Mine safety early warning method, system, equipment and medium

Abstract

The present application relates to a mine safety early warning method, system, equipment and medium, the method specifically comprising: obtaining an aerial global image of a mine and its surrounding areas through an unmanned aerial vehicle; establishing a three-dimensional model of the area where the mine is located and the area around the mine based on the aerial global image; obtaining precipitation information of the area where the mine is located and the area around the mine; importing the three-dimensional model and the precipitation information of the area where the mine is located and the area around the mine into a preset neural network prediction model to predict the probability of mine disasters; and issuing a safety early warning based on the probability of disasters. The present application obtains data on the main influencing factors that may cause mine disasters by establishing a three-dimensional model of the mine and its surrounding areas, inputs the relevant data into the neural network prediction model to predict the water accumulation level of the mine, realizes the early prediction of flood disasters in the mine, and sends early warnings to personnel in time before the disaster occurs to prompt them to take corresponding measures.

Description

Technical Field

[0001] The present application relates to the field of safety warning, and in particular to a mine safety warning method, system, equipment and medium. Background Art

[0002] With the advancement of science and technology, digitalization and intelligence have gradually been integrated into all aspects of life and production. For the current mining industry, traditional mines are gradually transforming into digital mines. As the main production facilities for mining and processing, mines are usually located in mountainous areas rich in mineral resources. It is precisely because of their special geographical location that natural disasters have a huge impact on mines. Once natural disasters such as waterlogging or flooding occur in mines, it will cause huge disasters and losses to industrial and agricultural production and people’s lives and property.

[0003] This requires an effective early warning system for mine hazards. At present, mine safety warnings are usually carried out manually on a regular basis to inspect the amount of water accumulated in the mine, the drainage system, etc., or to collect various data from the mine through various sensors to analyze possible dangers in the mine.

[0004] Through the above method, only the mine itself is inspected, and the impact of factors such as rainfall on the upstream and downstream areas or surrounding areas of the mine is not considered, and it is impossible to accurately provide safety reminders to personnel. Summary of the invention

[0005] In order to accurately provide safety reminders to personnel, the present application provides a mine safety early warning method, system, equipment and medium.

[0006] In the first aspect, the present application provides a mine safety early warning method using the following technical solution:

[0007] Use drones to obtain global aerial images of the mine area and surrounding areas;

[0008] Establishing a three-dimensional model of the area where the mine is located and the area surrounding the mine based on the aerial global image;

[0009] Obtain precipitation information in the area where the mine is located and the area surrounding the mine;

[0010] Importing the three-dimensional model and precipitation information of the area where the mine is located and the area surrounding the mine into a preset neural network prediction model, and outputting a prediction result, wherein the prediction result is the water accumulation level of the mine;

[0011] A safety warning is issued based on the prediction result.

[0012] By adopting the above technical solution, the basic data for mine disaster prediction is obtained through drone global photography, and the mine disaster prediction model is generated through neural network model learning feedback. The obtained basic disaster prediction data is imported into the prediction model to complete the mine disaster prediction, and safety warning information is generated in time according to the prediction results output by the prediction model to prompt relevant personnel to respond quickly and prevent and control possible flood disasters in mines.

[0013] Preferably, the camera device of the drone includes four inclined cameras distributed around the drone and forming a certain inclination angle with the ground, and a vertical camera arranged at the bottom of the drone and perpendicular to the ground.

[0014] By adopting the above technical solution, the drone camera device can obtain global image data of the mine and its surrounding areas in all directions. Compared with the setting of a single camera, the regular setting of multiple cameras can more accurately obtain the terrain information of the mine and its surrounding areas, the overall structure of the mine and other information, and provide accurate data for subsequent three-dimensional modeling.

[0015] Preferably, the step of obtaining an aerial global image of the area where the mine is located and the area surrounding the mine by using a drone specifically includes the following steps:

[0016] Determine the drone cruise cycle according to the preset drone cruise table;

[0017] According to the UAV cruise cycle, aerial global images of the mine area and the mine surrounding area are taken.

