A method and device for monitoring and positioning a mountain fire of an ultra-high voltage transmission line

Abstract

The application relates to a method and device for monitoring and positioning a mountain fire of an ultrahigh-voltage power transmission line, and relates to the technical field of image processing.The method comprises the following steps: acquiring elevation information and satellite images of a target mountain area; establishing a three-dimensional grid model of the target mountain area based on the elevation information and the satellite images; regularly shooting actual images of the target mountain area through a pre-deployed image acquisition device; identifying a mountain fire range of the actual images by using a preset mountain fire detection model to obtain image two-dimensional information corresponding to the mountain fire range; and determining three-dimensional positioning information corresponding to the mountain fire range in the target mountain area according to the image two-dimensional information and the three-dimensional grid model.The application can fully utilize the higher recognition rate of a ground monitoring terminal compared with a satellite for a mountain fire, and can map the mountain fire coordinates in a mountain area picture back to latitude, longitude and altitude information in combination with the elevation information, so that the purpose of more accurately positioning the mountain fire can be achieved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method and device for monitoring and locating wildfires on ultra-high voltage transmission lines. Background Technology

[0002] Ultra-high voltage (UHV) transmission lines are a crucial component of the power grid system. Due to their need to traverse significant distances, their construction often involves crossing forest or mountainous areas. These areas are susceptible to severe wildfires caused by activities such as tomb sweeping, burning of vegetation, or indiscriminate arson, posing a serious threat to the stability and safety of the UHV transmission lines. Furthermore, UHV transmission line failures caused by wildfires are characterized by their suddenness, insidious nature, and seasonality. Therefore, strengthening wildfire monitoring is essential to effectively control the adverse effects of wildfires on the operational stability of transmission lines.

[0003] Currently, wildfire monitoring primarily relies on weather radar or satellites, or the scientific deployment of smoke sensors within the monitoring area. However, weather satellites have relatively low resolution and are susceptible to cloud cover or fog, leading to missed detections and negatively impacting the reliability of wildfire monitoring and location results near ultra-high-voltage transmission lines. Furthermore, ultra-high-voltage transmission lines are widely distributed and involve complex environments when crossing forest or mountainous areas, making fixed smoke alarm devices ineffective for overall monitoring and location. Summary of the Invention

[0004] To achieve more accurate location of wildfires, this application provides a method and device for monitoring and locating wildfires along ultra-high voltage transmission lines. The technical solution is as follows:

[0005] In a first aspect, embodiments of this application provide a method for monitoring and locating wildfires along ultra-high voltage transmission lines, the method comprising:

[0006] Acquire elevation information and satellite imagery of the target mountainous area;

[0007] A three-dimensional mesh model of the target mountain area is established based on the elevation information and satellite imagery;

[0008] The actual images of the target mountainous area are periodically captured using pre-deployed image acquisition devices;

[0009] The wildfire range is identified in the actual image using a preset wildfire detection model, and the corresponding two-dimensional image information is obtained.

[0010] According to the image two-dimensional information and the three-dimensional grid model, three-dimensional positioning information corresponding to the range of the forest fire in the target mountain area is determined.

[0011] In a second aspect, the embodiments of the present application provide a device for monitoring and positioning forest fires in an ultra-high voltage power transmission line, the device comprising:

[0012] an information acquisition module configured to acquire elevation information and satellite images of a target mountain area;

[0013] a model establishment module configured to establish a three-dimensional grid model of the target mountain area based on the elevation information and the satellite images;

[0014] an image acquisition module configured to periodically capture actual images of the target mountain area by using a pre-deployed image acquisition device;

[0015] a forest fire identification module configured to identify a range of forest fires in the actual images by using a pre-set forest fire detection model, and obtain image two-dimensional information corresponding to the range of the forest fires;

[0016] a forest fire positioning module configured to determine three-dimensional positioning information corresponding to the range of the forest fires in the target mountain area according to the image two-dimensional information and the three-dimensional grid model.

[0017] In a third aspect, the embodiments of the present application provide a server device, which comprises a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by the processor to implement the method for monitoring and positioning forest fires in an ultra-high voltage power transmission line according to the first aspect.

