Forest fire detection method and device, storage medium and electronic equipment

Abstract

The invention discloses a forest fire detection method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring a satellite remote sensing image corresponding to a target area, and generating a first fire point distribution diagram corresponding to the target area according to the satellite remote sensing image; obtaining an unmanned aerial vehicle aerial image corresponding to the target area, and processing the unmanned aerial vehicle aerial image and the satellite remote sensing image through a target algorithm to obtain a second fire point distribution diagram corresponding to the target area; and according to the first fire point distribution diagram and the second fire point distribution diagram, whether a fire exists in the target area is judged, and a judgment result is obtained. Through the forest fire detection method and device, the problem that the accuracy of forest fire detection is relatively low when forest fire detection is performed manually on the ground in related technologies is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and more specifically, to a forest fire detection method and device, a storage medium, and an electronic device. Background Art

[0002] With the development of agricultural loans, forest tree loans are also increasing. However, forests are prone to fire problems. Once a fire occurs, if it is not handled in time, it will cause great economic losses to loan users and economic impacts on financial institutions. At the same time, fires will also have a certain negative impact on ecological balance and the environment. It changes the physical condition of vegetation and releases various greenhouse gases into the atmosphere. Therefore, it is particularly important to monitor and locate forest fires in a timely and accurate manner. In the existing technology, forest fire monitoring often relies on ground manual and aircraft monitoring. This monitoring method is not only expensive, but also has blind spots and poor timeliness.

[0003] With regard to the problem that the accuracy of forest fire detection is relatively low when forest fire detection is performed manually on the ground in related technologies, no effective solution has been proposed so far. Summary of the invention

[0004] The main purpose of the present application is to provide a forest fire detection method and device, a storage medium and an electronic device to solve the problem in the related art that the accuracy of forest fire detection is relatively low when forest fire detection is performed manually on the ground.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a forest fire detection method is provided. The method comprises: obtaining a satellite remote sensing image corresponding to a target area, and generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; obtaining a drone aerial image corresponding to the target area, processing the drone aerial image and the satellite remote sensing image through a target algorithm, and obtaining a second fire point distribution map corresponding to the target area; judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and obtaining a judgment result.

[0006] Furthermore, generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image includes: acquiring first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range based on the satellite remote sensing image; performing calculations based on the first brightness temperature data and the second brightness temperature data to obtain a fire point intensity value corresponding to each pixel point in the satellite remote sensing image; and generating the first fire point distribution map based on the fire point intensity value corresponding to each pixel point.

[0007] Furthermore, calculations are performed based on the first brightness temperature data and the second brightness temperature data to obtain a fire point intensity value corresponding to each pixel in the satellite remote sensing image, including: calculating the difference between the first brightness temperature data and the second brightness temperature data to obtain a target difference; calculating the ratio between the first brightness temperature data and the second brightness temperature data to obtain a target ratio; and determining the fire point intensity value based on the target difference and / or the target ratio.

[0008] Furthermore, generating the first fire point distribution map based on the fire point intensity value corresponding to each pixel point includes: obtaining a preset fire point intensity threshold; for each pixel point, if the fire point intensity value corresponding to the pixel point is greater than the fire point intensity threshold, determining the pixel point as a first value; if the fire point intensity value corresponding to the pixel point is less than or equal to the fire point intensity threshold, determining the pixel point as a second value; generating the first fire point distribution map based on the first value and the second value.

[0009] Furthermore, the UAV aerial image and the satellite remote sensing image are processed by a target algorithm to obtain a second fire point distribution map corresponding to the target area, including: performing image enhancement processing on the UAV aerial image to obtain a target UAV aerial image, and performing image enhancement processing on the satellite remote sensing image to obtain a target satellite remote sensing image; performing feature extraction on the target UAV aerial image and the target satellite remote sensing image by the target algorithm to obtain target feature information; and obtaining a second fire point distribution map corresponding to the target area based on the target feature information.

[0010] Furthermore, feature extraction is performed on the target UAV aerial image and the target satellite remote sensing image by the target algorithm to obtain target feature information, including: feature extraction is performed on the target UAV aerial image by the target algorithm to obtain first feature information; feature extraction is performed on the target satellite remote sensing image by the target algorithm to obtain second feature information; feature fusion is performed on the first feature information and the second feature information by the target algorithm to obtain the target feature information.

[0011] Furthermore, the first feature information and the second feature information are feature fused through the target algorithm to obtain the target feature information, including: determining a first weight value for each eigenvalue in the first feature information through a self-attention module in the target algorithm; determining a second weight value for each eigenvalue in the second feature information through the self-attention module; and calculating each eigenvalue in the first feature information and each eigenvalue in the second feature information based on the first weight value and the second weight value to obtain the target feature information.

[0012] Furthermore, after determining whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map and obtaining a determination result, the method further includes: acquiring a first remote sensing image and a second remote sensing image of the target area, wherein the first remote sensing image is an image of the target area before the fire occurs, and the second remote sensing image is an image of the target area after the fire occurs; performing calculations based on the first remote sensing image to obtain a first normalized vegetation index; performing calculations based on the second remote sensing image to obtain a second normalized vegetation index; determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, wherein the degree of impact is used to determine a fire protection strategy for the target area.

[0013] Furthermore, after determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, the method also includes: acquiring multiple third remote sensing images of the target area, wherein the multiple third remote sensing images are images within a preset time period after the fire occurs in the target area; performing calculations based on the multiple third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image; establishing a change curve between the normalized vegetation index and time based on the third normalized vegetation index; and evaluating the vegetation recovery speed of the target area based on the change curve to obtain an evaluation result, wherein the evaluation result provides data reference for the ecological restoration of the target area.

