An intelligent early warning system and method for coal mine safety accidents

Abstract

The present invention relates to the technical field of coal mine safety early warning, and specifically to an intelligent early warning system and method for coal mine safety accidents. The method comprises: acquiring underground monitoring images of coal mines; extracting dust points belonging to high-frequency features; grayscale processing and division to obtain image areas, and determining the grayscale discrete index of each image area based on the numerical distribution discreteness of the grayscale values ​​of all pixels in each image area; analyzing the dust quality, and determining the quality correction index based on the grayscale discrete index of the image area at any height and below it; correcting the grayscale discrete index to obtain a corrected discrete value; combining the number of dust points in each image area and the corrected discrete value to determine the possibility index, and making early warning judgments based on the numerical values ​​of the possibility index of adjacent acquisition time frames. The present invention can improve the reliability of the overall intelligent hidden danger investigation.

Description

Technical Field

[0001] The present invention relates to the technical field of coal mine safety early warning technology, and in particular to an intelligent early warning system and method for coal mine safety accidents. Background Art

[0002] During the mining process inside a coal mine, mechanical coal crushing and transportation are accompanied by the generation of dust, which floats in the air. Dust is a major hazard in mines, and dust can easily cause respiratory diseases. When dust control management is inadequate and the dust concentration in the space reaches a certain level, if the equipment ages or malfunctions, sparks can easily occur, causing casualties and property damage. Existing technologies analyze the grayscale difference between the captured image and the standard image to analyze the impact of dust and provide early warnings. However, due to the large space and complex ventilation conditions, the air flow is unstable, and the dust concentration varies at different heights and locations. This makes it difficult for a single monitoring point to accurately represent the entire dust situation, and dust analysis based solely on image grayscale changes is less reliable. Therefore, effective dust identification and intelligent hidden danger detection are needed within mines. Summary of the Invention

[0003] To address the technical problem in related technologies of poor reliability in detecting dust hazards based solely on image grayscale changes, due to the different impacts of dust at different heights and inconsistent concentrations at different locations, the present invention provides an intelligent early warning system and method for coal mine safety accidents. The technical solutions employed are as follows:

[0004] The present invention proposes an intelligent early warning system and method for coal mine safety accidents, the method comprising:

[0005] Obtain underground coal mine monitoring images at each acquisition time frame; obtain the overall high-frequency and low-frequency information of the underground coal mine monitoring images based on Gaussian filtering, and extract dust points with high-frequency characteristics;

[0006] Each coal mine underground monitoring image is grayscale processed and evenly divided to obtain different image areas. The grayscale discrete index of each image area is determined based on the numerical distribution discreteness of the grayscale values ​​of all pixels in each image area.

[0007] Analyze the dust quality and determine the quality correction index of each image area based on the grayscale discrete index of the image area at any height and below it; correct the grayscale discrete index based on the quality correction index to obtain a corrected discrete value;

[0008] Combined with the number of dust points and the corrected discrete value in each image area, the possibility index of dust hazard occurring in the corresponding acquisition time frame is determined, and early warning judgment is made based on the values ​​of the possibility index in adjacent acquisition time frames.

[0009] Furthermore, the method of obtaining the overall high-frequency information and low-frequency information of the underground coal mine monitoring image based on Gaussian filtering and extracting dust points with high-frequency features includes:

[0010] The coal mine underground monitoring image is smoothed based on Gaussian filtering and used as a low-frequency image;

[0011] Calculate the brightness ratio of the coal mine underground monitoring image and the low-frequency image at the same point as the high-frequency feature index at the corresponding point;

[0012] The dust point is determined according to the numerical value of the high-frequency characteristic index.

[0013] Furthermore, determining the dust point according to the numerical value of the high-frequency characteristic index includes:

[0014] The points where the high-frequency characteristic index is greater than a preset index threshold are regarded as dust points under high-frequency characteristics.

[0015] Furthermore, the grayscale dispersion index of each image region is determined based on the numerical distribution discreteness of the grayscale values ​​of all pixels in each image region, including:

[0016] Calculate the standard deviation of the grayscale values ​​of all pixels in the same image area as the grayscale dispersion index.

