A method and device for spontaneous combustion monitoring and fire extinguishing in goaf based on multi-parameter fusion

Abstract

The present application provides a method and device for goaf spontaneous combustion monitoring and fire extinguishing based on multi-parameter fusion, which relates to the technical field of coal seam spontaneous combustion monitoring. The method and device obtain the spontaneous combustion area image of the goaf, use the spontaneous combustion area image of the goaf, use an edge detection algorithm to obtain the edge pixel points of the edge of the high-temperature area, and obtain the temperature grayscale value of the pixel points in the high-temperature area, calculate the distance from each non-high-temperature area pixel point outside the edge of the high-temperature area to the high-temperature area, obtain the wind factor, calculate the wind force through the wind factor, obtain the spontaneous combustion spread factor, determine the probability of spontaneous combustion spread of the pixel points in the non-high-temperature area in the future through the spontaneous combustion spread factor, divide the non-high-temperature area into grids, use the probability of spontaneous combustion spread of the pixel points in the non-high-temperature area to calculate the average probability of spontaneous combustion spread in the grid, cluster the grids according to the average probability of spontaneous combustion spread in the grid, obtain the high-risk area for spontaneous combustion spread, and realize early warning and precise rescue of spontaneous combustion in the goaf.

Description

Technical Field

[0001] The present application relates to the technical field of coal seam spontaneous combustion monitoring, and in particular to a method and device for spontaneous combustion monitoring and fire extinguishing in goafs based on multi-parameter fusion. Background Art

[0002] Goaf is the space formed after coal mining. Since the residual coal seam is prone to spontaneous combustion, it can cause fires and bring hidden dangers to coal mine production safety. Spontaneous combustion in goaf is affected by many factors such as ventilation and coal seam composition. The fire often spreads quickly and over a wide range, making it difficult to extinguish the fire.

[0003] In the existing goaf spontaneous combustion monitoring system, only a single parameter such as temperature is usually monitored. However, due to the complex environment of the goaf and the fact that the spontaneous combustion process is affected by many factors, it is difficult for a single parameter to accurately assess the development trend of the fire. For example, changes in wind flow will aggravate the spontaneous combustion of coal seams, and the concentration of toxic and harmful gases generated by combustion can also reflect the fire situation. It is impossible to accurately predict the high-risk areas of spontaneous combustion with a single parameter alone. Therefore, it is necessary to integrate multiple parameters such as temperature, wind speed, gas concentration, and combine them with intelligent algorithms to achieve accurate early warning of spontaneous combustion fires. In addition, most of the existing systems lack linkage with fire extinguishing devices, making it difficult to achieve accurate fire extinguishing. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems in the related art to a certain extent. To this end, one purpose of the present application is to propose a method and device for monitoring and extinguishing spontaneous combustion in goaf based on multi-parameter fusion, so as to achieve early warning and precise rescue of spontaneous combustion in goaf.

[0005] The first aspect of the present application provides a method for spontaneous combustion monitoring and fire extinguishing in goaf based on multi-parameter fusion, comprising:

[0006] Step S1: Acquire the spontaneous combustion area image of the goaf;

[0007] The method for obtaining the spontaneous combustion area image of the goaf is:

[0008] Obtain historical high temperature area images and historical normal temperature images, and use the historical high temperature area images and historical normal temperature images to train a neural network model for determining the category of spontaneous combustion fire;

[0009] The goaf image is used as the input of the neural network model to determine the spontaneous combustion fire category of the current goaf. If there is spontaneous combustion, a fire warning is triggered;

[0010] When a fire warning is triggered, the drone equipment responds to the fire warning and flies over the goaf to take images of the spontaneous combustion area of ​​the goaf;

[0011] The historical high temperature area images include spontaneous combustion area images of goaf under different environmental conditions;

[0012] The historical normal temperature images include normal temperature images of goaf areas under different environmental conditions;

[0013] The spontaneous combustion fire category is divided into: 1 means spontaneous combustion, 0 means no spontaneous combustion;

[0014] The training process of the neural network model for determining the category of spontaneous combustion fire is as follows:

[0015] Perform data enhancement on historical high temperature area images and historical normal temperature images;

[0016] The historical high temperature area images and historical normal temperature images are respectively labeled with corresponding spontaneous combustion fire category labels, and the historical high temperature area images and historical normal temperature images with spontaneous combustion fire category labels are used as input image data for training the neural network model. For each input historical high temperature area image and historical normal temperature image, its spontaneous combustion fire category is determined as the output of the neural network model; the number of the i-th input image data is recorded as i, and there are n input image data in total, y i represents the true spontaneous combustion fire category of the i-th input image data, It indicates that the neural network model outputs the spontaneous combustion fire category of the i-th input image data. The accurate judgment of the true spontaneous combustion fire category of the goaf is taken as the training goal of the model, which will reduce the loss function between the prediction result and the true label. As the prediction target for training; the calculation formula of the loss function is

[0017]

[0018] Step S2: using the spontaneous combustion area image of the goaf, using an edge detection algorithm to obtain edge pixel points at the edge of the high-temperature area, and obtaining the temperature grayscale value H of the pixel points in the high-temperature area;

[0019] The specific steps of using the edge detection algorithm to obtain edge pixel points at the edge of the high temperature area and obtaining the temperature grayscale value H of the pixel points in the high temperature area are:

[0020] Step S21: converting the spontaneous combustion area image of the goaf into a grayscale image;

[0021] Step S22: performing smoothing filtering on the grayscale image, removing noise with Gaussian filtering, and obtaining a smoothed image;

[0022] Step S23: Based on the smoothed image, the gradient magnitude and gradient direction are calculated using the Sobel gradient operator;

[0023] Step S24: non-maximum suppression is performed on the pixel points of the spontaneous combustion area image of the goaf along the gradient direction, the pixel points with the largest gradient amplitude are retained, and the pixel points in other directions are suppressed;

[0024] Step S25: dividing the spontaneous combustion area image of the goaf after non-maximum suppression into three categories: strong edge pixels, weak edge pixels and non-edge pixels;

[0025] Step S26: connecting the strong edge pixel points and the weak edge pixel points adjacent thereto to form a continuous edge line;

[0026] Step S27: outputting an image including a high temperature area, a non-high temperature area, and edge lines after edge detection processing;

[0027] Among them, the edge pixel points of the edge lines of the spontaneous combustion area image of the goaf after edge detection processing are highlighted;

[0028] Step S28: using the edge lines and the grayscale image to obtain the temperature grayscale value H of the pixel point in the high temperature area.