[0014] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a forest fire detection device is provided. The device comprises: a first acquisition unit, used to acquire a satellite remote sensing image corresponding to a target area, and generate a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; a second acquisition unit, used to acquire a drone aerial image corresponding to the target area, and process the drone aerial image and the satellite remote sensing image through a target algorithm to obtain a second fire point distribution map corresponding to the target area; a judgment unit, used to judge whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and obtain a judgment result.

[0015] Furthermore, the first acquisition unit includes: a first acquisition subunit, used to acquire first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range based on the satellite remote sensing image; a calculation subunit, used to perform calculations based on the first brightness temperature data and the second brightness temperature data to obtain a fire point intensity value corresponding to each pixel in the satellite remote sensing image; and a generation subunit, used to generate the first fire point distribution map based on the fire point intensity value corresponding to each pixel.

[0016] Furthermore, the calculation subunit includes: a first calculation module, used to calculate the difference between the first brightness temperature data and the second brightness temperature data to obtain a target difference; a second calculation module, used to calculate the ratio between the first brightness temperature data and the second brightness temperature data to obtain a target ratio; a first determination module, used to determine the fire point intensity value based on the target difference and / or the target ratio.

[0017] Furthermore, the generation subunit includes: an acquisition module for acquiring a preset fire point intensity threshold; a second determination module for determining, for each pixel point, the first value if the fire point intensity value corresponding to the pixel point is greater than the fire point intensity threshold; a third determination module for determining the pixel point as the second value if the fire point intensity value corresponding to the pixel point is less than or equal to the fire point intensity threshold; and a generation module for generating the first fire point distribution map based on the first value and the second value.

[0018] Furthermore, the second acquisition unit includes: a processing subunit, used to perform image enhancement processing on the UAV aerial image to obtain a target UAV aerial image, and to perform image enhancement processing on the satellite remote sensing image to obtain a target satellite remote sensing image; an extraction subunit, used to perform feature extraction on the target UAV aerial image and the target satellite remote sensing image through the target algorithm to obtain target feature information; a determination subunit, used to obtain a second fire point distribution map corresponding to the target area based on the target feature information.

[0019] Furthermore, the extraction subunit includes: a first extraction module, used to perform feature extraction on the target UAV aerial image through the target algorithm to obtain first feature information; a second extraction module, used to perform feature extraction on the target satellite remote sensing image through the target algorithm to obtain second feature information; and a fusion module, used to perform feature fusion on the first feature information and the second feature information through the target algorithm to obtain the target feature information.

[0020] Furthermore, the fusion module includes: a first determination submodule, used to determine the first weight value of each eigenvalue in the first feature information through the self-attention module in the target algorithm; a second determination submodule, used to determine the second weight value of each eigenvalue in the second feature information through the self-attention module; and a calculation submodule, used to calculate each eigenvalue in the first feature information and each eigenvalue in the second feature information according to the first weight value and the second weight value to obtain the target feature information.

[0021] Furthermore, the device also includes: a third acquisition unit, used to judge whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and after obtaining the judgment result, acquire the first remote sensing image and the second remote sensing image of the target area, wherein the first remote sensing image is an image of the target area before the fire occurs, and the second remote sensing image is an image of the target area after the fire occurs; a first calculation unit, used to calculate based on the first remote sensing image to obtain a first normalized vegetation index; a second calculation unit, used to calculate based on the second remote sensing image to obtain a second normalized vegetation index; a determination unit, used to determine the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, wherein the degree of impact is used to determine the fire protection strategy for the target area.

[0022] Furthermore, the device also includes: a fourth acquisition unit, used to determine the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, and then acquire multiple third remote sensing images of the target area, wherein the multiple third remote sensing images are images within a preset time period after the fire occurs in the target area; a third calculation unit, used to perform calculations based on the multiple third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image; an establishment unit, used to establish a change curve between the normalized vegetation index and time based on the third normalized vegetation index; an evaluation unit, used to evaluate the vegetation recovery speed of the target area based on the change curve to obtain an evaluation result, wherein the evaluation result provides data reference for the ecological restoration of the target area.

[0023] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the storage medium stores a program, wherein when the program is executed, the device where the storage medium is located is controlled to execute any one of the above-mentioned forest fire detection methods.

[0024] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising a memory storing an executable program; and a processor for running the program, wherein the program executes any one of the above-mentioned forest fire detection methods when running.

[0025] Through this application, the following steps are adopted: obtaining a satellite remote sensing image corresponding to the target area, and generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; obtaining a drone aerial image corresponding to the target area, processing the drone aerial image and the satellite remote sensing image through a target algorithm, and obtaining a second fire point distribution map corresponding to the target area; judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and obtaining a judgment result, which solves the problem of low accuracy of forest fire detection obtained by manual forest fire detection on the ground in the related technology. In this scheme, the first fire point distribution map is obtained through the satellite remote sensing image corresponding to the target area, and the second fire point distribution map of the target area is obtained through the drone aerial image. Finally, according to the first fire point distribution map and the second fire point distribution map, it is judged whether there is a fire in the target area, avoiding the manual judgment of forest fires, and improving the accuracy of fire detection. When the financial institution judges that there is a fire in the target area through this scheme, it will promptly remind the loan users corresponding to the target area, thereby achieving the effect of avoiding property losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0027] Figure 1 is a flow chart of a forest fire detection method provided according to an embodiment of the present application;

[0028] Figure 2 is a schematic diagram of a target algorithm provided according to an embodiment of the present application;

[0029] Figure 3 is a schematic diagram of a fire detection method provided according to an embodiment of the present application;

[0030] Figure 4 is a schematic diagram of a forest fire detection device provided according to an embodiment of the present application;

[0031] Figure 5 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0033] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0036] The present invention is described below in conjunction with preferred implementation steps. Figure 1 is a flow chart of a forest fire detection method provided in accordance with an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0037] Step S101, obtaining a satellite remote sensing image corresponding to the target area, and generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image;

[0038] Optionally, obtain satellite remote sensing images of the target area to be detected. It should be noted that different satellite remote sensing data have different spatial resolutions, temporal resolutions, spectral resolutions, and radiation resolutions, and it is necessary to select a suitable data source according to the monitoring purpose and requirements. Since resource satellites (such as the Gaofen series) can provide high-resolution, multi-spectral fire identification and assessment data, but the revisit period is long, it is difficult to achieve dynamic monitoring. In this solution, meteorological satellites can be selected to provide real-time, continuous, and full-coverage satellite remote sensing images.