[0017] Furthermore, determining the quality correction index of each image area based on the grayscale discrete index of any height and the image area below it includes:

[0018] The height of any point is taken as the target height, and the heights of all points below the target height are taken as candidate heights;

[0019] Calculate the sum of the grayscale discrete indexes at the target height and all candidate heights, and normalize them as the quality correction index.

[0020] Furthermore, the grayscale discrete index is corrected in combination with the quality correction index to obtain a corrected discrete value, including:

[0021] The product value of the quality correction index and the grayscale discrete index of the image area is calculated as the correction discrete value.

[0022] Furthermore, the method of combining the number of dust points and the corrected discrete value in each image area to determine the probability index of dust hazard occurring in the corresponding acquisition time frame includes:

[0023] The product value of the number of dust points in the image area and the corrected discrete value is calculated and normalized to be used as a possibility indicator.

[0024] Furthermore, the making of early warning judgment according to the values ​​of the possibility indicators of adjacent acquisition time frames includes:

[0025] Determine the image area whose likelihood index of all image areas in the same acquisition frame is greater than a preset likelihood threshold as the warning area;

[0026] Whether to issue a warning is determined based on the change in area of ​​the warning area between the current and previous acquisition time frames, and the average value of the possibility indicators of all warning areas in the current acquisition time frame.

[0027] Furthermore, the determining whether to issue a warning based on the change in area of ​​the warning area between the current and previous acquisition time frames and the average value of the possibility index of all warning areas in the current acquisition time frame includes:

[0028] Calculate the difference in the warning area between the current acquisition time frame and the previous acquisition time frame, and normalize it as the area impact indicator;

[0029] The product of the average value of all warning area possibility indicators and the area impact indicator is normalized and used as the warning judgment value;

[0030] When the warning judgment value is greater than the preset warning threshold, it is determined to perform warning processing; otherwise, it is determined not to perform warning processing.

[0031] On the other hand, an intelligent early warning system for coal mine safety accidents is also provided, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any of the methods described above are implemented.

[0032] The present invention has the following beneficial effects:

[0033] The embodiment of the present invention obtains the underground monitoring image of the coal mine on each acquisition time frame; and according to the high-frequency characteristics of the dust, performs high-frequency information and low-frequency information analysis to obtain dust points; then, combines the distribution discrete indicators of dust in different blocks, as well as the quality characteristics and lifting characteristics of the dust itself, to perform dust distribution analysis to obtain a corrected discrete value. The corrected discrete value can effectively integrate the quality characteristics and dust lifting characteristics, thereby improving the reliability of dust aggregation analysis; then, uses the number of dust points as a density analysis feature, combined with the corrected discrete value, to determine the possibility index of dust hazards, so that the analysis of the possibility index can effectively combine multi-dimensional data features, avoid inaccurate effects caused by grayscale feature analysis alone, and perform early warning judgments through the numerical changes of the possibility indicators of adjacent acquisition time frames, combined with the change characteristics of adjacent acquisition time frames, so that the early warning judgment is combined with dynamic change analysis, thereby improving the reliability of the overall intelligent hidden danger detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 A flow chart of an intelligent early warning method for coal mine safety accidents provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0036] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of an intelligent early warning system and method for coal mine safety accidents proposed in accordance with the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0038] The following describes in detail a specific solution of an intelligent early warning method for coal mine safety accidents provided by the present invention in conjunction with the accompanying drawings.

[0039] See also Figure 1, which shows a flow chart of an intelligent early warning method for coal mine safety accidents provided by one embodiment of the present invention, the method comprising:

[0040] S101: Acquire an underground coal mine monitoring image at each acquisition time frame; acquire high-frequency information and low-frequency information of the entire underground coal mine monitoring image based on Gaussian filtering, and extract dust points with high-frequency characteristics.

[0041] Mining operations often generate dust, a significant risk factor for dust inside mines. Dust can easily lead to respiratory illnesses. When dust control is inadequate and dust concentrations reach a certain level, sparks from aging or malfunctioning equipment can easily cause flash explosions, resulting in casualties and property damage. Therefore, effective dust identification and intelligent risk detection within mines are essential.