[0018] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, and the storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by a processor to implement the method for monitoring and positioning forest fires in an ultra-high voltage power transmission line according to the first aspect.

[0019] In summary, the present application has the following beneficial effects:

[0020] The method for monitoring and positioning a mountain fire near an ultra-high voltage transmission line disclosed in the application comprises the following steps: constructing a three-dimensional grid model of a mountainous area near the ultra-high voltage transmission line by using elevation information and satellite images; taking actual images of the mountainous area; when a mountain fire is detected in the actual images, mapping two-dimensional information of the mountain fire range in the actual images to the three-dimensional grid model to determine three-dimensional positioning information corresponding to the mountain fire range. In this way, when monitoring a mountain fire in the mountainous area near the ultra-high voltage transmission line, the ground monitoring terminal (image acquisition device) is used to take the actual images, and the higher recognition rate of the mountain fire by the ground monitoring terminal compared with the satellite is fully utilized. In addition, the coordinates of the mountain fire in the actual images are mapped back to the latitude, longitude and altitude information (three-dimensional positioning information) by using the elevation information, so that the positioning is more accurate. Furthermore, the three-dimensional positioning information of the mountain fire range can be directly drawn in the actual images taken by the ground monitoring terminal, which is convenient for users to view. Moreover, since only the existing elevation information and satellite image information are used, the satellite or radar resources do not need to be used in real time when the mountain fire is recognized and positioned, and the cost can be greatly saved. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 A flow chart of the method for monitoring and positioning a mountain fire near an ultra-high voltage transmission line in the embodiments of the application;

[0022] Figure 2 A mountainous area elevation map containing elevation information in the embodiments of the application;

[0023] Figure 3 A three-dimensional grid diagram generated based on the mountainous area elevation map in the embodiments of the application;

[0024] Figure 4 A diagram of a three-dimensional grid model of a mountainous area in the embodiments of the application;

[0025] Figure 5 A diagram for detecting and identifying a mountain fire range in the embodiments of the application;

[0026] Figure 6 A diagram of the architecture of a device for monitoring and positioning a mountain fire near an ultra-high voltage transmission line in the embodiments of the application. DETAILED DESCRIPTION

[0027] In order to make the objectives, technical solutions and advantages of the application clearer, the application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application and should not be used to limit the application. Figures 1-6

[0028] ​The embodiment of the application provides a mountain fire monitoring and positioning method for an ultrahigh-voltage power transmission line, and the execution subject of the method can be a management center of the ultrahigh-voltage power transmission line. The management center can monitor and manage the operation and maintenance state of the ultrahigh-voltage power transmission line, and can also be used for monitoring the environment near the erection area of the ultrahigh-voltage power transmission line, so as to ensure the operation stability of the ultrahigh-voltage power transmission line. In the step of monitoring the environment, if the erection area of the ultrahigh-voltage power transmission line passes through a mountainous area or a forest area, the management center can monitor the mountain fire that can occur in the mountainous area or the forest area through a pre-deployed image acquisition device, so as to timely find the mountain fire and accurately position the mountain fire position when the mountain fire occurs.

[0029] The processing flow of the mountain fire monitoring and positioning for the ultrahigh-voltage power transmission line shown in the specific embodiment will be described in detail below, and the content can be as follows: Figure 1

[0030] Step 101, obtaining the elevation information and satellite image of a target mountainous area.

[0031] The target mountainous area can be any adjacent mountainous area along the ultrahigh-voltage power transmission line.

[0032] In the implementation, after the ultrahigh-voltage power transmission line is erected, the management center can determine the mountainous area near the ultrahigh-voltage power transmission line according to a map through which the ultrahigh-voltage power transmission line passes, so as to monitor the mountain fire in the mountainous area in real time. For the target mountainous area, after the approximate range of the target mountainous area is determined, the management center can obtain the elevation information and satellite image of the target mountainous area, which can be collected by a third party and then be publicly disclosed. The management center can download the elevation information and satellite image from the third party through a network. The elevation information can be the distance of each point on the ground in the target mountainous area to a preset base surface along the plumb line direction, which can refer to the elevation map shown in Figure 2 The satellite image can be the image content of the target mountainous area obtained by a satellite.

[0033] Step 102, establishing a three-dimensional grid model of the target mountainous area based on the elevation information and the satellite image.