[0018] By adopting the above technical solution, the terrain information of the mine and its surrounding areas is photographed regularly to prevent the terrain information of the mine and its surrounding areas from being affected by climate change or human factors, ensure that the global image obtained is not much different from the actual situation, and ensure the accuracy of subsequent prediction results.

[0019] Preferably, the three-dimensional model of the area where the mine is located and the area surrounding the mine and the neural network prediction model are iterated based on the real-time aerial global images taken by the drone during regular cruises.

[0020] By adopting the above technical solution, the three-dimensional model of the mine area and the surrounding area of ​​the mine and the neural network prediction model are iterated according to changes in actual conditions, ensuring that the prediction results output by the prediction model are relatively accurate and making the prediction results more consistent with the actual conditions.

[0021] Preferably, in the step of importing the three-dimensional model and the precipitation information of the region where the mine is located and the region surrounding the mine into the neural network prediction model, the prediction of the disaster probability of the mine specifically includes the following steps:

[0022] Extracting the topographic features of the mine and the area surrounding the mine according to the three-dimensional model;

[0023] Calculate the maximum water storage capacity and drainage capacity of the mine based on the three-dimensional model;

[0024] The maximum water storage capacity of the mine, the drainage capacity of the mine, the terrain characteristics of the mine and the surrounding area of ​​the mine and the precipitation information are input as input data into the neural network prediction model to output the prediction result.

[0025] By adopting the above technical scheme, the maximum water storage capacity of the mine, the drainage capacity of the mine, the terrain characteristics of the mine and the surrounding area, and precipitation information are introduced as the main influencing factors affecting the occurrence of mine flood disasters. The above input values ​​directly represent the relationship between the overall water storage capacity of the mine and the drainage capacity of the mine. Through the analysis of the mine's water storage capacity and precipitation, the water storage situation of the mine within a certain period of time after rainfall in the surrounding area is predicted; various factors that lead to mine disasters are considered from multiple angles to ensure the accuracy of the prediction model.

[0026] Preferably, the neural network prediction model uses the historical maximum water storage capacity of the mine, the drainage capacity of the mine, the terrain characteristics of the mine and the surrounding area, and the precipitation information data as a training set, and uses the historical mine waterlogging level as the target result through error correction feedback learning.

[0027] By adopting the above technical solution, during the learning process of the neural network prediction model, the error is fed back and the weight of each input sample is adjusted in time according to the fed back error information, so that the neural network prediction model can more accurately reflect the relationship between the input value and the output value, and the prediction result can be closer to the true value.

[0028] Preferably, the step of issuing a safety warning according to the disaster probability specifically includes the following steps:

[0029] Determine the safety warning level corresponding to the prediction result according to a preset safety level table;

[0030] Corresponding warning information is issued according to the safety warning level to prompt personnel to respond.

[0031] By adopting the above technical solution, different safety warning information is issued according to different prediction results, so that personnel can plan targeted strategies according to different safety warning information, effectively warn personnel about possible flood disasters, and enable personnel to respond to dangerous situations in a timely manner.

[0032] In the second aspect, a mine safety early warning system provided by the present application includes the following modules:

[0033] Aerial global image acquisition module, used to obtain aerial global images of the mine area and the mine surrounding area;

[0034] A three-dimensional model generation module, used to establish a three-dimensional model of the area where the mine is located and the area surrounding the mine based on the aerial global image;

[0035] A precipitation information acquisition module is used to obtain precipitation information in the area where the mine is located and the area surrounding the mine;

[0036] A neural network prediction module is used to predict the water accumulation level of the mine through a neural network prediction model;

[0037] The safety warning module is used to issue safety warnings based on the prediction results.

[0038] In a third aspect, the present application provides a computer device, which adopts the following technical solution: it includes a memory and a processor, and the memory stores a computer program that can be loaded by the processor and execute any of the above-mentioned mine safety early warning methods.

[0039] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: storing any of the above-mentioned mine safety warning programs that can be loaded and executed by a processor.