[0029] The Sobel gradient operator is an operator used for edge detection, which is divided into a horizontal kernel and a vertical kernel, and is used to detect edge pixel points in the horizontal and vertical directions in the spontaneous combustion area image of the goaf, respectively;

[0030] The method for calculating the image gradient amplitude and gradient direction using the Sobel gradient operator is as follows: placing the Sobel kernel at a certain pixel position of the spontaneous combustion area image of the goaf, taking the pixel as the center, performing element-by-element multiplication and addition of the horizontal kernel and the vertical kernel with the corresponding pixel on the image, respectively, to obtain the horizontal gradient and the vertical gradient; calculating the square root of the sum of the squares of the horizontal gradient and the vertical gradient to obtain the gradient amplitude, and calculating the inverse tangent function values ​​of the horizontal and vertical gradients to obtain the gradient direction;

[0031] The method for performing non-maximum suppression on the pixel points of the spontaneous combustion area image of the goaf along the gradient direction is: obtaining edge pixel points, if the gradient amplitude of the current pixel point is greater than the gradient amplitude of the two adjacent pixel points in the positive and negative directions of the gradient, then the point is considered to be an edge pixel point, and its corresponding position is marked as 1, otherwise the point is suppressed as a non-edge pixel point, and its corresponding position is marked as 0.

[0032] The method for finding two pixel points adjacent to each other in the gradient direction is: judging the positions of two pixel points adjacent to the pixel point according to the angle between the pixel point gradient direction and the horizontal direction.

[0033] The classification method for dividing the spontaneous combustion area image of the goaf after non-maximum suppression into three categories: strong edge pixels, weak edge pixels and non-edge pixels is:

[0034] Set a high threshold and a low threshold, and compare the gradient amplitude of the pixel points in the spontaneous combustion area image of the goaf with the size of the two thresholds:

[0035] If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf exceeds the high threshold, it is set as a strong edge pixel point;

[0036] If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf is between the high threshold and the low threshold, it is set as a weak edge pixel point;

[0037] If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf is lower than the low threshold, it is set as a non-edge pixel point.

[0038] The method of connecting strong edge pixels and their adjacent weak edge pixels to form continuous edge lines is:

[0039] Step S210: Finding pixels marked as strong edge pixels according to the classification results of strong edge pixels, weak edge pixels and non-edge pixels;

[0040] Step S220: for each strong edge pixel, check its eight neighboring pixels to find the neighboring pixels marked as weak edge pixels;

[0041] Step S230: performing edge tracing for each strong edge pixel and its adjacent weak edge pixel, connecting them to form a continuous edge line; preferably, edge tracing is performed in a recursive manner;

[0042] Step S240: During edge tracing, the connected pixels are marked as connected to avoid repeated tracing;

[0043] Step S250: Repeat steps S220 to S240 until all strong edge pixels are processed.

[0044] Furthermore, the method of selecting a recursive method to perform edge tracking in step S230 is:

[0045] Step S231: Select a strong edge pixel point as a starting point;

[0046] Step S232: starting from the starting point, searching for adjacent weak edge pixels;

[0047] Step S233: for each adjacent weak edge pixel point, mark it as a connected edge, and recursively continue edge tracing with the weak edge pixel point as a new starting point;

[0048] Step S234: Repeat step S233 until there are no more weak edge pixels to track.

[0049] The method for obtaining the temperature grayscale value H of the high-temperature area pixel point is: according to the grayscale image converted from the spontaneous combustion area image of the goaf, the temperature grayscale value H of each high-temperature area pixel point within the edge of the high-temperature area on the grayscale image is obtained.

[0050] Step S3: Calculate the distance D between each pixel point in the non-high temperature area outside the edge of the high temperature area and the high temperature area;

[0051] The method for calculating the distance D from each pixel point in the non-high temperature area outside the edge of the high temperature area to the high temperature area is:

[0052] Calculate the distance from each non-high temperature area pixel outside the edge of the high temperature area to each edge pixel at the edge of the high temperature area, select the edge pixel at the edge of the high temperature area with the smallest distance, and use the distance between the edge pixel and the non-high temperature area pixel as the distance D from the non-high temperature area pixel to the high temperature area.

[0053] Step S4: Obtain wind force factors, and calculate wind force F based on the wind force factors;

[0054] The wind factors include: wind speed data s, wind direction data x, and air density data m;

[0055] The method for obtaining the wind force factor and calculating the wind force F by the wind force factor is:

[0056] Obtain real-time wind speed data s, wind direction data x, and air density data m;

[0057] The wind force is calculated using wind speed data s, wind direction data x, and air density data m. The formula is wind force F = 0.5 × m × s 2 ×cosx;

[0058] The wind direction data x refers to the angle x between the wind direction and the direction from the edge pixel point to the non-high temperature area pixel point;

[0059] The wind force F is a wind vector in the distance direction from the edge pixel point to the pixel point in the non-high temperature area.

[0060] Step S5: Obtain the spontaneous combustion propagation factor, and determine the probability P of spontaneous combustion propagation of each pixel point in the non-high temperature area in the future time through the spontaneous combustion propagation factor;

[0061] The factors for the spread of spontaneous combustion include: the distance D from the pixel point in the non-high temperature area to the high temperature area, the wind force F, the temperature data T, the air visibility data V, and the temperature gray value H;

[0062] The calculation formula for the probability P of spontaneous combustion spreading of each pixel point in the non-high temperature area in the future time is: Where ε and δ are small constants used to avoid the denominator being zero, k is a scaling factor, and the temperature grayscale value H used for calculation is the temperature grayscale value H of the pixel in the high temperature area that is closest to the pixel in the non-high temperature area.

[0063] Step S6: Divide the non-high temperature area into grids, and use the probability of spontaneous combustion propagation of each pixel point in the non-high temperature area to calculate the average probability of spontaneous combustion propagation in each grid.