[0034] In the implementation, after the management center obtains the elevation information and satellite image of the target mountainous area, the management center can first generate a three-dimensional basic grid of the target mountainous area by using the elevation information, as shown in Figure 3 Then, the three-dimensional grid is mapped by using the satellite image, so that the three-dimensional grid model of the target mountainous area can be established, as shown in Figure 4

[0035] Step 103, regularly shooting actual images of the target mountainous area by using a pre-deployed image acquisition device.

[0036] ​​In implementation, after the three-dimensional grid model of the target mountain area is established, the management center can regularly shoot actual images of the target mountain area through the image collection device pre-deployed in the target mountain area. It can be understood that the image collection device can be pre-deployed in the target mountain area by the maintenance personnel of the ultra-high voltage transmission line, one or more image collection devices can be deployed, the shooting range of all image collection devices can cover the entire area of the target mountain area, and the maintenance personnel can evaluate the mountain fire prone area of the target mountain area. For the mountain fire high-risk area, multiple image collection devices can shoot actual images of the mountain fire high-risk area from multiple angles.

[0037] Step 104, using a preset mountain fire detection model to identify the mountain fire range of the actual image, and obtaining the image two-dimensional information corresponding to the mountain fire range.

[0038] In implementation, the management center can pre-store an artificial intelligence model for detecting mountain fires, which can be referred to as a mountain fire detection model. Through the mountain fire detection model, whether the specified mountain area image contains mountain fire and the range covered by the mountain fire can be effectively identified. Therefore, the management center can use the mountain fire detection model to identify the actual image after shooting the actual image of the target mountain area each time to detect whether there is mountain fire in the target mountain area. If there is mountain fire, the management center can further identify the mountain fire range in the actual image through the mountain fire detection model, so as to obtain the image two-dimensional information corresponding to the mountain fire range, that is, a two-dimensional coordinate system is established in the actual image, and the image two-dimensional information corresponding to the mountain fire range is constructed by two-dimensional coordinates. It can be referred to as shown in FIG. 1, wherein the mountain fire range containing three rectangles corresponds to the image two-dimensional information [x1, x2]&[y1, y4], [x3, x4]&[y2, y3] and [x5, x6]&[y5, y6]. Figure 5

[0039] Step 105, determining the three-dimensional positioning information corresponding to the mountain fire range in the target mountain area according to the image two-dimensional information and the three-dimensional grid model.

[0040] In implementation, after the management center obtains the image two-dimensional information corresponding to the mountain fire range, the image two-dimensional information can be substituted into the three-dimensional grid model of the target mountain area constructed in step 102, that is, the image two-dimensional information is mapped and transformed into information in the three-dimensional space through the positioning relationship between the two-dimensional image and the three-dimensional model, so as to determine the three-dimensional positioning information corresponding to the mountain fire range in the target mountain area. It can be understood that the positioning relationship between the two-dimensional image and the three-dimensional model can be pre-set, and each pixel point in the two-dimensional image can correspond to a space point in the three-dimensional model.

[0041] ​Optionally, the shooting angle and position of the two-dimensional image can be introduced when determining the three-dimensional positioning information, and correspondingly, the processing of step 105 can be as follows: determining a three-dimensional shooting space corresponding to the actual image according to the deployment position and the angle direction of the image acquisition device; mapping the three-dimensional shooting space to the three-dimensional grid model to obtain a three-dimensional model space corresponding to the actual image; and determining the three-dimensional positioning information corresponding to the mountain fire range in the target mountain area according to the mapping relationship between the two-dimensional image information and the three-dimensional model space.

[0042] In implementation, in the process of inputting the two-dimensional image information into the three-dimensional grid model by the management center, the shooting source of the actual image can be determined first, that is, the image acquisition device for shooting the image is determined, and then the deployment position and the angle direction of the image acquisition device can be obtained. It is not difficult to think that the staff of the management center can record the deployment position and the shooting angle direction of each image acquisition device when deploying the image acquisition device, and upload the deployment position and the angle direction to the management center. Then, the management center can determine the three-dimensional shooting space corresponding to the actual image in combination with the deployment position and the angle direction of the image acquisition device. Further, the management center can map the three-dimensional shooting space to the three-dimensional grid model, and specifically, a plurality of space points in the target mountain area can be taken as mapping points, and then all the mapping points contained in the three-dimensional shooting space are located in the three-dimensional grid model to obtain the three-dimensional model space corresponding to the actual image. Finally, the management center can determine the three-dimensional positioning information corresponding to the mountain fire range in the target mountain area from the two-dimensional image information of the mountain fire in the actual image according to the mapping relationship between the two-dimensional image information and the three-dimensional model space.

