordinates of the predicted fire position, the second loss function term represents the offset between the size corresponding to the predicted identification box and the size corresponding to the real identification box, the third loss function term represents the offset between the confidence corresponding to the predicted identification box and the confidence corresponding to the real identification box, the fourth loss function term represents the offset between the confidence corresponding to the predicted background box and the confidence corresponding to the real background box, and the fifth loss function term represents the offset between the category probability corresponding to the predicted identification box and the category probability corresponding to the real identification box; Based on the loss function, the initial model is trained using the sample data to obtain the fire detection model.
5. The method according to claim 1, characterized in that Determining the first wildfire feature and the second wildfire feature from the target wildfire feature includes: Obtaining geographic location information of the target wildfire characteristics; Determining, based on the geographic location information, a fire characteristic index corresponding to the target wildfire area; Determining an indicator weight corresponding to the ignition characteristic indicator; Determining a fire characteristic index corresponding to the target wildfire characteristic according to the fire characteristic indicator and the corresponding indicator weight; It is determined that a mountain fire characteristic whose corresponding fire characteristic index is greater than the first threshold is the first mountain fire characteristic, and a mountain fire characteristic whose corresponding fire characteristic index is less than the first threshold is the second mountain fire characteristic.
6. The method according to claim 1, characterized in that After determining the fire result of whether the target wildfire area is on fire according to the second fire index, the method further includes: Determining a target wildfire data collection frequency of the target wildfire area according to the second fire index; The wildfire data collection frequency of the target wildfire area is updated to the target wildfire data collection frequency.
7. The method according to any one of claims 1 to 6, characterized in that: After determining the first fire index corresponding to the first wildfire characteristic, the method further includes: When the first ignition index is greater than or equal to the second threshold, a ignition result of the target wildfire area is determined.
8. A device for determining the ignition result of a wildfire area, characterized in that: include: An acquisition module, used to acquire target wildfire data of a target wildfire area; An extraction module, used to extract target wildfire features of the target wildfire data; A first determination module is used to determine a first wildfire feature and a second wildfire feature from the target wildfire feature, wherein the first wildfire feature represents a wildfire feature whose corresponding ignition feature index is greater than or equal to a first threshold, and the second wildfire feature represents a wildfire feature whose corresponding ignition feature index is less than the first threshold; A second determination module, configured to determine a first fire index corresponding to the first wildfire characteristic; A third determination module, configured to determine a second fire index corresponding to the second wildfire feature when the first fire index is less than a second threshold; The fourth determination module is used to determine the fire result of whether the target wildfire area is on fire according to the second fire index.
9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the method for determining the fire result of a wildfire area according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for determining the ignition result of a wildfire area according to any one of claims 1 to 7.