[0039] After obtaining the above-mentioned satellite remote sensing images, in order to improve the accuracy of subsequent fire point judgment, the satellite remote sensing images can be corrected.

[0040] Optionally, a first fire point distribution map of the target area is generated based on the satellite remote sensing image, for example, the satellite remote sensing image is processed by a deep learning model to obtain the first fire point distribution map. The number of fire points and the density of fire points in the fire point distribution map can be used to accurately determine whether a fire exists.

[0041] Step S102, obtaining an unmanned aerial image corresponding to the target area, and processing the unmanned aerial image and the satellite remote sensing image by a target algorithm to obtain a second fire point distribution map corresponding to the target area;

[0042] Optionally, the target area is photographed by a drone to obtain the above-mentioned drone aerial image, and then the drone aerial image and the satellite remote sensing image are processed by a target algorithm to obtain a second fire point distribution map corresponding to the target area. By combining the ground image and the remote sensing image, it is possible to more accurately detect whether there are fire points in the target area.

[0043] Step S103, judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map to obtain a judgment result.

[0044] Optionally, after obtaining the first fire point distribution map and the second fire point distribution map, whether there is a fire in the target area is judged based on the first fire point distribution map and the second fire point distribution map. For example, the first fire point distribution map and the second fire point distribution map are fused to obtain a target fire point distribution map, and then whether there is a fire is judged based on the fire point density in the target fire point distribution map.

[0045] To summarize, the first fire point distribution map is obtained through the satellite remote sensing images corresponding to the target area, and the second fire point distribution map of the target area is obtained through the drone aerial images. Finally, based on the first fire point distribution map and the second fire point distribution map, it is judged whether there is a fire in the target area, thereby avoiding manual judgment of forest fires and improving the accuracy of fire detection. Through this solution, financial institutions can promptly remind loan users corresponding to the target area when it is judged that there is a fire in the target area, thereby achieving the effect of avoiding property losses.

[0046] Optionally, in the forest fire detection method provided in an embodiment of the present application, generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image includes: obtaining first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range based on the satellite remote sensing image; performing calculations based on the first brightness temperature data and the second brightness temperature data to obtain a fire point intensity value corresponding to each pixel point in the satellite remote sensing image; and generating a first fire point distribution map based on the fire point intensity value corresponding to each pixel point.

[0047] In an optional embodiment, the following steps are used to generate a first fire point distribution map corresponding to the target area: extract first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range from a satellite remote sensing image, such as brightness temperature data in a thermal infrared channel (11 μm and 12 μm). It should be noted that brightness temperature data (BrightnessTemperature) is an important concept in the field of remote sensing, which is used to describe the radiation characteristics of objects on the earth's surface or in the atmosphere within a specific wavelength range.

[0048] After obtaining the above-mentioned first brightness temperature data and second brightness temperature data, the fire point intensity value corresponding to each pixel point in the satellite remote sensing image is calculated by the brightness temperature data, and finally a first fire point distribution map is generated according to the fire point intensity value.

[0049] In an optional embodiment, the above-mentioned fire point intensity value can be obtained by calculating the brightness temperature difference between the 11 μm and 12 μm channels.

[0050] The fire intensity is calculated using the brightness temperature data between different bands, and the fire intensity can be used to accurately determine whether there is a fire in the target area.

[0051] Optionally, in the forest fire detection method provided in an embodiment of the present application, calculation is performed based on the first brightness temperature data and the second brightness temperature data to obtain the fire point intensity value corresponding to each pixel point in the satellite remote sensing image, including: calculating the difference between the first brightness temperature data and the second brightness temperature data to obtain a target difference; calculating the ratio between the first brightness temperature data and the second brightness temperature data to obtain a target ratio; determining the fire point intensity value based on the target difference and / or the target ratio.

[0052] In an optional embodiment, the following steps are used to calculate the fire point intensity value corresponding to each pixel point: the difference between the first brightness temperature data and the second brightness temperature data is calculated, that is, the brightness temperature difference between the 11μm and 12μm channels (that is, the above-mentioned target difference) is calculated. Then the ratio between the first brightness temperature data and the second brightness temperature data is calculated, that is, the brightness temperature ratio between the 11μm and 12μm channels (that is, the above-mentioned target ratio) is calculated, and finally the fire point intensity value is determined according to the target difference and / or the target ratio.

[0053] In an optional embodiment, the target difference and the target ratio may be normalized and then summed to obtain the above-mentioned fire point intensity value. The target difference or the target ratio may also be directly determined as the fire point intensity value.

[0054] The fire point intensity value can be accurately obtained by calculating the difference or ratio between the brightness temperature data.

[0055] Optionally, in the forest fire detection method provided in the embodiment of the present application, generating a first fire point distribution map based on the fire point intensity value corresponding to each pixel point includes: obtaining a preset fire point intensity threshold; for each pixel point, if the fire point intensity value corresponding to the pixel point is greater than the fire point intensity threshold, determining the pixel point as a first value; if the fire point intensity value corresponding to the pixel point is less than or equal to the fire point intensity threshold, determining the pixel point as a second value; generating a first fire point distribution map based on the first value and the second value.

[0056] In an optional embodiment, generating a first fire point distribution map based on the fire point intensity value includes the following steps: setting a fire point intensity threshold, wherein the user can set the fire point intensity threshold based on information such as parameters of the actual device, and then judging whether there is a fire point at each pixel based on the fire point intensity value corresponding to the pixel and the fire point intensity threshold, thereby obtaining the above-mentioned first fire point distribution map.