[0043] Optionally, in order to improve the positioning accuracy, the angle direction of the image acquisition device can be updated periodically, and correspondingly, the following processing can be performed before determining the three-dimensional shooting space: determining the latest angle direction of the image acquisition device according to the actual image and the satellite image of the target mountain area.

[0044] In implementation, considering that the wind near the mountain area is relatively strong and the birds frequently appear and disappear, which can affect the angle direction of the image acquisition device deployed in the mountain area, therefore, the management center needs to regularly correct the angle direction of the image acquisition device. Specifically, after obtaining the actual image of the target mountain area, the management center can call the satellite image of the target mountain area, and then select common features in the actual image and the satellite image, such as selecting the connecting line of random colors in the image as the common feature, or selecting part of the things in the image as the common feature, and then mapping the actual image and the satellite image through the common features to finally determine the latest angle direction of the image acquisition device. It is worth mentioning that when determining the latest angle direction, a time period with good weather condition, such as sunny day and no fog, should be selected, so as to ensure the clarity of the satellite image and help to more accurately determine the angle direction of the image acquisition device.

[0045] Optionally, the embodiment also discloses a training process of the wildfire detection model, and the specific process can be as follows: an image feature extraction model based on a neural network is established; a wildfire picture set is established according to the positional relationship of the image acquisition device relative to the target mountain area; the wildfire picture set is input into the image feature extraction model to train and generate the wildfire detection model.

[0046] In implementation, the management center can establish an initial image feature extraction model based on neural network technology, and the image feature extraction model can also directly use a machine learning model disclosed on the network. Then, the management center can set a matching wildfire picture set for the target mountain area, and can refer to the positional relationship of the image acquisition device relative to the target mountain area to select wildfire pictures and non-wildfire pictures with similar or similar shooting angles to form the wildfire picture set, and mark whether each picture contains wildfire.

[0047] In this way, the management center can input the pictures in the wildfire picture set into the image feature extraction model to train and generate the wildfire detection model. Furthermore, wildfire detection models suitable for different seasons can also be set according to different seasons, and correspondingly, the wildfire picture set can use wildfire pictures and non-wildfire pictures in the corresponding season.

[0048] Optionally, the image acquisition device can be set according to the wildfire-prone area, and the corresponding process can be as follows: the wildfire-prone area of the target mountain area is determined based on the three-dimensional grid model; at least one image acquisition device is set in the target mountain area for the position of all wildfire-prone areas, wherein the image acquisition periods of different image acquisition devices are the same or different.

[0049] In implementation, after the management center constructs the three-dimensional grid model of the target mountain area, the management center can evaluate the wildfire-prone area of the target mountain area based on the three-dimensional grid model, such as setting the vicinity of the high-frequency activity range of residents (such as the entrance and exit of the target mountain area and the vicinity of the mountainous road) as the wildfire-prone area, or setting the area with high altitude (easily struck by lightning) as the wildfire-prone area. Then, the management center can summarize all the wildfire-prone areas and analyze the positions of the wildfire-prone areas, so that at least one image acquisition device can be set in the target mountain area, so that the shooting range of all image acquisition devices can cover all wildfire-prone areas, and the area with high probability of wildfire occurrence or fast wildfire spread can be simultaneously in the shooting range of multiple image acquisition devices, so as to timely discover wildfire. Here, the image acquisition periods of different image acquisition devices can be independently set according to actual needs, which can be set to be the same or different.

[0050] Optionally, after detecting the wildfire, the spread trend of the wildfire can be estimated in combination with the three-dimensional grid model, and the corresponding processing can be as follows: when it is detected that there is a wildfire in the target mountain area, meteorological information of the target mountain area is acquired; the latest image of the target mountain area is captured at a preset interval; and the three-dimensional diffusion trend of the wildfire range is determined according to the meteorological information, the latest image and the three-dimensional grid model.