[0042] In related technologies, the grayscale difference between the captured image and the standard image is analyzed to analyze the dust impact and issue an early warning. In this method, since the concentration of dust is inconsistent at different heights and locations, and the reliability of dust analysis based only on image grayscale changes is poor, further optimization analysis is needed.

[0043] In an embodiment of the present invention, a high-definition camera is installed in a mine to capture images at different time frames. Specifically, underground coal mine monitoring images can be acquired at a rate of one frame per second, or one frame every 10 seconds, for periodic analysis. Each time frame of the acquired underground coal mine monitoring images can be referred to as an acquisition time frame. To improve comparison accuracy, it is necessary to ensure that the position of the camera and the elements in the object being photographed do not change throughout the entire capture process, and the brightness of the captured image must also remain consistent.

[0044] In the embodiment of the present invention, since the dust distribution in the mine has a strong regularity, specifically in that longer and narrow tunnels may cause dust to accumulate in local areas, resulting in increased dust concentration, while large-space mines help to dilute the dust distribution; smaller dust particles will remain suspended in the air for a longer time, while larger particles are more likely to settle and usually concentrate in places closer to the source; by analyzing the number of pixels corresponding to dust in certain local areas of the entire image and the relationship between pixels, the specific situation of dust distribution in the area can be obtained.

[0045] To achieve the desired effect, it's necessary to first separate the dust from the mine, essentially identifying the specific locations of the dust, referred to as dust locations in this embodiment. Because dust appears as tiny particles in an image and exhibits high-frequency characteristics, this embodiment determines dust locations through high-frequency feature analysis.

[0046] Furthermore, in some embodiments of the present invention, the overall high-frequency information and low-frequency information of the underground monitoring image of the coal mine are obtained based on Gaussian filtering, and the dust points belonging to the high-frequency characteristics are extracted, including: smoothing the underground monitoring image of the coal mine based on Gaussian filtering as a low-frequency image; calculating the brightness value ratio of the underground monitoring image of the coal mine and the low-frequency image at the same point as the high-frequency feature index at the corresponding point; and determining the dust point according to the numerical value of the high-frequency feature index.

[0047] Because Gaussian filtering blurs the image through smoothing in the spatial domain, reducing rapidly changing details and preserving the general structural characteristics of the image, it is used to obtain low-frequency information in the image. The coal mine monitoring image is smoothed using Gaussian filtering to produce a low-frequency image. Since dust characteristics are high-frequency information, in embodiments of the present invention, the brightness ratio of the coal mine monitoring image and the low-frequency image at the same point can be directly calculated as the high-frequency feature indicator at the corresponding point.

[0048] Since the larger the value of the high-frequency feature index is, the greater the difference between the normal brightness and the low-frequency information brightness at the corresponding point is, which is also manifested as the more obvious the high-frequency grayscale feature is, and the more effective it is in distinguishing white noise points in the image, the greater the possibility that it is a dust point. Therefore, in this embodiment of the present invention, the point whose high-frequency feature index is greater than the preset index threshold is regarded as the dust point under the high-frequency feature.

[0049] Among them, the preset indicator threshold is the threshold value of the high-frequency feature indicator. In an embodiment of the present invention, the preset indicator threshold can be specifically set to 0.8, that is, when the high-frequency feature index is greater than 0.8, the corresponding point is used as a dust point, and when the high-frequency feature index is less than or equal to 0.8, the corresponding point is used as a background point.

[0050] S102: grayscale processing is performed on each coal mine underground monitoring image and evenly divided to obtain different image areas. The grayscale discrete index of each image area is determined based on the numerical distribution discreteness of the grayscale values ​​of all pixels in each image area.

[0051] Dust inside mines is typically generated by tiny particles from ore crushing and equipment operation. These particles are typically very small and appear as floating and dispersed particles in images. Consequently, the distribution of dust inside mines is irregular, appearing as fuzzy dots or smoke. Furthermore, dust distribution within different areas is irregular. Therefore, detailed dust analysis can be performed based on discrete dust distribution indicators.