[0057] In an optional embodiment, judging whether a fire point exists at a pixel based on the fire point intensity value corresponding to each pixel and the fire point intensity threshold includes the following steps: if the fire point intensity value corresponding to the pixel is greater than the fire point intensity threshold, the pixel is determined as a first value (for example, 1), and it is considered that a fire point exists at the pixel; if the fire point intensity value corresponding to the pixel is less than or equal to the fire point intensity threshold, the pixel is determined as a second value (for example, 0), and finally a first fire point distribution map is generated based on the first value and the second value.

[0058] The fire point intensity value and the fire point intensity threshold can be used to accurately determine whether a fire point exists.

[0059] Optionally, in the forest fire detection method provided in the embodiment of the present application, the UAV aerial image and the satellite remote sensing image are processed by a target algorithm to obtain a second fire point distribution map corresponding to the target area, including: performing image enhancement processing on the UAV aerial image to obtain the target UAV aerial image, performing image enhancement processing on the satellite remote sensing image to obtain the target satellite remote sensing image; performing feature extraction on the target UAV aerial image and the target satellite remote sensing image by a target algorithm to obtain target feature information; and obtaining a second fire point distribution map corresponding to the target area based on the target feature information.

[0060] In an optional embodiment, feature extraction is performed on the target UAV aerial image and the target satellite remote sensing image through a target algorithm to obtain target feature information, including: feature extraction is performed on the target UAV aerial image through a target algorithm to obtain first feature information; feature extraction is performed on the target satellite remote sensing image through a target algorithm to obtain second feature information; feature fusion is performed on the first feature information and the second feature information through a target algorithm to obtain target feature information.

[0061] In an optional embodiment, the process of obtaining the second fire point distribution map by the target algorithm is as follows: In order to improve the robustness of the target algorithm to image processing, image enhancement processing can be performed on the UAV aerial images and the satellite remote sensing images to obtain the target UAV aerial images and the target satellite remote sensing images.

[0062] In an optional embodiment, the purpose of increasing the image can be achieved by folding, cropping and superimposing the drone aerial images. Figure 2 As shown, the target UAV aerial image and the target satellite remote sensing image are input into the target algorithm, and the first feature extraction module of the target algorithm performs feature extraction on the target UAV aerial image to obtain the first feature information, and the second feature extraction module performs feature extraction on the target satellite remote sensing image to obtain the second feature information.

[0063] Then, the first feature information and the second feature information are feature-fused by the feature fusion module in the target algorithm to obtain the target feature information. Finally, the target feature information is predicted by the output layer in the target algorithm to obtain the second fire point distribution map corresponding to the target area.

[0064] In an optional embodiment, the feature fusion module may directly perform splicing processing on the first feature information and the second feature information, and then perform convolution calculation on the spliced ​​feature information again to obtain the above-mentioned target feature information.

[0065] The target algorithm can be used to discover the potential features between UAV aerial images and satellite remote sensing images, thereby improving the accuracy of the fire distribution map.

[0066] Optionally, in the forest fire detection method provided in an embodiment of the present application, the first feature information and the second feature information are feature fused through a target algorithm to obtain target feature information, including: determining a first weight value for each feature value in the first feature information through a self-attention module in the target algorithm; determining a second weight value for each feature value in the second feature information through the self-attention module; and calculating each feature value in the first feature information and each feature value in the second feature information based on the first weight value and the second weight value to obtain the target feature information.

[0067] In an optional embodiment, in order to improve the accuracy of feature fusion, the feature fusion module of the target algorithm can be a self-attention module, which calculates the first feature information and the second feature information respectively through the self-attention module to obtain the first weight value of each feature value in the first feature information and the second weight value of each feature value.

[0068] Then, each eigenvalue in the first feature information and each eigenvalue in the second feature information are calculated according to the weight value to obtain the above-mentioned target feature information.

[0069] Feature fusion through weights can more accurately grasp the importance of each eigenvalue and improve the accuracy of target feature information.

[0070] Optionally, in the forest fire detection method provided in the embodiment of the present application, after judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map and obtaining the judgment result, the method also includes: acquiring a first remote sensing image and a second remote sensing image of the target area, wherein the first remote sensing image is an image of the target area before the fire occurs, and the second remote sensing image is an image of the target area after the fire occurs; performing calculations based on the first remote sensing image to obtain a first normalized vegetation index; performing calculations based on the second remote sensing image to obtain a second normalized vegetation index; determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, wherein the degree of impact is used to determine the fire protection strategy for the target area.

[0071] In an optional embodiment, after obtaining the above-mentioned judgment result, if a fire occurs in the target area, the following steps are also included: collecting a first remote sensing image and a second remote sensing image of the target area. It should be noted that the first remote sensing image is an image of the target area before the fire occurs, and the second remote sensing image is an image of the target area after the fire occurs.

[0072] The impact of the fire on the target area is determined by analyzing the Normalized Difference Vegetation Index (NDVI) in the first remote sensing image and the second remote sensing image.

[0073] In an optional embodiment, the fire impact coefficient (FIC) is calculated using the normalized vegetation index (NDVI) before and after the fire, which indicates the degree of impact of the fire on vegetation. FIC = (NDVI1-NDVI2) / NDVI1, where NDVI1 and NDVI2 are the NDVI values ​​before and after the fire (i.e., the first normalized vegetation index and the second normalized vegetation index mentioned above). According to the value of FIC, the fire impact level is divided into mild, moderate, severe, etc.

[0074] The normalized vegetation index can be used to accurately assess the impact of a fire on a target area, which in turn helps to determine the fire protection strategy for the target area based on the impact level, thereby improving the safety of the target area.