[0064] The non-high temperature area is divided into grids, and the probability of spontaneous combustion propagation of each pixel point in the non-high temperature area is used to calculate the average probability of spontaneous combustion propagation in each grid. The method is:

[0065] The non-high temperature area of ​​the spontaneous combustion area image of the goaf is divided into grids according to the grid size; preferably, the non-high temperature area of ​​the spontaneous combustion area image of the goaf is divided into a plurality of grids according to the grid size a×a;

[0066] Map irregular non-high temperature areas to a grid;

[0067] Randomly select n non-high temperature area pixel points in a non-high temperature area of ​​a grid; preferably, the random method selects a random number generator;

[0068] Calculate the probability P of spontaneous combustion spreading for each pixel point in the selected non-high temperature area;

[0069] Calculate the average probability of spontaneous combustion spread of n non-high temperature area pixels in the future as the average probability of spontaneous combustion spread in the grid

[0070] Step S7: According to the average probability of spontaneous combustion spread in the grid Cluster the grids to obtain high-risk areas for spontaneous combustion spread;

[0071] The average probability of spontaneous combustion propagation in the grid The method to cluster the raster is:

[0072] The average probability of spontaneous combustion spread of each grid is taken as a feature vector to form a data set;

[0073] Use clustering algorithms to divide the rasters in the dataset into different clusters according to probability;

[0074] For all clusters, the average probability of spontaneous combustion spread is To classify;

[0075] According to the classification level, high-risk areas for spontaneous combustion spread are displayed to users in a visual way;

[0076] The classification method is as follows: the cluster with level A is set as a risk-free area; the cluster with level B is set as a low-risk area; the cluster with level C is set as a medium-risk area; the cluster with level D is set as a high-risk area; preferably, the risk-free area refers to the average probability The range is between 0 and 20%; the low risk area refers to the average probability The range is between 20% and 50%; the medium risk area refers to the average probability The range is between 50% and 80%; the high-risk area refers to the average probability The range is between 80% and 100%.

[0077] The second aspect of the present application provides a goaf spontaneous combustion monitoring and fire extinguishing device based on multi-parameter fusion, comprising:

[0078] A spontaneous combustion image acquisition module, used to acquire the spontaneous combustion area image of the goaf;

[0079] The edge detection module is used to obtain edge pixel points at the edge of the high-temperature area using the image of the spontaneous combustion area of ​​the goaf and the edge detection algorithm, and obtain the temperature gray value H of the pixel points in the high-temperature area;

[0080] A distance calculation module is used to calculate the distance D between each pixel point in the non-high temperature area outside the edge of the high temperature area and the high temperature area;

[0081] A wind force calculation module is used to obtain wind force factors and calculate the wind force F based on the wind force factors;

[0082] The spread probability calculation module is used to obtain the spontaneous combustion spread factor and determine the probability P of spontaneous combustion spread of each pixel point in the non-high temperature area in the future through the spontaneous combustion spread factor;

[0083] The average probability calculation module is used to divide the non-high temperature area into grids, and use the probability of spontaneous combustion spread of each pixel point in the non-high temperature area to calculate the average probability of spontaneous combustion spread in each grid;

[0084] The high-risk area clustering module is used to cluster the grids according to the average probability of spontaneous combustion spread in the grids to obtain the high-risk areas for spontaneous combustion spread.

[0085] One aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in a method for spontaneous combustion monitoring and extinguishing of goaf based on multi-parameter fusion are implemented.

[0086] One aspect of the present application provides a readable storage medium storing a computer program suitable for loading by a processor to execute steps in a method for spontaneous combustion monitoring and extinguishing of goaf based on multi-parameter fusion.

[0087] Compared with the prior art, the goaf spontaneous combustion monitoring and fire extinguishing method based on multi-parameter fusion proposed in this application has the following advantages:

[0088] The present invention integrates multiple parameters such as distance, wind speed, temperature, visibility, spontaneous combustion high temperature gray value, etc., and comprehensively considers various factors that affect the spread of spontaneous combustion in goaf areas. By comprehensively analyzing these parameters, the development trend and spread probability of spontaneous combustion in goaf areas can be more accurately predicted, providing a reliable basis for timely and effective fire extinguishing measures.

[0089] The present invention can discretize continuous areas by dividing non-spontaneous combustion areas into grids and calculating the average probability of spontaneous combustion spread in each grid, so as to facilitate analysis and processing. At the same time, the grids are grouped by clustering algorithms, which can automatically identify high-risk areas for spontaneous combustion spread, thus realizing refined identification and management of spontaneous combustion areas.

[0090] The present invention proposes to classify the areas according to the probability of spontaneous combustion spread and present it to the user in a visual way. Through intuitive visualization methods such as heat maps and contour maps, users can quickly understand the spontaneous combustion spread inside the goaf, timely grasp the distribution of high-risk areas, and facilitate the formulation of targeted fire extinguishing plans.

[0091] Since the present invention uses real-time data such as wind speed, wind direction, and temperature, it can dynamically reflect environmental changes within the goaf. When changes in the spontaneous combustion area are detected, the system can quickly recalculate the probability of spontaneous combustion spread in each area, update the prediction results in real time, and provide support for timely adjustment of fire extinguishing strategies.

[0092] The various steps and calculation models in the present invention are highly versatile and can adapt to different goaf environments. By adjusting the model parameters and threshold settings, the system can be adapted to different geographical conditions, climate characteristics and spontaneous combustion characteristics, thereby improving the practicality and scalability of the solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 It is a flow chart of a method for spontaneous combustion monitoring and fire extinguishing in goaf based on multi-parameter fusion;

[0094] Figure 2 It is a visualization diagram of clustering results of a method for spontaneous combustion monitoring and extinguishing in goaf based on multi-parameter fusion;

[0095] Figure 3It is a functional module diagram of a goaf spontaneous combustion monitoring and fire extinguishing device based on multi-parameter fusion;

[0096] Figure 4 A schematic diagram of the structure of an electronic device provided in this application;

[0097] Figure 5 It is a schematic diagram of the structure of a readable storage medium provided by this application. DETAILED DESCRIPTION

[0098] In order to better understand the present application, various aspects of the present application will be described in more detail with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present application and do not limit the scope of the present application in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0099] In the accompanying drawings, the size, dimensions and shapes of the elements have been slightly adjusted for ease of illustration. The drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in this application, the order in which the steps are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0100] It should also be understood that expressions such as "including", "comprising", "having", "containing" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present application, "may" is used to mean "one or more embodiments of the present application". And, the term "exemplary" is intended to refer to an example or illustration.

[0101] Unless otherwise specified, all words (including engineering terms and scientific and technological terms) used in this article have the same meaning as those commonly understood by ordinary technicians in the field to which this application belongs. It should also be understood that, unless clearly stated in this application, words defined in common dictionaries should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0102] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in 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.