[0052] Since the distribution position and state of dust of different diameters in the mine are different, in order to analyze different blocks, the coal mine underground monitoring image can be divided first. In an embodiment of the present invention, the coal mine underground monitoring image can be specifically divided into different image areas. In an embodiment of the present invention, the coal mine underground monitoring image can be directly divided into a fixed number of image areas, or the size of the image area can be preset and divided according to the size. For example, the coal mine underground monitoring image is segmented according to a pixel size of 25×25. It should be noted that when any subsequent column or row is less than 25 pixels, it is necessary to perform a pixel point supplement with a grayscale value of 255. This is because this embodiment is aimed at coal mine safety analysis, and the grayscale value of coal is close to the grayscale value of 0, so the pixel point supplement with a grayscale value of 255 is selected to facilitate subsequent analysis.

[0053] Furthermore, in some embodiments of the present invention, the grayscale discrete index of each image area is determined based on the numerical distribution discreteness of the grayscale values ​​of all pixels in each image area, including: calculating the standard deviation of the grayscale values ​​of all pixels in the same image area as the grayscale discrete index.

[0054] In the underground monitoring images of coal mines, due to the different diameters of dust particles, the sedimentation degree of dust of different diameters is also different, which means that the dust aggregation information at different positions in the same image is also different. The details and features of the local areas of the underground monitoring images of coal mines can reveal the local dust aggregation situation. In the embodiment of the present invention, the standard deviation is used for discrete analysis. The larger the value of the standard deviation, the poorer the aggregation in the same image area. That is, the larger the value of the grayscale discrete index, the worse the dust aggregation effect in the image area.

[0055] S103: Analyze the dust quality and determine a quality correction index for each image area based on the grayscale discrete index of the image area at any height and below it; and correct the grayscale discrete index in combination with the quality correction index to obtain a corrected discrete value.

[0056] It should be noted that the grayscale discrete index of each area calculated using the above logic is obtained without considering the dust mass. In reality, dust of different masses is distributed differently in space. In the same scene, the heavier the dust, the closer it is to the bottom, while the lighter the dust, the more likely it is to float in the air. Therefore, the aggregation information obtained above (i.e., the grayscale discrete index) needs to be corrected.

[0057] Furthermore, in some embodiments of the present invention, the quality correction index of each image area is determined based on the grayscale discrete index of any height and the image area below it, including: taking the height of any point as the target height and the heights of all points below the target height as the selected heights; calculating the target height and the sum of the grayscale discrete indexes at all the selected heights, and normalizing it as the quality correction index.

[0058] It should be noted that in actual scenarios, dust particles of different mass sizes have different settling velocities. The larger the mass, the faster it sinks. Therefore, the image shows that the dust in the area closer to the bottom of the image has better aggregation. As time changes, the possibility of dust near the bottom area adhering to the ground becomes higher. Therefore, the proportion of disasters caused by dust in this area is far less than that of dust above the image, that is, the probability of dust disasters below and above will change.

[0059] However, in actual scenarios, dust from below may be lifted up due to various factors such as wind, human activities, and machine operation. Taking this aspect into consideration, in an embodiment of the present invention, the grayscale discrete indicators of the target height and all the selected heights below it are directly accumulated. When close to the bottom, due to the small number of accumulated parameters, the proportion of disasters caused by the dust below is not as good as that of the dust above. When the dust distribution below is more discrete and easier to be lifted up, the accumulated grayscale discrete indicator value is larger, which also represents the lifting feature. Therefore, the quality correction indicator of the embodiment of the present invention can accurately characterize the quality feature and the lifting feature, and has higher reliability.

[0060] The grayscale discrete index is corrected in combination with the quality correction index to obtain a corrected discrete value, including: calculating a product value of the quality correction index and the grayscale discrete index of the image area as the corrected discrete value.

[0061] In the embodiment of the present invention, the grayscale discrete index is further corrected by the quality correction index, so that the corrected discrete value can integrate the quality characteristics and the dust raising characteristics, thereby improving the reliability of the dust aggregation analysis.