[0040] In summary, the present application includes at least one of the following beneficial technical effects:

[0041] 1. The neural network prediction model is used to predict the possibility of mine disasters, and timely predicts mine flood disasters caused by precipitation in surrounding areas, reminding personnel to deal with disasters in a timely manner;

[0042] 2. Issue different safety warnings to personnel based on different prediction results, which can prompt personnel to take corresponding measures according to different disaster probabilities, so that personnel can solve possible dangerous situations in a targeted manner;

[0043] 3. Use drones to obtain global aerial images of the mine and its surroundings, build a three-dimensional model of the mine and its surroundings, and use the three-dimensional model to analyze data such as the terrain characteristics of the mine and its surrounding areas, the maximum water storage capacity of the mine, etc., so that personnel can have a more intuitive understanding of the geographical conditions of the mine. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a method flow chart of a mine safety early warning method in an embodiment of the present application.

[0045] Figure 2 It is a system block diagram of a mine safety early warning method according to an embodiment of the present application.

[0046] Figure 3It is a flow chart of a climate drone cruise strategy of a mine safety early warning method in an embodiment of the present application.

[0047] Figure 4 It is a structural schematic diagram of an electronic device according to an embodiment of the present application.

[0048] Explanation of reference numerals: 201, aerial global image acquisition module; 202, three-dimensional model generation module; 203, precipitation information acquisition module; 204, neural network prediction module; 205, safety warning module; 206, unmanned aerial vehicle; 400, electronic device; 401, processor; 402, communication bus; 403, user interface; 404, network interface; 405, memory. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0050] The embodiments of the present application disclose a mine safety early warning method, system, equipment and medium.

[0051] Reference Figure 1 , a mine safety early warning method comprises the following steps:

[0052] S101: Obtaining an aerial global image of the area where the mine is located and the area surrounding the mine by using the drone 206;

[0053] Specifically, since the mine and its surrounding areas may be affected by climate or human factors, such as personnel construction, which may cause changes in the terrain characteristics of the mine and its surrounding areas, the drone 206 can be set to obtain the aerial global image of the mine and its surrounding areas on a regular basis. The drone 206 cruises the mine and its surrounding areas at fixed intervals according to the preset drone 206 cruise table. For example, the drone 206 can be set to cruise regularly every six months. The drone 206 shoots the aerial global image of the mine and its surrounding areas according to the pre-planned shooting path. The set drone 206 cruise mode can realize the regular grasp of the real-time situation of the mine and its surrounding areas, providing an accurate and practical data basis for subsequent steps.

[0054] The mine and its surrounding areas are photographed by the camera mechanism set on the drone 206. The camera mechanism of the drone 206 is provided with five cameras, one of which is installed at the bottom of the drone 206, and the camera angle is perpendicular to the ground. The remaining four cameras are distributed around the drone 206, and their camera angles are at a certain inclination angle to the ground. The drone 206 takes specific images of the mine and its surrounding areas by oblique photography, including but not limited to the dam of the mine, the terrain features of the surrounding areas of the mine, the detailed texture of the mine and its surrounding areas, etc. The surrounding areas of the mine can be set as all areas within 10 kilometers around the mine. The multiple cameras set on the drone 206 can more clearly reflect the real situation of the mine and its surrounding areas, and the elements contained in the mine and its surrounding areas can be directly measured through the camera images obtained by the drone 206.

[0055] The aerial global image taken by the drone 206 can be first stored on a storage card inside the drone 206. When the drone 206 returns, personnel receive the captured control global image by reading the data on the storage card. A wireless communication module can also be installed on the drone 206, and the aerial global image taken by the drone 206 can be directly wirelessly transmitted to the aerial global image acquisition module 201.

[0056] S102: Establishing a three-dimensional model of the area where the mine is located and the area surrounding the mine based on the global aerial image;

[0057] Specifically, after obtaining images of the mine and its surrounding areas, it is necessary to use the acquired data to perform three-dimensional modeling of the mine and its surrounding areas; import the acquired image information into three-dimensional modeling software, and establish a three-dimensional model of the mine and its surrounding areas in the three-dimensional modeling software. The three-dimensional modeling software can be PhotoScan.