[0103] Example 1

[0104] like Figure 1 As shown, a method for spontaneous combustion monitoring and fire extinguishing in goaf based on multi-parameter fusion includes the following steps:

[0105] Step S1: Acquire the spontaneous combustion area image of the goaf;

[0106] The method for obtaining the spontaneous combustion area image of the goaf is:

[0107] Obtain historical high temperature area images and historical normal temperature images, and use the historical high temperature area images and historical normal temperature images to train a neural network model for determining the category of spontaneous combustion fire;

[0108] The goaf image is used as the input of the neural network model to determine the spontaneous combustion fire category of the current goaf. If there is spontaneous combustion, a fire warning is triggered;

[0109] When a fire warning is triggered, the drone equipment responds to the fire warning and flies over the goaf to take images of the spontaneous combustion area of ​​the goaf;

[0110] The historical high temperature area images include spontaneous combustion area images of goaf under different environmental conditions;

[0111] The historical normal temperature images include normal temperature images of goaf areas under different environmental conditions;

[0112] The spontaneous combustion fire category is divided into: 1 means spontaneous combustion, 0 means no spontaneous combustion;

[0113] The training process of the neural network model for determining the category of spontaneous combustion fire is as follows:

[0114] Perform data enhancement on historical high temperature area images and historical normal temperature images;

[0115] The historical high temperature area images and historical normal temperature images are respectively labeled with corresponding spontaneous combustion fire category labels, and the historical high temperature area images and historical normal temperature images with spontaneous combustion fire category labels are used as input image data for training the neural network model. For each input historical high temperature area image and historical normal temperature image, its spontaneous combustion fire category is determined as the output of the neural network model; the number of the i-th input image data is recorded as i, and there are n input image data in total, y i represents the true spontaneous combustion fire category of the i-th input image data, It indicates that the neural network model outputs the spontaneous combustion fire category of the i-th input image data. The accurate judgment of the true spontaneous combustion fire category of the goaf is taken as the training goal of the model, which will reduce the loss function between the prediction result and the true label. As the prediction target for training; the calculation formula of the loss function is

[0116]

[0117] The data enhancement of the historical high temperature area images and the historical normal temperature images refers to: scaling and cropping the historical high temperature area images and the historical normal temperature images, and randomly flipping them horizontally or vertically to increase the diversity of the historical high temperature area images and the historical normal temperature images;

[0118] The labeling of the historical high temperature area images and the historical normal temperature images with the corresponding spontaneous combustion fire category labels respectively means labeling the historical high temperature area images as having spontaneous combustion, and labeling the historical normal temperature images as having no spontaneous combustion;

[0119] The goaf spontaneous combustion area image is an image of the entire goaf including the spontaneous combustion area, including the high temperature area and the normal temperature area around the high temperature area;

[0120] The goaf image is obtained by monitoring equipment;

[0121] The images of the spontaneous combustion area in the goaf are obtained by taking pictures using drones. This technology is an existing technology and will not be described in detail here.

[0122] Step S2: using the spontaneous combustion area image of the goaf, using an edge detection algorithm to obtain edge pixel points at the edge of the high-temperature area, and obtaining the temperature grayscale value H of the pixel points in the high-temperature area;

[0123] Preferably, the edge detection algorithm adopts the Canny edge detection algorithm;

[0124] The specific steps of using the edge detection algorithm to obtain edge pixel points at the edge of the high temperature area and obtaining the temperature grayscale value H of the pixel points in the high temperature area are:

[0125] Step S21: converting the spontaneous combustion area image of the goaf into a grayscale image;

[0126] Step S22: performing smoothing filtering on the grayscale image, removing noise with Gaussian filtering, and obtaining a smoothed image;

[0127] Step S23: Based on the smoothed image, the gradient magnitude and gradient direction are calculated using the Sobel gradient operator;

[0128] Step S24: non-maximum suppression is performed on the pixel points of the spontaneous combustion area image of the goaf along the gradient direction, the pixel points with the largest gradient amplitude are retained, and the pixel points in other directions are suppressed;

[0129] Step S25: dividing the spontaneous combustion area image of the goaf after non-maximum suppression into three categories: strong edge pixels, weak edge pixels and non-edge pixels;

[0130] Step S26: connecting the strong edge pixel points and the weak edge pixel points adjacent thereto to form a continuous edge line;

[0131] Step S27: outputting an image including a high temperature area, a non-high temperature area, and edge lines after edge detection processing; wherein edge pixel points of edge lines of the spontaneous combustion area image of the goaf after edge detection processing are highlighted;

[0132] Step S28: using the edge lines and the grayscale image to obtain the temperature grayscale value H of the pixel point in the high temperature area.

[0133] The grayscale image refers to an image containing only brightness information. The value of each pixel of the spontaneous combustion area image of the goaf represents the grayscale level of the pixel of the spontaneous combustion area image of the goaf in terms of brightness, and is represented by an integer between 0 and 255, where 0 represents black, 255 represents white, and the indicators in between are different grayscale levels;

[0134] The Sobel gradient operator is an operator used for edge detection, which is divided into a horizontal kernel and a vertical kernel, and is used to detect edge pixel points in the horizontal and vertical directions in the spontaneous combustion area image of the goaf, respectively;

[0135] The gradient direction refers to the direction in which the grayscale changes fastest at a certain pixel point in the spontaneous combustion area image of the goaf;

[0136] The gradient amplitude refers to: the spatial variation amplitude of the gray value of a certain pixel point in the spontaneous combustion area image of the goaf;

[0137] The method for calculating the image gradient amplitude and gradient direction using the Sobel gradient operator is as follows: placing the Sobel kernel at a certain pixel position of the spontaneous combustion area image of the goaf, taking the pixel as the center, performing element-by-element multiplication and addition of the horizontal kernel and the vertical kernel with the corresponding pixel on the image, respectively, to obtain the horizontal gradient and the vertical gradient; calculating the square root of the sum of the squares of the horizontal gradient and the vertical gradient to obtain the gradient amplitude, and calculating the inverse tangent function values ​​of the horizontal and vertical gradients to obtain the gradient direction;

[0138] The method for performing non-maximum suppression on the pixel points of the spontaneous combustion area image of the goaf along the gradient direction is: obtaining edge pixel points, if the gradient amplitude of the current pixel point is greater than the gradient amplitude of the two adjacent pixel points in the positive and negative directions of the gradient, then the point is considered to be an edge pixel point, and its corresponding position is marked as 1, otherwise the point is suppressed as a non-edge pixel point, and its corresponding position is marked as 0.

[0139] It should be noted that non-maximum suppression is a conventional technical means of edge detection algorithm, and the present invention does not explain the principle here; but in order to facilitate the implementation of the present invention, the present invention provides the following exemplary description of non-maximum suppression on the spontaneous combustion area image of the goaf:

[0140] Assume that there is a pixel point (i, j) on the spontaneous combustion area image of the goaf, its gradient amplitude is G(i, j), and the gradient direction is θ.