[0062] S104: Determine the probability index of dust hazard occurring in the corresponding acquisition time frame based on the number of dust points and the corrected discrete value in each image area, and make an early warning judgment based on the values ​​of the probability index of adjacent acquisition time frames.

[0063] Among them, since dust has a diffusion effect, the relevant technology only analyzes the dust situation at a single moment and does not take its changes into consideration. Therefore, in the embodiment of the present invention, a specific analysis can be performed in combination with the changes in different acquisition time frames. First, it is necessary to analyze and obtain the possibility index.

[0064] Furthermore, in some embodiments of the present invention, the number of dust points and the corrected discrete value in each image area are combined to determine the possibility index of dust hazard occurring in the corresponding acquisition time frame, including: calculating the product value of the number of dust points and the corrected discrete value in the image area, and normalizing it as the possibility index.

[0065] Among them, since the number of dust points objectively represents the dust concentration, the more dust points there are in an area of ​​the same area, the greater the dust concentration, and the corrected discrete value represents the state of the dust. The larger the value of the corrected discrete value, the more likely the dust distribution state is to cause explosion hazards. Therefore, the product value of the number of dust points in the image area and the corrected discrete value is directly calculated, and the normalized processing is performed to obtain the possibility index of dust hazard.

[0066] Furthermore, in some embodiments of the present invention, a warning judgment is made based on the numerical values ​​of the possibility indicators of adjacent acquisition time frames, including: determining the image area in which the possibility indicators of all image areas in the same acquisition frame are greater than a preset possibility threshold as a warning area; and determining whether to issue a warning based on the change in the area of ​​the warning area between the current and previous acquisition time frames, and the average value of the possibility indicators of all warning areas in the current acquisition time frame.

[0067] The preset possibility threshold is a threshold value of the possibility index. In the embodiment of the present invention, the preset possibility threshold can be, for example, 0.75. That is, when the possibility index is greater than 0.75, the corresponding image area is designated as a warning area. The warning area represents an area with a relatively dangerous dust condition.

[0068] Determine whether to issue a warning based on the change in the area of ​​the warning area between the current and previous acquisition time frames, and the average of all warning area possibility indicators within the current acquisition time frame, including: calculating the difference in the warning area area between the current acquisition time frame and the previous acquisition time frame, and normalizing it as the area impact indicator; normalizing the product of the average of all warning area possibility indicators and the area impact indicator as the warning judgment value; when the warning judgment value is greater than the preset warning threshold, determine to issue a warning; otherwise, determine not to issue a warning.

[0069] In this embodiment of the present invention, while directly analyzing the mean value of the warning area likelihood index, we also incorporate the area change characteristic. A decreasing area indicates a decrease in dust concentration, while an increasing area indicates an increase in dust concentration, requiring special attention. Therefore, the product of the mean value of all warning area likelihood indices and the area impact index is used as the warning judgment value.

[0070] Among them, the preset warning threshold is the threshold value of the warning judgment value, which can be 0.8, for example. It is analyzed according to the actual scenario and there is no restriction on this. When the warning judgment value is greater than 0.8, it is determined to perform warning processing, otherwise, it is determined not to perform warning processing. The warning processing in the embodiment of the present invention can be specifically, for example, air humidification to allow dust to settle, a buzzer to remind relevant inspection personnel, etc., and there is no restriction on this.

[0071] The embodiment of the present invention obtains the underground monitoring image of the coal mine on each acquisition time frame; and according to the high-frequency characteristics of the dust, performs high-frequency information and low-frequency information analysis to obtain dust points; then, combines the distribution discrete indicators of dust in different blocks, as well as the quality characteristics and lifting characteristics of the dust itself, to perform dust distribution analysis to obtain a corrected discrete value. The corrected discrete value can effectively integrate the quality characteristics and dust lifting characteristics, thereby improving the reliability of dust aggregation analysis; then, uses the number of dust points as a density analysis feature, combined with the corrected discrete value, to determine the possibility index of dust hazards, so that the analysis of the possibility index can effectively combine multi-dimensional data features, avoid inaccurate effects caused by grayscale feature analysis alone, and perform early warning judgments through the numerical changes of the possibility indicators of adjacent acquisition time frames, combined with the change characteristics of adjacent acquisition time frames, so that the early warning judgment is combined with dynamic change analysis, thereby improving the reliability of the overall intelligent hidden danger detection.