[0058] The three-dimensional model is regularly updated based on the images of the current mine and its surrounding areas taken by the UAV 206 during the cruise, so that the three-dimensional model is closer to the actual situation and the accuracy of subsequent predictions is guaranteed.

[0059] S103: Obtain precipitation information of the area where the mine is located and the area surrounding the mine;

[0060] Specifically, precipitation information of the surrounding areas is obtained, which includes but is not limited to data such as precipitation amount and precipitation time. All precipitation information of the areas surrounding the mine can be fully obtained, or only precipitation information of several areas can be obtained after roughly analyzing the areas where the surrounding precipitation has the greatest impact on the mine based on the three-dimensional model.

[0061] Precipitation information can be obtained through climate drones, which regularly patrol specific nodes set by personnel in the area around the mine. These specific nodes are delineated based on the three-dimensional modeling of the area around the mine; reference Figure 3 The cruising strategy of the climate drone for characteristic nodes is as follows: when precipitation occurs in the area where the mine is located, the climate drone immediately starts to cruise the specific node, and the interval between each cruise is 1 hour; if there is no precipitation in the area where the mine is located, the climate drone cruises according to the start command, and the start command can be generated based on three criteria: first, when personnel perceive that there is abnormal climate or possible flooding in the area where the mine is located, they send a signal through a controller such as a mobile phone to command the climate drone to cruise; second, receiving weather forecast information, by identifying keywords in the weather forecast information, determining that precipitation occurs in the designated area around the mine, generating a start command, starting the climate drone to cruise, and periodically monitoring the designated nodes according to the preset path, and the monitoring period can be 2 hours; third, when the temperature sensor installed in the mine detects a large temperature change in the mine, a start command is generated and the climate drone cruises. Since precipitation in the area surrounding the mine will have an impact on the area where the mine is located, when the temperature in the area where the mine is located changes, it means that precipitation may be occurring or will occur in the area surrounding the mine. The temperature changes mentioned above will vary depending on the season and time of the area where the mine is located, so it can be set by technical personnel in this field according to actual conditions.

[0062] After the climate drone arrives at the designated node, the designated node can be arranged by personnel, and its penetration effect is small and negligible. The ground water accumulation information is photographed, and the ground water accumulation information is image information. The ground water accumulation information is sent to the computer for analysis through wireless transmission, and the precipitation is obtained through a preset analysis model, which can be a pre-trained neural network model.

[0063] S104: importing the three-dimensional model and the precipitation information of the area where the mine is located and the area surrounding the mine into a preset neural network prediction model to predict the water accumulation level of the mine;

[0064] Specifically, before predicting the probability of mine disasters through the neural network prediction model, it is necessary to process various data of the mine. According to the generated three-dimensional model, the maximum water storage capacity of the mine is calculated, the drainage capacity of the mine is analyzed, and the terrain characteristics of the area around the mine are analyzed. The drainage capacity of the mine and the terrain characteristics of the area around the mine are evaluated by personnel according to the indicator system. The maximum water storage capacity of the mine can be constructed through the data analysis plug-in in the three-dimensional model. The irregular triangulated network calculation model is made to make the calculation model close to the mine terrain, and then the volume of the triangulated network is calculated to obtain the maximum water storage capacity of the mine.

[0065] The preset neural network prediction model adopts BP neural network, takes the five main data that affect the probability of mine disasters, namely historical precipitation, precipitation time, mine terrain information, maximum mine water storage capacity and mine drainage capacity, as the training set, and the historical mine water accumulation level as the target output value. The neural network prediction model is trained until the mean square error between the generated result and the ideal value is less than a certain value, completing the initialization of the preset neural network prediction model.

[0066] After the above process, the preset neural network prediction model completed the initialization during the error correction process. At this time, the current precipitation information, mine terrain information, maximum mine water storage capacity and mine drainage capacity data are input into the neural network prediction model to obtain the water accumulation level of the mine within a certain period of time after precipitation occurs in the mine area and the surrounding areas.