[0075] Optionally, in the forest fire detection method provided in the embodiment of the present application, after determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, the method also includes: acquiring multiple third remote sensing images of the target area, wherein the multiple third remote sensing images are images within a preset time period after the fire occurs in the target area; performing calculations based on the multiple third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image; establishing a change curve between the normalized vegetation index and time based on the third normalized vegetation index; and evaluating the vegetation recovery speed of the target area based on the change curve to obtain an evaluation result, wherein the evaluation result provides data reference for the ecological restoration of the target area.

[0076] In an optional embodiment, in order to facilitate rapid ecological restoration of the target area, it also includes: collecting remote sensing images of the target area within a preset time period (for example, half a month) after the fire occurs, obtaining multiple third remote sensing images, and calculating through the third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image.

[0077] Then, the curve fitting is performed based on the third normalized vegetation index to obtain the change curve between the normalized vegetation index and time, so as to evaluate the vegetation recovery speed of the target area through the change curve, for example, at what time the vegetation in the target area is expected to be fully recovered. The speed and degree of vegetation recovery after the fire in the evaluation results can provide a reference for fire prevention and ecological restoration.

[0078] The normalized vegetation index can be used to accurately assess the rate of vegetation recovery so that the ecological restoration of the target area can be managed subsequently.

[0079] In an optional embodiment, the Figure 3 The flowchart shown achieves accurate detection of forest fires: Step 1, select appropriate satellite remote sensing data. Different satellite remote sensing data have different spatial resolutions, temporal resolutions, spectral resolutions, and radiation resolutions. It is necessary to select the appropriate data source according to the monitoring purpose and requirements. Since resource satellites (such as the Gaofen series, etc.) can provide high-resolution, multi-spectral fire identification and assessment data, but the revisit cycle is long, it is difficult to achieve dynamic monitoring. In this solution, meteorological satellites can be selected to provide real-time, continuous, and full-coverage fire point monitoring data.

[0080] Step 2: After obtaining the above satellite remote sensing data, preprocess the satellite remote sensing data.

[0081] In step 3, the satellite remote sensing data can be identified through the fire point identification algorithm to determine the location and intensity of the fire point. The fire point identification algorithm is a method that uses the spectral characteristics and brightness temperature characteristics of different bands in the satellite remote sensing data to determine the location and intensity of the fire point. For example, the brightness temperature data of the geostationary meteorological satellite in the thermal infrared channel (11μm and 12μm) is obtained. Then the fire point intensity value of each pixel in the remote sensing data is calculated, that is, the brightness temperature difference between the 11μm and 12μm channels.

[0082] Step 4, set the fire point issuance threshold, and determine whether the fire point intensity value is greater than the threshold according to the preset threshold. If so, the pixel is considered to be a fire point. Based on the judgment result, a fire point distribution map and fire point alarm information are generated.

[0083] The forest fire detection method provided in the embodiment of the present application obtains a satellite remote sensing image corresponding to the target area, and generates a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; obtains a drone aerial image corresponding to the target area, processes the drone aerial image and the satellite remote sensing image through a target algorithm, and obtains a second fire point distribution map corresponding to the target area; judges whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and obtains a judgment result, which solves the problem of low accuracy of forest fire detection obtained by manual forest fire detection on the ground in the related technology. In this scheme, the first fire point distribution map is obtained by the satellite remote sensing image corresponding to the target area, and the second fire point distribution map of the target area is obtained by the drone aerial image. Finally, it is judged whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, avoiding the manual judgment of forest fires and improving the accuracy of fire detection. When the financial institution judges that there is a fire in the target area through this scheme, it promptly reminds the loan users corresponding to the target area, thereby achieving the effect of avoiding property losses.

[0084] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0085] The embodiment of the present application also provides a forest fire detection device. It should be noted that the forest fire detection device of the embodiment of the present application can be used to execute the forest fire detection method provided in the embodiment of the present application. The forest fire detection device provided in the embodiment of the present application is introduced below.

[0086] Figure 4Schematic diagram of a forest fire detection device according to an embodiment of the present application. Figure 4 As shown, the device includes: a first acquisition unit 401, a second acquisition unit 402 and a judgment unit 403.

[0087] The first acquisition unit 401 is used to acquire a satellite remote sensing image corresponding to the target area, and generate a first fire point distribution map corresponding to the target area based on the satellite remote sensing image;

[0088] The second acquisition unit 402 is used to acquire the drone aerial image corresponding to the target area, and process the drone aerial image and the satellite remote sensing image through a target algorithm to obtain a second fire point distribution map corresponding to the target area;

[0089] The judgment unit 403 is used to judge whether there is a fire in the target area according to the first fire point distribution map and the second fire point distribution map to obtain a judgment result.

[0090] The forest fire detection device provided in the embodiment of the present application obtains the satellite remote sensing image corresponding to the target area through the first acquisition unit 401, and generates the first fire point distribution map corresponding to the target area based on the satellite remote sensing image; the second acquisition unit 402 obtains the drone aerial image corresponding to the target area, processes the drone aerial image and the satellite remote sensing image through the target algorithm, and obtains the second fire point distribution map corresponding to the target area; the judgment unit 403 judges whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and obtains the judgment result, which solves the problem of low accuracy of forest fire detection obtained by manual forest fire detection on the ground in the related technology. In this scheme, the first fire point distribution map is obtained through the satellite remote sensing image corresponding to the target area, and the second fire point distribution map of the target area is obtained through the drone aerial image. Finally, it is judged whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, avoiding the manual judgment of forest fires and improving the accuracy of fire detection. When the financial institution judges that there is a fire in the target area through this scheme, it will promptly remind the loan users corresponding to the target area, thereby achieving the effect of avoiding property losses.