[0141] Find two adjacent pixel points in the high temperature area in the gradient direction, such as neighboring point 1 (i+1, j) and neighboring point 2 (i-1, j). The gradient amplitude of neighboring point 1 is G(i+1, j), and the gradient amplitude of neighboring point 2 is G(i-1, j).

[0142] Compare the gradient amplitude G(i, j) of the pixel with the gradient amplitudes G(i-1, j) and G(i+1, j) of the two adjacent pixels. If the gradient amplitude G(i, j) of the pixel is greater than the gradient amplitudes G(i+1, j) and G(i-1, j) of the two adjacent pixels, the pixel (i, j) is an edge pixel and is marked as 1. If the gradient amplitude G(i, j) of the pixel is less than the gradient amplitudes G(i+1, j) and G(i-1, j) of the two adjacent pixels, the pixel (i, j) is a non-edge pixel and is marked as 0.

[0143] The method for finding two pixel points adjacent to each other in the gradient direction is: judging the positions of two pixel points adjacent to the pixel point according to the angle between the pixel point gradient direction and the horizontal direction.

[0144] Assume that there is a pixel point (i, j) on the spontaneous combustion area image of the goaf, and the gradient direction is θ.

[0145] If the angle between θ and the horizontal direction is If the gradient direction is close to horizontal, check the pixels on the left and right sides: (i, j-1) and (i, j+1);

[0146] If the angle between θ and the horizontal direction is If the pixel is between (i-1, j) and (i+1, j), check the pixels above and below: (i-1, j) and (i+1, j).

[0147] If the angle between θ and the horizontal direction is If the pixels are between , check the pixels on both sides of the diagonal: (i-1, j-1) and (i+1, j+1) or (i-1, j+1) and (i+1, j-1).

[0148] The classification method for dividing the spontaneous combustion area image of the goaf after non-maximum suppression into three categories: strong edge pixels, weak edge pixels and non-edge pixels is:

[0149] Set a high threshold and a low threshold, and compare the gradient amplitude of the pixel points in the spontaneous combustion area image of the goaf with the size of the two thresholds:

[0150] If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf exceeds the high threshold, it is set as a strong edge pixel point;

[0151] If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf is between the high threshold and the low threshold, it is set as a weak edge pixel point;

[0152] If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf is lower than the low threshold, it is set as a non-edge pixel point.

[0153] The high threshold and the low threshold are set according to actual needs;

[0154] The method of connecting strong edge pixels and their adjacent weak edge pixels to form continuous edge lines is:

[0155] Step S210: Finding pixels marked as strong edge pixels according to the classification results of strong edge pixels, weak edge pixels and non-edge pixels;

[0156] Step S220: for each strong edge pixel, check its eight neighboring pixels to find the neighboring pixels marked as weak edge pixels;

[0157] Step S230: performing edge tracing for each strong edge pixel and its adjacent weak edge pixel, connecting them to form a continuous edge line; preferably, edge tracing is performed in a recursive manner;

[0158] Step S240: During edge tracing, the connected pixels are marked as connected to avoid repeated tracing;

[0159] Step S250: Repeat steps S220 to S240 until all strong edge pixels are processed.

[0160] Furthermore, the method of selecting a recursive method to perform edge tracking in step S230 is:

[0161] Step S231: Select a strong edge pixel point as a starting point;

[0162] Step S232: starting from the starting point, searching for adjacent weak edge pixels;

[0163] Step S233: for each adjacent weak edge pixel point, mark it as a connected edge, and recursively continue edge tracing with the weak edge pixel point as a new starting point;

[0164] Step S234: Repeat step S233 until there are no more weak edge pixels to track.

[0165] The method for obtaining the temperature grayscale value H of the high-temperature area pixel point is: according to the grayscale image converted from the spontaneous combustion area image of the goaf, the temperature grayscale value H of each high-temperature area pixel point within the edge of the high-temperature area on the grayscale image is obtained.

[0166] Step S3: Calculate the distance D between each pixel point in the non-high temperature area outside the edge of the high temperature area and the high temperature area;

[0167] The method for calculating the distance D from each pixel point in the non-high temperature area outside the edge of the high temperature area to the high temperature area is:

[0168] Calculate the distance from each non-high temperature area pixel outside the edge of the high temperature area to each edge pixel at the edge of the high temperature area, select the edge pixel at the edge of the high temperature area with the smallest distance, and use the distance between the edge pixel and the non-high temperature area pixel as the distance D from the non-high temperature area pixel to the high temperature area.

[0169] Step S4: Obtain wind force factors, and calculate wind force F based on the wind force factors;

[0170] The wind factors include: wind speed data s, wind direction data x, and air density data m;

[0171] The method for obtaining the wind force factor and calculating the wind force F by the wind force factor is:

[0172] Obtain real-time wind speed data s, wind direction data x, and air density data m;

[0173] The wind force is calculated using wind speed data s, wind direction data x, and air density data m. The formula is wind force F = 0.5 × m × s 2 ×cosx;

[0174] The wind direction data x refers to the angle x between the wind direction and the direction from the edge pixel point to the non-high temperature area pixel point;

[0175] The wind force F is a wind vector in the distance direction from the edge pixel point to the pixel point in the non-high temperature area.

[0176] Step S5: Obtain the spontaneous combustion propagation factor, and determine the probability P of spontaneous combustion propagation of each pixel point in the non-high temperature area in the future time through the spontaneous combustion propagation factor;

[0177] The factors for the spread of spontaneous combustion include: the distance D from the pixel point in the non-high temperature area to the high temperature area, the wind force F, the temperature data T, the air visibility data V, and the temperature gray value H;

[0178] The calculation formula for the probability P of spontaneous combustion spreading of each pixel point in the non-high temperature area in the future time is: Where ε and δ are small constants used to avoid the denominator being zero, k is a scaling factor, and the temperature grayscale value H used for calculation is the temperature grayscale value H of the pixel in the high temperature area that is closest to the pixel in the non-high temperature area.

[0179] Step S6: Divide the non-high temperature area into grids, and use the probability of spontaneous combustion propagation of each pixel point in the non-high temperature area to calculate the average probability of spontaneous combustion propagation in each grid.