[0051] In implementation, when the management center detects that there is a wildfire in the target mountain area by using the wildfire detection model, the meteorological information of the target mountain area can be acquired first, which can include temperature, rainfall, wind power and wind direction. Then, the management center can control the image acquisition device of the target mountain area to capture the latest image of the target mountain area at a shorter preset interval. In this way, the management center can use the latest image in combination with the meteorological information and the three-dimensional grid model of the target mountain area to evaluate the spread trend of the wildfire, so as to determine the three-dimensional diffusion trend of the wildfire range, and thus the wildfire rescue operation for the target mountain area can be planned and carried out based on the three-dimensional diffusion trend.

[0052] Optionally, when identifying the wildfire, the environment of the target mountain area can be considered comprehensively to select the wildfire detection model that best matches the environment, and the processing of step 104 can be as follows: current environmental state information of the target mountain area is acquired, a target environment category to which the environmental state information belongs is determined; and the wildfire range of the actual image is identified by using the wildfire detection model corresponding to the target environment category to obtain the image two-dimensional information corresponding to the wildfire range.

[0053] In implementation, considering that the wildfire detection model uses an image feature detection model and mainly uses image recognition technology, the images of the mountain areas in different environmental states are quite different, such as the images of the mountain areas in different time periods within a day, the images of the mountain areas in different seasons within a year, or the images of the mountain areas in different weathers. Therefore, when training the wildfire detection model, the management center can classify the possible environmental states of the mountain areas to generate multiple environment categories, wherein the images of the mountain areas in each environment category are basically similar, and the images of the mountain areas in different environment categories are obviously different. Based on this, after acquiring the actual image of the target mountain area, the management center can further acquire the current environmental state information of the target mountain area, and then determine the target environment category to which the environmental state information belongs. Then, the management center can call the wildfire detection model corresponding to the target environment category to identify the wildfire range of the actual image captured to obtain the image two-dimensional information corresponding to the wildfire range.

[0054] The method for monitoring and positioning a mountain fire of an ultra-high voltage transmission line disclosed in the application is used for a mountain area near an ultra-high voltage transmission line. A three-dimensional grid model of the mountain area is first constructed using elevation information and satellite images. Then, actual images of the mountain area are taken. When a mountain fire is detected in the actual images, the two-dimensional information of the image corresponding to the range of the mountain fire is mapped to the three-dimensional grid model that has been established to determine the three-dimensional positioning information corresponding to the range of the mountain fire. In this way, when monitoring a mountain fire in a mountain area near an ultra-high voltage transmission line, the higher recognition rate of a mountain fire by a ground monitoring terminal (image acquisition device) compared with a satellite is fully utilized, and the coordinates of the mountain fire in the mountain area picture (actual image) are mapped back to the latitude, longitude and altitude information (three-dimensional positioning information) in combination with the elevation information, so that the purpose of more accurate positioning can be achieved. Furthermore, the three-dimensional positioning information of the range of the mountain fire can also be directly drawn in the actual images taken by the ground monitoring terminal, which is convenient for users to view. Moreover, since only the existing elevation information and satellite image information are used, the satellite or radar resources do not need to be used in real time when identifying and positioning the mountain fire compared with other solutions, and the cost can be greatly saved.

[0055] Based on the same technical concept, the embodiments of the application further provide an apparatus for monitoring and positioning a mountain fire of an ultra-high voltage transmission line, as shown in Figure 6 The apparatus comprises:

[0056] An information acquisition module 601 is configured to acquire elevation information and satellite images of a target mountain area.

[0057] A model establishment module 602 is configured to establish a three-dimensional grid model of the target mountain area based on the elevation information and the satellite images.

[0058] An image acquisition module 603 is configured to periodically take actual images of the target mountain area by using a pre-deployed image acquisition device.

[0059] A mountain fire identification module 604 is configured to identify a range of a mountain fire in the actual images by using a preset mountain fire detection model to obtain two-dimensional image information corresponding to the range of the mountain fire.

[0060] A mountain fire positioning module 605 is configured to determine three-dimensional positioning information corresponding to the range of the mountain fire in the target mountain area according to the two-dimensional image information and the three-dimensional grid model.