[0067] It should be noted that the water accumulation level is an assessment of the overall water accumulation in a mine. Taking into account the maximum water storage capacity of the mine, the water accumulation level can be divided into level one, level two, level three and level four. The water accumulation level reflects the water accumulation situation in a mine under the influence of precipitation. The higher the water accumulation level, the greater the impact of precipitation in the area where the mine is located or in the surrounding area, and the greater the probability of flood disasters in the mine.

[0068] S105: Issue a safety warning based on the prediction results;

[0069] Specifically, after the mine's water accumulation level is determined, the mine's maximum water storage capacity and drainage capacity are comprehensively considered to assess possible flooding in the mine after precipitation in the mine or its surrounding areas; a safety warning is issued based on the predicted mine water accumulation level and the safety warning rules formulated in advance. For example, when the mine's water accumulation level is predicted to be level one, it can be determined that the precipitation has no effect on the mine's drainage capacity and the probability of a disaster is low. In this case, no safety warning will be issued to personnel; when the mine's water accumulation level is predicted to be level two, A preliminary safety warning is sent to the smart terminal of mine managers to remind them to pay continuous attention to the water accumulation level of subsequent mines and be aware of the occurrence of dangerous situations. When the water accumulation level of the mine is predicted to be level three, a medium safety warning is sent to the smart terminals of mine and upstream mining factory personnel to explain the current water storage situation of the mine and remind personnel to make flood prevention preparations as soon as possible. When the water accumulation level of the mine is predicted to be level four, the highest safety warning is issued, an obvious sound alarm is emitted, prompting personnel to evacuate in time and sending the current status information of the mine to relevant disaster handling units.

[0070] The implementation principle of a mine safety early warning method in an embodiment of the present application is as follows: obtaining an aerial global image of the mine and its surrounding area by means of a drone 206 cruise, importing the obtained image data into a three-dimensional modeling software to generate a three-dimensional model of the mine and its surrounding area, obtaining precipitation information of the mine and its surrounding area, extracting characteristic values ​​that may cause mine disasters based on the three-dimensional model and the precipitation information, inputting the characteristic values ​​as input quantities into a preset neural network prediction model, predicting the water accumulation level of the mine, and generating different early warning information based on the prediction results to prompt personnel to take targeted measures for possible dangerous situations.

[0071] The present application also discloses a mine safety early warning system. Figure 2 , a mine safety early warning system includes the following modules:

[0072] Aerial global image acquisition module 201, used to acquire aerial global images of the mine area and the mine surrounding area;

[0073] A three-dimensional model generation module 202 is used to establish a three-dimensional model of the area where the mine is located and the area surrounding the mine based on the global aerial image;

[0074] The precipitation information acquisition module 203 is used to acquire precipitation information in the area where the mine is located and the area surrounding the mine;

[0075] A neural network prediction module 204 is used to predict the water accumulation level of the mine through a neural network prediction model;

[0076] The safety warning module 205 is used to issue a safety warning according to the prediction result.

[0077] See also Figure 4 , is a schematic diagram of the structure of an electronic device 400 provided in an embodiment of the present application. Figure 4 As shown, the electronic device 400 may include: at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 405 , and at least one communication bus 402 .

[0078] The communication bus 402 is used to realize the connection and communication between these components.

[0079] The user interface 403 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0080] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0081] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 401 can integrate one or a combination of a central processing unit 401 (Central Processing Unit, CPU), an image processor 401 (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented separately through a chip.

[0082] Among them, the memory 405 may include a random access memory 405 (Random Access Memory, RAM), and may also include a read-only memory 405 (Read-Only Memory). Optionally, the memory 405 includes a non-transitory computer-readable storage medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 405 may also be optionally at least one storage device located away from the aforementioned processor 401. As Figure 4 As shown, the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface 403 module and an application program of a mine safety early warning method.

[0083] exist Figure 4In the electronic device 400 shown, the user interface 403 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 401 can be used to call an application of a mine safety warning method stored in the memory 405. When executed by one or more processors 401, the electronic device 400 executes one or more methods described in the above embodiments.