[0091] Optionally, in the forest fire detection device provided in the embodiment of the present application, the first acquisition unit includes: a first acquisition subunit, used to acquire first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range based on a satellite remote sensing image; a calculation subunit, used to perform calculations based on the first brightness temperature data and the second brightness temperature data to obtain a fire point intensity value corresponding to each pixel in the satellite remote sensing image; and a generation subunit, used to generate a first fire point distribution map based on the fire point intensity value corresponding to each pixel.

[0092] Optionally, in the forest fire detection device provided in the embodiment of the present application, the calculation subunit includes: a first calculation module, used to calculate the difference between the first brightness temperature data and the second brightness temperature data to obtain a target difference; a second calculation module, used to calculate the ratio between the first brightness temperature data and the second brightness temperature data to obtain a target ratio; a first determination module, used to determine the fire point intensity value based on the target difference and / or target ratio.

[0093] Optionally, in the forest fire detection device provided in the embodiment of the present application, the generation subunit includes: an acquisition module, used to obtain a preset fire point intensity threshold; a second determination module, used to, for each pixel point, if the fire point intensity value corresponding to the pixel point is greater than the fire point intensity threshold, determine the pixel point as a first value; a third determination module, used to determine the pixel point as a second value if the fire point intensity value corresponding to the pixel point is less than or equal to the fire point intensity threshold; a generation module, used to generate a first fire point distribution map based on the first value and the second value.

[0094] Optionally, in the forest fire detection device provided in the embodiment of the present application, the second acquisition unit includes: a processing subunit, used to perform image enhancement processing on the UAV aerial image to obtain the target UAV aerial image, and perform image enhancement processing on the satellite remote sensing image to obtain the target satellite remote sensing image; an extraction subunit, used to perform feature extraction on the target UAV aerial image and the target satellite remote sensing image through a target algorithm to obtain target feature information; a determination subunit, used to obtain a second fire point distribution map corresponding to the target area based on the target feature information.

[0095] Optionally, in the forest fire detection device provided in the embodiment of the present application, the extraction subunit includes: a first extraction module, used to perform feature extraction on the target UAV aerial image through a target algorithm to obtain first feature information; a second extraction module, used to perform feature extraction on the target satellite remote sensing image through a target algorithm to obtain second feature information; and a fusion module, used to perform feature fusion on the first feature information and the second feature information through a target algorithm to obtain target feature information.

[0096] Optionally, in the forest fire detection device provided in the embodiment of the present application, the fusion module includes: a first determination submodule, used to determine the first weight value of each eigenvalue in the first feature information through the self-attention module in the target algorithm; a second determination submodule, used to determine the second weight value of each eigenvalue in the second feature information through the self-attention module; a calculation submodule, used to calculate each eigenvalue in the first feature information and each eigenvalue in the second feature information based on the first weight value and the second weight value to obtain target feature information.

[0097] Optionally, in the forest fire detection device provided in the embodiment of the present application, the device also includes: a third acquisition unit, used to judge whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and after obtaining the judgment result, acquire the first remote sensing image and the second remote sensing image of the target area, wherein the first remote sensing image is an image of the target area before the fire occurs, and the second remote sensing image is an image of the target area after the fire occurs; a first calculation unit, used to calculate based on the first remote sensing image to obtain a first normalized vegetation index; a second calculation unit, used to calculate based on the second remote sensing image to obtain a second normalized vegetation index; a determination unit, used to determine the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, wherein the degree of impact is used to determine the fire protection strategy for the target area.

[0098] Optionally, in the forest fire detection device provided in the embodiment of the present application, the device also includes: a fourth acquisition unit, used to determine the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, and then acquire multiple third remote sensing images of the target area, wherein the multiple third remote sensing images are images within a preset time period after the fire occurs in the target area; a third calculation unit, used to perform calculations based on the multiple third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image; an establishment unit, used to establish a change curve between the normalized vegetation index and time based on the third normalized vegetation index; an evaluation unit, used to evaluate the vegetation recovery speed of the target area based on the change curve to obtain an evaluation result, wherein the evaluation result provides data reference for the ecological restoration of the target area.

[0099] The forest fire detection device includes a processor and a memory. The first acquisition unit 401, the second acquisition unit 402 and the judgment unit 403 are all stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.

[0100] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and accurate detection of forest fires can be achieved by adjusting kernel parameters.

[0101] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0102] An embodiment of the present invention provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, a method for detecting forest fires is implemented.

[0103] An embodiment of the present invention provides a processor, which is used to run a program, wherein the program executes a forest fire detection method when it is run.

[0104] like Figure 5 As shown, an embodiment of the present invention provides an electronic device, the device includes a processor, a memory, and a program stored in the memory and executable on the processor, and the processor implements the following steps when executing the program: obtaining a satellite remote sensing image corresponding to a target area, and generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; obtaining a drone aerial image corresponding to the target area, processing the drone aerial image and the satellite remote sensing image through a target algorithm, and obtaining a second fire point distribution map corresponding to the target area; judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and obtaining a judgment result.

[0105] Optionally, generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image includes: acquiring first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range based on the satellite remote sensing image; performing calculations based on the first brightness temperature data and the second brightness temperature data to obtain fire point intensity values ​​corresponding to each pixel point in the satellite remote sensing image; and generating a first fire point distribution map based on the fire point intensity values ​​corresponding to each pixel point.

[0106] Optionally, calculation is performed based on the first brightness temperature data and the second brightness temperature data to obtain the fire point intensity value corresponding to each pixel in the satellite remote sensing image, including: calculating the difference between the first brightness temperature data and the second brightness temperature data to obtain a target difference; calculating the ratio between the first brightness temperature data and the second brightness temperature data to obtain a target ratio; determining the fire point intensity value based on the target difference and / or the target ratio.

[0107] Optionally, generating a first fire point distribution map based on the fire point intensity value corresponding to each pixel point includes: obtaining a preset fire point intensity threshold; for each pixel point, if the fire point intensity value corresponding to the pixel point is greater than the fire point intensity threshold, determining the pixel point as a first value; if the fire point intensity value corresponding to the pixel point is less than or equal to the fire point intensity threshold, determining the pixel point as a second value; generating a first fire point distribution map based on the first value and the second value.