[0061] Optionally, the mountain fire positioning module 605 is specifically configured to

[0062] determine a three-dimensional shooting space corresponding to the actual images according to a deployment position and a viewing direction of the image acquisition device;

[0063] map the three-dimensional shooting space to the three-dimensional grid model to obtain a three-dimensional model space corresponding to the actual images.

[0064] According to the mapping relationship between the image two-dimensional information and the three-dimensional model space, three-dimensional positioning information corresponding to a mountain fire range in the target mountain area is determined.

[0065] Optionally, the image acquisition module 603 is further configured to:

[0066] According to the actual image and the satellite image of the target mountain area, a latest view direction of the image acquisition device is determined.

[0067] Optionally, the mountain fire recognition module 604 is further configured to:

[0068] An image feature extraction model based on a neural network is established.

[0069] According to a position relationship of the image acquisition device relative to the target mountain area, a mountain fire picture set is established.

[0070] The mountain fire picture set is input into the image feature extraction model, and a mountain fire detection model is trained and generated.

[0071] Optionally, the image acquisition module 603 is further configured to:

[0072] A mountain fire prone area of the target mountain area is determined based on the three-dimensional grid model.

[0073] At least one image acquisition device is arranged in the target mountain area for all positions of the mountain fire prone area, and image acquisition periods of different image acquisition devices are the same or different.

[0074] Optionally, the mountain fire positioning module 605 is further configured to:

[0075] When it is detected that there is a mountain fire in the target mountain area, meteorological information of the target mountain area is acquired.

[0076] Newest images of the target mountain area are taken at preset intervals.

[0077] According to the meteorological information, the newest images and the three-dimensional grid model, a three-dimensional diffusion trend of a mountain fire range is determined.

[0078] Optionally, the mountain fire recognition module 604 is specifically configured to:

[0079] Current environmental state information of the target mountain area is acquired, and a target environmental category to which the environmental state information belongs is determined.

[0080] A mountain fire range in the actual image is recognized by using a mountain fire detection model corresponding to the target environmental category, and image two-dimensional information corresponding to the mountain fire range is obtained.

[0081] The embodiment of the present application further provides a server device, which comprises a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to realize the method for monitoring and positioning of mountain fire of ultra-high voltage transmission line as described in steps 101-105.

[0082] Those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by program instructing relevant hardware, and the program can be stored in a computer readable storage medium, and the storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk.

[0083] The above are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application, and any feature disclosed in the specification (including the abstract and the drawings) can be replaced by other equivalent or similar purpose alternative features, unless specifically described. That is, each feature is only an example of a series of equivalent or similar features, unless specifically described.

Claims

1. A method for monitoring and locating wildfires along ultra-high voltage transmission lines, characterized in that, The method includes: The elevation information and satellite imagery of the target mountain area are obtained; the target mountain area is any adjacent mountain area along the overvoltage transmission line; the elevation information is the distance from each point on the ground in the target mountain area to a preset base plane along the vertical direction; the satellite imagery is the image content of the target mountain area obtained by satellite. A three-dimensional mesh model of the target mountain area is established based on the elevation information and satellite imagery; including: generating a three-dimensional basic mesh of the target mountain area using the elevation information, and then performing texture processing on the three-dimensional mesh using satellite imagery; The actual images of the target mountain area are captured periodically by pre-deployed image acquisition devices; wildfire-prone areas of the target mountain area are determined based on the three-dimensional mesh model; for the location of all wildfire-prone areas, at least one image acquisition device is set up in the target mountain area, wherein the image acquisition cycles of different image acquisition devices are the same or different; The method involves identifying the wildfire range in the actual image using a pre-defined wildfire detection model to obtain two-dimensional image information corresponding to the wildfire range. This includes: establishing a neural network-based image feature extraction model; creating a wildfire image set based on the positional relationship between the image acquisition device and the target mountain area; inputting the wildfire image set into the image feature extraction model to train and generate a wildfire detection model; and acquiring the current environmental state information of the target mountain area and determining the target environment category to which the environmental state information belongs. Finally, the method involves using the wildfire detection model corresponding to the target environment category to identify the wildfire range in the actual image to obtain two-dimensional image information corresponding to the wildfire range. Based on the two-dimensional information of the image and the three-dimensional mesh model, determine the three-dimensional positioning information corresponding to the wildfire range within the target mountainous area; including: The three-dimensional shooting space corresponding to the actual image is determined based on the deployment location and viewing angle of the image acquisition device. The three-dimensional shooting space is mapped onto the three-dimensional mesh model to obtain the three-dimensional model space corresponding to the actual image; Based on the mapping relationship between the two-dimensional information of the image and the three-dimensional model space, the three-dimensional positioning information corresponding to the wildfire range within the target mountainous area is determined; multiple spatial points within the target mountainous area are used as mapping markers, and then all mapping markers contained in the three-dimensional shooting space are located in the three-dimensional mesh model to obtain the three-dimensional model space corresponding to the actual image; it also includes: Based on the actual images and satellite imagery of the target mountain area, the latest viewing angle of the image acquisition device is determined.