[0180] The non-high temperature area is divided into grids, and the probability of spontaneous combustion propagation of each pixel point in the non-high temperature area is used to calculate the average probability of spontaneous combustion propagation in each grid. The method is:

[0181] Divide the non-high temperature area of ​​the spontaneous combustion area image of the goaf into grids according to the grid size;

[0182] Preferably, the non-high temperature area of ​​the spontaneous combustion area image of the goaf is divided into a plurality of grid grids according to the grid size a×a;

[0183] Map irregular non-high temperature areas to a grid;

[0184] Randomly select n non-high temperature area pixel points in a non-high temperature area of ​​a grid; preferably, the random method selects a random number generator;

[0185] Calculate the probability P of spontaneous combustion spreading for each pixel point in the selected non-high temperature area;

[0186] Calculate the average probability of spontaneous combustion spread of n non-high temperature area pixels in the future as the average probability of spontaneous combustion spread in the grid

[0187] Step S7: According to the average probability of spontaneous combustion spread in the grid Cluster the grids to obtain high-risk areas for spontaneous combustion spread;

[0188] The average probability of spontaneous combustion propagation in the grid The method to cluster the raster is:

[0189] The average probability of spontaneous combustion spread of each grid is taken as a feature vector to form a data set;

[0190] Use clustering algorithms to divide the rasters in the dataset into different clusters according to probability;

[0191] For all clusters, the average probability of spontaneous combustion spread is To classify;

[0192] According to the classification level, high-risk areas for spontaneous combustion spread are displayed to users in a visual way;

[0193] The classification method is as follows: the cluster with level A is set as a risk-free area; the cluster with level B is set as a low-risk area; the cluster with level C is set as a medium-risk area; the cluster with level D is set as a high-risk area; preferably, the risk-free area refers to the average probability The range is between 0% and 20%; the low-risk area refers to the average probability P range between 20% and 50%; the medium-risk area refers to the average probability The range is between 50% and 80%; the high-risk area refers to the average probability The range is between 80% and 100%;

[0194] The clustering algorithm adopts the k-means clustering algorithm, which is a conventional technical means in this field and will not be described in detail here.

[0195] Example 2

[0196] like Figure 2 As shown in the figure, a visualization diagram of clustering results of a method for spontaneous combustion monitoring and extinguishing in goaf based on multi-parameter fusion is shown. The overall image represents the grayscale image of the spontaneous combustion area image of the goaf. The middle gray area 1, gray area 2, and gray area 3 represent the grayscale images of the high-temperature area. The spontaneous combustion is weakened from gray area 1 to gray area 3. The irregular non-high-temperature area is mapped to a number of grid grids with a grid size of a×a. The grids are divided into different clusters according to probability using a clustering algorithm. There are four types of clusters, namely, level A, level B, level C, and level D. Each cluster is composed of a number of grids, and the grids in each cluster have similar probabilities of spontaneous combustion spreading.

[0197] Among them, the cluster with level A is a risk-free area, and the probability of spontaneous combustion spreading is between 0 and 20%;

[0198] Among them, the cluster with level B is a low-risk area, and the probability of spontaneous combustion spreading is between 20% and 50%;

[0199] Among them, the cluster with level C is a medium-risk area, and the probability of spontaneous combustion spreading is between 50% and 80%;

[0200] Among them, the cluster with level D is a high-risk area, and the probability of spontaneous combustion spreading is between 80% and 100%.

[0201] Example 3

[0202] like Figure 3 A goaf spontaneous combustion monitoring and fire extinguishing device based on multi-parameter fusion is shown, comprising:

[0203] A spontaneous combustion image acquisition module, used to acquire the spontaneous combustion area image of the goaf;

[0204] The edge detection module is used to obtain edge pixel points at the edge of the high-temperature area using the image of the spontaneous combustion area of ​​the goaf and the edge detection algorithm, and obtain the temperature gray value H of the pixel points in the high-temperature area;

[0205] A distance calculation module is used to calculate the distance D between each pixel point in the non-high temperature area outside the edge of the high temperature area and the high temperature area;

[0206] A wind force calculation module is used to obtain wind force factors and calculate the wind force F based on the wind force factors;

[0207] The spread probability calculation module is used to obtain the spontaneous combustion spread factor and determine the probability P of spontaneous combustion spread of each pixel point in the non-high temperature area in the future through the spontaneous combustion spread factor;

[0208] The average probability calculation module is used to divide the non-high temperature area into grids, and use the probability of spontaneous combustion spread of each pixel in the non-high temperature area to calculate the average probability of spontaneous combustion spread in each grid.

[0209] High-risk area clustering module, used to cluster the areas according to the average probability of spontaneous combustion spread in the grid Cluster the grids to obtain high-risk areas for spontaneous combustion spread.

[0210] For example, the spontaneous combustion image acquisition module collects historical high-temperature area images and historical normal temperature images, and uses the historical high-temperature area images and historical normal temperature images to train a neural network model for judging the category of spontaneous combustion fire. The monitoring equipment collects goaf images, and uses the trained neural network model to judge whether there is spontaneous combustion in the goaf images. If there is spontaneous combustion, a fire warning is triggered. The drone equipment responds to the fire warning, enters the air above the goaf to take images of the spontaneous combustion area of ​​the goaf, and transmits the images of the spontaneous combustion area of ​​the goaf to the edge detection module.

[0211] The edge detection module uses the spontaneous combustion area image of the goaf and the edge detection algorithm to obtain the edge pixels of the high-temperature area and the temperature grayscale value H of the pixels in the high-temperature area, and transmits the spontaneous combustion area image of the goaf with edge lines after edge detection to the distance calculation module, and transmits the temperature grayscale value H to the spread probability calculation module.

[0212] The distance calculation module uses the spontaneous combustion area image of the goaf with edge lines to calculate the distance D from each pixel point in the non-high temperature area outside the edge of the high temperature area to the high temperature area, and transmits the calculated distance D from the pixel point in the non-high temperature area to the high temperature area to the spread probability calculation module.

[0213] The wind force calculation module is used to obtain wind force factors, calculate the wind force F according to the wind force factors, and transmit the calculated wind force F to the spreading probability calculation module.

[0214] The spread probability calculation module is used to obtain the spontaneous combustion spread factor, determine the probability P of spontaneous combustion spread of each pixel point in the non-high temperature area in the future time through the spontaneous combustion spread factor, and transmit the probability P of spontaneous combustion spread of each pixel point in the non-high temperature area to the average probability calculation module.

[0215] The average probability calculation module divides the non-high temperature area into grids, and uses the probability of spontaneous combustion spread of each pixel in the non-high temperature area to calculate the average probability of spontaneous combustion spread in each grid. The average probability of spontaneous combustion spreading in each grid Transmitted to the high-risk area clustering module.