[0084] A storage medium readable by an electronic device 400 stores instructions. When executed by one or more processors 401, the electronic device 400 executes one or more of the methods described in the above embodiments.

[0085] The embodiment of the present application also discloses a computer-readable storage medium.

[0086] Specifically, the computer-readable storage medium stores a computer program that can be loaded by the processor 401 and execute a mine safety warning method such as the above-mentioned one. The computer-readable storage medium includes, for example: a U disk, a mobile hard disk, a read-only memory 405 (Read-Only Memory, ROM), a random access memory 405 (Random Access Memory, RAM), a disk or an optical disk, and other media that can store program codes.

[0087] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A mine safety early warning method, characterized in that: The method comprises the following steps: Obtaining an aerial global image of the area where the mine is located and the area surrounding the mine by using a drone (206); Establishing a three-dimensional model of the area where the mine is located and the area surrounding the mine based on the aerial global image; Obtain precipitation information in the area where the mine is located and the area surrounding the mine; Importing the three-dimensional model and precipitation information of the area where the mine is located and the area surrounding the mine into a preset neural network prediction model, and outputting a prediction result, wherein the prediction result is the water accumulation level of the mine; issuing a safety warning based on the prediction results, In the step of importing the three-dimensional model and the precipitation information of the area where the mine is located and the area surrounding the mine into a preset neural network prediction model, and outputting the prediction result, the steps specifically include: Extracting the topographic features of the mine and the area surrounding the mine according to the three-dimensional model; Calculate the maximum water storage capacity and drainage capacity of the mine based on the three-dimensional model; The maximum water storage capacity of the mine, the drainage capacity of the mine, the terrain characteristics of the mine and the surrounding area of ​​the mine, and the precipitation information are input into the neural network prediction model as input data, and the prediction result is output. The neural network prediction model uses the historical maximum water storage capacity of the mine, the drainage capacity of the mine, the terrain characteristics of the mine and the surrounding area, and the precipitation information data as a training set, and uses the historical mine waterlogging level as the target result through error correction feedback learning.

2. A mine safety early warning method according to claim 1, characterized in that: The camera device of the drone (206) comprises four inclined cameras distributed around the drone (206) and forming a certain inclination angle with the ground, and a vertical camera arranged at the bottom of the drone (206) and perpendicular to the ground.

3. A mine safety early warning method according to claim 1, characterized in that: The step of obtaining the global aerial image of the mine area and the mine surrounding area by using the drone (206) specifically includes the following steps: Determining a cruise cycle of the drone (206) according to a preset cruise table of the drone (206); According to the cruise cycle of the drone (206), aerial global images of the area where the mine is located and the area surrounding the mine are taken.

4. A mine safety early warning method according to claim 1, characterized in that: The three-dimensional model of the area where the mine is located and the area surrounding the mine and the neural network prediction model are iterated based on the real-time global aerial images taken by the drone (206) during regular patrols.

5. A mine safety early warning method according to claim 1, characterized in that: The step of issuing a safety warning according to the disaster probability specifically includes the following steps: Determine the safety warning level corresponding to the prediction result according to a preset safety level table; Corresponding warning information is issued according to the safety warning level to prompt personnel to respond.

6. A mine safety early warning system based on the mine safety early warning method according to any one of claims 1 to 5, the system comprising the following modules: An aerial global image acquisition module (201) is used to acquire an aerial global image of the area where the mine is located and the area surrounding the mine; A three-dimensional model generation module (202) is used to establish a three-dimensional model of the area where the mine is located and the area surrounding the mine based on the aerial global image; A precipitation information acquisition module (203) is used to acquire precipitation information in the area where the mine is located and the area surrounding the mine; A neural network prediction module (204), used to predict the water accumulation level of the mine through a neural network prediction model; The safety warning module (205) is used to issue a safety warning according to the prediction result.

7. A computer device, characterized in that: The invention comprises a memory (405) and a processor (401), wherein the memory (405) stores a computer program which can be loaded by the processor (401) and executes any one of the methods according to claims 1 to 5.

8. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor (401) and executes the method according to any one of claims 1 to 5.

Images