2. The method according to claim 1, characterized in that, The method further includes: When a wildfire is detected in the target mountainous area, the meteorological information of the target mountainous area is obtained; Take the latest images of the target mountainous area at preset intervals; Based on the meteorological information, the latest images, and the three-dimensional mesh model, the three-dimensional spread trend of the wildfire area is determined.

3. A wildfire monitoring and location device for ultra-high voltage transmission lines, characterized in that, The device includes: The information acquisition module is used to acquire elevation information and satellite imagery of the target mountain area; the target mountain area is any adjacent mountain area along the overvoltage transmission line; the elevation information is the distance from each point on the ground in the target mountain area to a preset base plane along the vertical line; the satellite imagery is the image content of the target mountain area obtained by satellite. The model building module is used to build a three-dimensional mesh model of the target mountain area based on the elevation information and satellite imagery; including: generating a three-dimensional basic mesh of the target mountain area using elevation information, and performing texture processing on the three-dimensional mesh using satellite imagery; An image acquisition module is used to periodically capture actual images of the target mountain area using pre-deployed image acquisition devices; it is also used to determine the latest viewing angle of the image acquisition devices based on the actual images and satellite imagery of the target mountain area; it is also used to determine wildfire-prone areas of the target mountain area based on the three-dimensional mesh model; for the location of all wildfire-prone areas, at least one image acquisition device is set up in the target mountain area, wherein the image acquisition cycles of different image acquisition devices are the same or different; The wildfire identification module is used to identify the wildfire range in the actual image using a preset wildfire detection model, obtaining two-dimensional image information corresponding to the wildfire range; establish an image feature extraction model based on a neural network; establish a wildfire image set according to the positional relationship of the image acquisition device relative to the target mountain area; input the wildfire image set into the image feature extraction model to train and generate the wildfire detection model; specifically, it is used to obtain the current environmental state information of the target mountain area, determine the target environment category to which the environmental state information belongs; and use the wildfire detection model corresponding to the target environment category to identify the wildfire range in the actual image, obtaining two-dimensional image information corresponding to the wildfire range. The wildfire location module is used to determine the three-dimensional location information corresponding to the wildfire range within the target mountainous area based on the two-dimensional information of the image and the three-dimensional mesh model. Specifically, it is used to determine the three-dimensional shooting space corresponding to the actual image based on the deployment location and viewing direction of the image acquisition device; map the three-dimensional shooting space to the three-dimensional mesh model to obtain the three-dimensional model space corresponding to the actual image; determine the three-dimensional location information corresponding to the wildfire range within the target mountainous area based on the mapping relationship between the two-dimensional information of the image and the three-dimensional model space; use multiple spatial points within the target mountainous area as mapping markers, and then locate all the mapping markers contained in the three-dimensional shooting space in the three-dimensional mesh model to obtain the three-dimensional model space corresponding to the actual image.

4. A server-side device, characterized in that, The server-side device includes a processor and a memory. The memory stores at least one instruction, at least one program, a code set, or an instruction set. The at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the method for monitoring and locating wildfires on ultra-high voltage transmission lines as described in any one of claims 1 to 2.

5. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the method for monitoring and locating wildfires on ultra-high voltage transmission lines as described in any one of claims 1 to 2.

Images