[0216] The high-risk area clustering module is used to cluster the high-risk areas according to the average probability of spontaneous combustion spread in the grid. Cluster the grids and divide them into different clusters according to probability. The system divides the fire hazard levels of the non-high-temperature areas of the spontaneous combustion area of ​​the goaf into different levels and displays them to the user in a visual manner. The user can intuitively see the high-risk areas where spontaneous combustion may spread based on the visualization results, which helps the user to intuitively find the dangerous areas and improve the accuracy of manual fire extinguishing.

[0217] Example 4

[0218] Figure 4 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 4 As shown, according to another aspect of the present application, an electronic device is also provided. The electronic device may include one or more processors and one or more memories. The memory stores a computer readable code, and when the computer readable code is run by one or more processors, it can execute the above-mentioned method for goaf spontaneous combustion monitoring and fire extinguishing based on multi-parameter fusion.

[0219] The method or system according to the embodiment of the present application can also be used by Figure 4 The electronic device architecture shown in FIG. Figure 4As shown, the electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, etc. The storage device in the electronic device, such as a ROM or a hard disk, can store a method for monitoring and extinguishing spontaneous combustion in goafs based on multi-parameter fusion provided in the present application. A method for monitoring and extinguishing spontaneous combustion in goafs based on multi-parameter fusion may, for example, include: obtaining an image of the spontaneous combustion area of ​​the goaf; using the image of the spontaneous combustion area of ​​the goaf, using an edge detection algorithm to obtain edge pixels at the edge of the high-temperature area, and obtaining the temperature grayscale value H of the pixel points in the high-temperature area; calculating the distance D from each non-high-temperature area pixel point outside the edge of the high-temperature area to the high-temperature area; obtaining wind factors, and calculating the wind force F through the wind factors; obtaining spontaneous combustion propagation factors, and determining the probability P of spontaneous combustion propagation of each non-high-temperature area pixel point in the future time through the spontaneous combustion propagation factors; dividing the non-high-temperature area into grids, and using the probability of spontaneous combustion propagation of each non-high-temperature area pixel point to calculate the average probability of spontaneous combustion propagation in each grid. According to the average probability of spontaneous combustion spread in the grid The grids are clustered to obtain high-risk areas for spontaneous combustion. Furthermore, the electronic device may also include a user interface. Of course, Figure 4 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 4 One or more components of an electronic device are shown.

[0220] Example 5

[0221] Figure 5 Schematic diagram of a readable storage medium structure provided by an embodiment of the present application. Figure 5 As shown, it is a readable storage medium according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor, a method for spontaneous combustion monitoring and fire extinguishing in goaf based on multi-parameter fusion according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0222] In addition, according to the implementation mode of the present application, the process described in the above reference flowchart can be implemented as a computer software program. For example, the present application provides a non-temporary machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, such as: obtaining an image of the spontaneous combustion area of ​​the goaf; using the image of the spontaneous combustion area of ​​the goaf, using an edge detection algorithm to obtain edge pixels at the edge of the high-temperature area, and obtaining the temperature grayscale value H of the pixel points in the high-temperature area; calculating the distance D from each non-high-temperature area pixel point outside the edge of the high-temperature area to the high-temperature area; obtaining wind factors, and calculating wind force F through wind factors; obtaining spontaneous combustion propagation factors, and determining the probability P of spontaneous combustion propagation of each non-high-temperature area pixel point in the future time through spontaneous combustion propagation factors; dividing the non-high-temperature area into grids, and using the probability of spontaneous combustion propagation of each non-high-temperature area pixel point to calculate the average probability of spontaneous combustion propagation in each grid. According to the average probability of spontaneous combustion spread in the grid The grids are clustered to obtain high-risk areas for spontaneous combustion. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0223] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0224] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0225] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for spontaneous combustion monitoring and fire extinguishing in goaf based on multi-parameter fusion, characterized in that: include: Acquire the spontaneous combustion area image of the goaf; Using the spontaneous combustion area image of the goaf, the edge detection algorithm is used to obtain the edge pixel points of the edge of the high-temperature area, and the temperature gray value H of the pixel points in the high-temperature area is obtained; Calculate the distance D from each pixel point in the non-high temperature area outside the edge of the high temperature area to the high temperature area; Obtain wind force factors, and calculate wind force F based on the wind force factors; Obtain the spontaneous combustion spread factor, and determine the probability P of spontaneous combustion spread of each pixel point in the non-high temperature area in the future time through the spontaneous combustion spread factor; The non-high temperature area is divided into grids, and the probability of spontaneous combustion spread of each pixel in the non-high temperature area is used to calculate the average probability of spontaneous combustion spread in each grid. According to the average probability of spontaneous combustion spread in the grid Cluster the grids to obtain high-risk areas for spontaneous combustion spread; The wind factors include: wind speed data s, wind direction data x, and air density data m; The method for obtaining the wind force factor and calculating the wind force F by the wind force factor is: Obtain real-time wind speed data s, wind direction data x, and air density data m; The wind force is calculated using wind speed data s, wind direction data x, and air density data m. The formula is wind force F = 0.5 × m × s 2 ×cosx; The spontaneous combustion propagation factor is obtained, and the calculation formula for determining the probability P of spontaneous combustion propagation of each pixel point in the non-high temperature area in the future time is determined by the spontaneous combustion propagation factor: Wherein ε and δ are small constants used to avoid the denominator being zero, k is a scaling factor, the temperature grayscale value H used in the calculation is the temperature grayscale value H of the pixel point in the high temperature area closest to the pixel point in the non-high temperature area, and the spontaneous combustion spread factors include: the distance D from the pixel point in the non-high temperature area to the high temperature area, the wind force F, the temperature data T, the air visibility data V, and the temperature grayscale value H; The non-high temperature area is divided into grids, and the probability of spontaneous combustion propagation of each pixel point in the non-high temperature area is used to calculate the average probability of spontaneous combustion propagation in each grid. The method is: Divide the non-high temperature area of ​​the spontaneous combustion area image of the goaf into grids according to the grid size; Map irregular non-high temperature areas to a grid; Randomly select n non-high temperature area pixels in a grid's non-high temperature area; the random method uses a random number generator; Calculate the probability P of spontaneous combustion spreading for each pixel point in the selected non-high temperature area; The average probability of spontaneous combustion propagation of n pixels in the non-high temperature area in the future is calculated as the average probability P of spontaneous combustion propagation in the grid.