[0108] Optionally, the UAV aerial images and the satellite remote sensing images are processed by a target algorithm to obtain a second fire point distribution map corresponding to the target area, including: performing image enhancement processing on the UAV aerial images to obtain the target UAV aerial images, and performing image enhancement processing on the satellite remote sensing images to obtain the target satellite remote sensing images; performing feature extraction on the target UAV aerial images and the target satellite remote sensing images by a target algorithm to obtain target feature information; and obtaining a second fire point distribution map corresponding to the target area based on the target feature information.

[0109] Optionally, feature extraction is performed on the target UAV aerial image and the target satellite remote sensing image through a target algorithm to obtain target feature information, including: feature extraction is performed on the target UAV aerial image through a target algorithm to obtain first feature information; feature extraction is performed on the target satellite remote sensing image through a target algorithm to obtain second feature information; feature fusion is performed on the first feature information and the second feature information through a target algorithm to obtain target feature information.

[0110] Optionally, the first feature information and the second feature information are feature fused through a target algorithm to obtain target feature information, including: determining a first weight value for each eigenvalue in the first feature information through a self-attention module in the target algorithm; determining a second weight value for each eigenvalue in the second feature information through the self-attention module; and calculating each eigenvalue in the first feature information and each eigenvalue in the second feature information based on the first weight value and the second weight value to obtain the target feature information.

[0111] Optionally, after judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map and obtaining the judgment result, the method also includes: acquiring a first remote sensing image and a second remote sensing image of the target area, wherein the first remote sensing image is an image of the target area before the fire occurs, and the second remote sensing image is an image of the target area after the fire occurs; performing calculations based on the first remote sensing image to obtain a first normalized vegetation index; performing calculations based on the second remote sensing image to obtain a second normalized vegetation index; determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, wherein the degree of impact is used to determine the fire protection strategy for the target area.

[0112] Optionally, after determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, the method also includes: acquiring multiple third remote sensing images of the target area, wherein the multiple third remote sensing images are images within a preset time period after the fire occurs in the target area; performing calculations based on the multiple third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image; establishing a change curve between the normalized vegetation index and time based on the third normalized vegetation index; and evaluating the vegetation recovery speed of the target area based on the change curve to obtain an evaluation result, wherein the evaluation result provides data reference for the ecological restoration of the target area.

[0113] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0114] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that initializes the following method steps: obtaining a satellite remote sensing image corresponding to a target area, and generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; obtaining a drone aerial image corresponding to the target area, processing the drone aerial image and the satellite remote sensing image through a target algorithm, and obtaining a second fire point distribution map corresponding to the target area; judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map, and obtaining a judgment result.

[0115] Optionally, generating a first fire point distribution map corresponding to the target area based on the satellite remote sensing image includes: acquiring first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range based on the satellite remote sensing image; performing calculations based on the first brightness temperature data and the second brightness temperature data to obtain fire point intensity values ​​corresponding to each pixel point in the satellite remote sensing image; and generating a first fire point distribution map based on the fire point intensity values ​​corresponding to each pixel point.

[0116] Optionally, calculation is performed based on the first brightness temperature data and the second brightness temperature data to obtain the fire point intensity value corresponding to each pixel in the satellite remote sensing image, including: calculating the difference between the first brightness temperature data and the second brightness temperature data to obtain a target difference; calculating the ratio between the first brightness temperature data and the second brightness temperature data to obtain a target ratio; determining the fire point intensity value based on the target difference and / or the target ratio.

[0117] Optionally, generating a first fire point distribution map based on the fire point intensity value corresponding to each pixel point includes: obtaining a preset fire point intensity threshold; for each pixel point, if the fire point intensity value corresponding to the pixel point is greater than the fire point intensity threshold, determining the pixel point as a first value; if the fire point intensity value corresponding to the pixel point is less than or equal to the fire point intensity threshold, determining the pixel point as a second value; generating a first fire point distribution map based on the first value and the second value.

[0118] Optionally, the UAV aerial images and the satellite remote sensing images are processed by a target algorithm to obtain a second fire point distribution map corresponding to the target area, including: performing image enhancement processing on the UAV aerial images to obtain the target UAV aerial images, and performing image enhancement processing on the satellite remote sensing images to obtain the target satellite remote sensing images; performing feature extraction on the target UAV aerial images and the target satellite remote sensing images by a target algorithm to obtain target feature information; and obtaining a second fire point distribution map corresponding to the target area based on the target feature information.

[0119] Optionally, feature extraction is performed on the target UAV aerial image and the target satellite remote sensing image through a target algorithm to obtain target feature information, including: feature extraction is performed on the target UAV aerial image through a target algorithm to obtain first feature information; feature extraction is performed on the target satellite remote sensing image through a target algorithm to obtain second feature information; feature fusion is performed on the first feature information and the second feature information through a target algorithm to obtain target feature information.

[0120] Optionally, the first feature information and the second feature information are feature fused through a target algorithm to obtain target feature information, including: determining a first weight value for each eigenvalue in the first feature information through a self-attention module in the target algorithm; determining a second weight value for each eigenvalue in the second feature information through the self-attention module; and calculating each eigenvalue in the first feature information and each eigenvalue in the second feature information based on the first weight value and the second weight value to obtain the target feature information.

[0121] Optionally, after judging whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map and obtaining the judgment result, the method also includes: acquiring a first remote sensing image and a second remote sensing image of the target area, wherein the first remote sensing image is an image of the target area before the fire occurs, and the second remote sensing image is an image of the target area after the fire occurs; performing calculations based on the first remote sensing image to obtain a first normalized vegetation index; performing calculations based on the second remote sensing image to obtain a second normalized vegetation index; determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, wherein the degree of impact is used to determine the fire protection strategy for the target area.