2. The method for spontaneous combustion monitoring and extinguishing in goaf based on multi-parameter fusion according to claim 1 is characterized in that: The specific steps of using the edge detection algorithm to obtain edge pixel points at the edge of the high temperature area and obtaining the temperature grayscale value H of the pixel points in the high temperature area are: Convert the spontaneous combustion area image of the goaf into a grayscale image; Perform smoothing filtering on the grayscale image, use Gaussian filtering to remove noise, and obtain a smoothed image; Based on the smoothed image, the Sobel gradient operator is used to calculate the gradient magnitude and gradient direction; The pixels of the spontaneous combustion area image of the goaf are subjected to non-maximum suppression along the gradient direction, the pixels with the largest gradient amplitude are retained, and the pixels in other directions are suppressed; The spontaneous combustion area image of the goaf after non-maximum suppression is divided into three categories: strong edge pixels, weak edge pixels and non-edge pixels; Connect strong edge pixels and their adjacent weak edge pixels to form continuous edge lines; Output an image containing a high temperature area, a non-high temperature area, and edge lines after edge detection processing; wherein edge pixel points of edge lines of the spontaneous combustion area image of the goaf after edge detection processing are highlighted; The temperature grayscale value H of the pixel point in the high temperature area is obtained by using the edge lines and grayscale image.

3. The method for spontaneous combustion monitoring and fire extinguishing in goaf based on multi-parameter fusion according to claim 2 is characterized in that: The classification method for dividing the spontaneous combustion area image of the goaf after non-maximum suppression into three categories: strong edge pixels, weak edge pixels and non-edge pixels is: Set a high threshold and a low threshold, and compare the gradient amplitude of the pixel points in the spontaneous combustion area image of the goaf with the size of the two thresholds: If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf exceeds the high threshold, it is set as a strong edge pixel point; If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf is between the high threshold and the low threshold, it is set as a weak edge pixel point; If the gradient amplitude of the pixel point of the spontaneous combustion area image of the goaf is lower than the low threshold, it is set as a non-edge pixel point.

4. The method for spontaneous combustion monitoring and extinguishing in goaf based on multi-parameter fusion according to claim 3 is characterized in that: The method of connecting strong edge pixels and their adjacent weak edge pixels to form continuous edge lines is: Step S210: Finding pixels marked as strong edge pixels according to the classification results of strong edge pixels, weak edge pixels and non-edge pixels; Step S220: for each strong edge pixel, check its eight neighboring pixels to find the neighboring pixels marked as weak edge pixels; Step S230: performing edge tracing on each strong edge pixel and its adjacent weak edge pixel, and connecting them to form a continuous edge line; Step S240: During edge tracing, the connected pixels are marked as connected to avoid repeated tracing; Step S250: Repeat steps S220 to S240 until all strong edge pixels are processed.

5. The method for spontaneous combustion monitoring and extinguishing in goaf based on multi-parameter fusion according to claim 4 is characterized in that: The method for clustering grids according to the average probability of spontaneous combustion propagation in the grids is: The average probability of spontaneous combustion spread of each grid is taken as a feature vector to form a data set; Use clustering algorithms to divide the rasters in the dataset into different clusters according to probability; All clusters are graded according to the average probability of spontaneous combustion spread; According to the classification levels, high-risk areas for spontaneous combustion spread are displayed to users in a visual manner.

6. A goaf spontaneous combustion monitoring and fire extinguishing device based on multi-parameter fusion, characterized in that: include: A spontaneous combustion image acquisition module, used to acquire the spontaneous combustion area image of the goaf; The edge detection module is used to obtain edge pixel points at the edge of the high-temperature area using the image of the spontaneous combustion area of ​​the goaf and the edge detection algorithm, and obtain the temperature gray value H of the pixel points in the high-temperature area; A distance calculation module is used to calculate the distance D between each pixel point in the non-high temperature area outside the edge of the high temperature area and the high temperature area; A wind force calculation module is used to obtain wind force factors and calculate the wind force F based on the wind force factors; The spread probability calculation module is used to obtain the spontaneous combustion spread factor and determine the probability P of spontaneous combustion spread of each pixel point in the non-high temperature area in the future through the spontaneous combustion spread factor; The average probability calculation module is used to divide the non-high temperature area into grids, and use the probability of spontaneous combustion spread of each pixel point in the non-high temperature area to calculate the average probability of spontaneous combustion spread in each grid; The high-risk area clustering module is used to cluster the grids according to the average probability of spontaneous combustion spread in the grids to obtain the high-risk areas for spontaneous combustion spread; The wind factors include: wind speed data s, wind direction data x, and air density data m; The method for obtaining the wind force factor and calculating the wind force F by the wind force factor is: Obtain real-time wind speed data s, wind direction data x, and air density data m; The wind force is calculated using wind speed data s, wind direction data x, and air density data m. The formula is wind force F = 0.5 × m × s 2 ×cosx; The spontaneous combustion propagation factor is obtained, and the calculation formula for determining the probability P of spontaneous combustion propagation of each pixel point in the non-high temperature area in the future time is determined by the spontaneous combustion propagation factor: Wherein ε and δ are small constants used to avoid the denominator being zero, k is a scaling factor, the temperature grayscale value H used in the calculation is the temperature grayscale value H of the pixel point in the high temperature area closest to the pixel point in the non-high temperature area, and the spontaneous combustion spread factors include: the distance D from the pixel point in the non-high temperature area to the high temperature area, the wind force F, the temperature data T, the air visibility data V, and the temperature grayscale value H; The non-high temperature area is divided into grids, and the probability of spontaneous combustion propagation of each pixel point in the non-high temperature area is used to calculate the average probability of spontaneous combustion propagation in each grid. The method is: Divide the non-high temperature area of ​​the spontaneous combustion area image of the goaf into grids according to the grid size; Map irregular non-high temperature areas to a grid; Randomly select n non-high temperature area pixels in a grid's non-high temperature area; the random method uses a random number generator; Calculate the probability P of spontaneous combustion spreading for each pixel point in the selected non-high temperature area; Calculate the average probability of spontaneous combustion spread of n non-high temperature area pixels in the future as the average probability of spontaneous combustion spread in the grid 7. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in a method for spontaneous combustion monitoring and extinguishing of goaf based on multi-parameter fusion as described in any one of claims 1 to 5.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute the steps in a method for spontaneous combustion monitoring and extinguishing in goaf based on multi-parameter fusion as described in any one of claims 1 to 5.

Images