[0122] Optionally, after determining the degree of impact of the fire on the target area based on the first normalized vegetation index and the second normalized vegetation index, the method also includes: acquiring multiple third remote sensing images of the target area, wherein the multiple third remote sensing images are images within a preset time period after the fire occurs in the target area; performing calculations based on the multiple third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image; establishing a change curve between the normalized vegetation index and time based on the third normalized vegetation index; and evaluating the vegetation recovery speed of the target area based on the change curve to obtain an evaluation result, wherein the evaluation result provides data reference for the ecological restoration of the target area.

[0123] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0124] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0125] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0126] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0127] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0128] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0129] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0130] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0131] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0132] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for detecting forest fires, characterized in that: include: Acquire a satellite remote sensing image corresponding to the target area, and generate a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; Acquire a drone aerial image corresponding to the target area, and process the drone aerial image and the satellite remote sensing image by a target algorithm to obtain a second fire point distribution map corresponding to the target area; A judgment is made as to whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map to obtain a judgment result.

2. The method according to claim 1, characterized in that: Generating a first fire point distribution map corresponding to the target area according to the satellite remote sensing image includes: Acquire first brightness temperature data in a first wavelength range and second brightness temperature data in a second wavelength range according to the satellite remote sensing image; Calculating according to the first brightness temperature data and the second brightness temperature data to obtain a fire point intensity value corresponding to each pixel point in the satellite remote sensing image; The first fire point distribution map is generated according to the fire point intensity value corresponding to each pixel point.

3. The method according to claim 2, characterized in that Calculating according to the first brightness temperature data and the second brightness temperature data to obtain the fire point intensity value corresponding to each pixel point in the satellite remote sensing image includes: Calculating a difference between the first brightness temperature data and the second brightness temperature data to obtain a target difference; Calculating a ratio between the first brightness temperature data and the second brightness temperature data to obtain a target ratio; The fire point intensity value is determined based on the target difference and / or the target ratio.

4. The method according to claim 2, characterized in that: Generating the first fire point distribution map according to the fire point intensity value corresponding to each pixel point includes: Get the preset fire point intensity threshold; For each pixel point, if the fire point intensity value corresponding to the pixel point is greater than the fire point intensity threshold, the pixel point is determined as the first value; If the fire point intensity value corresponding to the pixel point is less than or equal to the fire point intensity threshold, the pixel point is determined as a second value; The first fire point distribution map is generated according to the first value and the second value.

5. The method according to claim 1, characterized in that The drone aerial image and the satellite remote sensing image are processed by a target algorithm to obtain a second fire point distribution map corresponding to the target area, including: Performing image enhancement processing on the drone aerial image to obtain a target drone aerial image, and performing image enhancement processing on the satellite remote sensing image to obtain a target satellite remote sensing image; Extracting features from the target UAV aerial image and the target satellite remote sensing image using the target algorithm to obtain target feature information; A second fire point distribution map corresponding to the target area is obtained based on the target feature information.

6. The method according to claim 5, characterized in that The target feature information obtained by extracting features from the target UAV aerial image and the target satellite remote sensing image through the target algorithm includes: Extracting features from the target drone aerial image using the target algorithm to obtain first feature information; Extracting features from the target satellite remote sensing image using the target algorithm to obtain second feature information; The first feature information and the second feature information are subjected to feature fusion by the target algorithm to obtain the target feature information.

7. The method according to claim 6, characterized in that Performing feature fusion on the first feature information and the second feature information by the target algorithm to obtain the target feature information includes: Determine a first weight value of each feature value in the first feature information by a self-attention module in the target algorithm; Determine a second weight value of each feature value in the second feature information by the self-attention module; Each feature value in the first feature information and each feature value in the second feature information are calculated according to the first weight value and the second weight value to obtain the target feature information.

8. The method according to claim 1, characterized in that After determining whether there is a fire in the target area based on the first fire point distribution map and the second fire point distribution map and obtaining a determination result, the method further includes: Acquire a first remote sensing image and a second remote sensing image of the target area, wherein the first remote sensing image is an image of the target area before a fire occurs, and the second remote sensing image is an image of the target area after a fire occurs; Calculating according to the first remote sensing image to obtain a first normalized vegetation index; performing calculation according to the second remote sensing image to obtain a second normalized vegetation index; The degree of impact of the fire on the target area is determined based on the first normalized difference vegetation index and the second normalized difference vegetation index, wherein the degree of impact is used to determine a fire protection strategy for the target area.

9. The method according to claim 8, characterized in that After determining the impact degree of the fire on the target area according to the first normalized vegetation index and the second normalized vegetation index, the method further includes: Acquire a plurality of third remote sensing images of the target area, wherein the plurality of third remote sensing images are images within a preset time period after a fire occurs in the target area; Calculating according to the plurality of third remote sensing images to obtain a third normalized vegetation index corresponding to each third remote sensing image; According to the third normalized vegetation index, a change curve between the normalized vegetation index and time is established; The vegetation recovery speed of the target area is evaluated according to the change curve to obtain an evaluation result, wherein the evaluation result provides data reference for ecological restoration of the target area.

10. A forest fire detection device, characterized in that: include: A first acquisition unit is used to acquire a satellite remote sensing image corresponding to a target area, and generate a first fire point distribution map corresponding to the target area based on the satellite remote sensing image; A second acquisition unit is used to acquire a drone aerial image corresponding to the target area, and process the drone aerial image and the satellite remote sensing image through a target algorithm to obtain a second fire point distribution map corresponding to the target area; The judgment unit is used to judge whether there is a fire in the target area according to the first fire point distribution map and the second fire point distribution map to obtain a judgment result.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the storage medium is controlled to execute the forest fire detection method according to any one of claims 1 to 9 on a device.

12. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the forest fire detection method described in any one of claims 1 to 9.

Images