[0072] The present invention also provides an intelligent early warning system for coal mine safety accidents. The system includes a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the steps of the aforementioned intelligent early warning method for coal mine safety accidents are implemented.

[0073] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent early warning method for coal mine safety accidents, characterized in that: The method comprises: Obtain underground coal mine monitoring images at each acquisition time frame; obtain the overall high-frequency and low-frequency information of the underground coal mine monitoring images based on Gaussian filtering, and extract dust points with high-frequency characteristics; Each coal mine underground monitoring image is grayscale processed and evenly divided to obtain different image areas. The grayscale discrete index of each image area is determined based on the numerical distribution discreteness of the grayscale values ​​of all pixels in each image area. Analyze the dust quality and determine the quality correction index of each image area based on the grayscale discrete index of the image area at any height and below it; correct the grayscale discrete index based on the quality correction index to obtain a corrected discrete value; Combined with the number of dust points and the corrected discrete value in each image area, the probability index of dust hazard occurring in the corresponding acquisition time frame is determined, and early warning judgment is made based on the values ​​of the probability index in adjacent acquisition time frames; Determining the quality correction index of each image area according to the grayscale discrete index of any height and the image area below it includes: The height of any point is taken as the target height, and the heights of all points below the target height are taken as candidate heights; Calculate the sum of the grayscale discrete indexes at the target height and all candidate heights, and normalize them as the quality correction index; The step of performing early warning judgment based on the values ​​of the possibility indicators of adjacent acquisition time frames includes: Determine the image area whose likelihood index of all image areas in the same acquisition frame is greater than a preset likelihood threshold as the warning area; Whether to issue a warning is determined based on the change in area of ​​the warning area between the current and previous acquisition time frames, and the average value of the possibility indicators of all warning areas in the current acquisition time frame.

2. The intelligent early warning method for coal mine safety accidents according to claim 1, characterized in that: The method of obtaining the overall high-frequency and low-frequency information of the underground coal mine monitoring image based on Gaussian filtering and extracting dust points with high-frequency features includes: The coal mine underground monitoring image is smoothed based on Gaussian filtering and used as a low-frequency image; Calculate the brightness ratio of the coal mine underground monitoring image and the low-frequency image at the same point as the high-frequency feature index at the corresponding point; The dust point is determined according to the numerical value of the high-frequency characteristic index.

3. The intelligent early warning method for coal mine safety accidents according to claim 2, characterized in that: Determining the dust point according to the numerical value of the high-frequency characteristic index includes: The points where the high-frequency characteristic index is greater than a preset index threshold are regarded as dust points under high-frequency characteristics.

4. The intelligent early warning method for coal mine safety accidents according to claim 1, characterized in that: Determining the grayscale discreteness index of each image region according to the numerical distribution discreteness of the grayscale values ​​of all pixels in each image region includes: Calculate the standard deviation of the grayscale values ​​of all pixels in the same image area as the grayscale dispersion index.

5. The intelligent early warning method for coal mine safety accidents according to claim 1, characterized in that: The grayscale discrete index is corrected in combination with the quality correction index to obtain a corrected discrete value, including: The product value of the quality correction index and the grayscale discrete index of the image area is calculated as the correction discrete value.

6. The intelligent early warning method for coal mine safety accidents according to claim 1, characterized in that: The method combines the number of dust points and the corrected discrete value in each image area to determine the probability index of dust hazard occurring in the corresponding acquisition time frame, including: The product value of the number of dust points in the image area and the corrected discrete value is calculated and normalized to be used as a possibility indicator.

7. The intelligent early warning method for coal mine safety accidents according to claim 6, characterized in that: The determining whether to issue a warning based on the change in the area of ​​the warning area between the current and previous acquisition time frames and the average of the possibility indicators of all warning areas in the current acquisition time frame includes: Calculate the difference in the warning area between the current acquisition time frame and the previous acquisition time frame, and normalize it as the area impact indicator; The product of the average value of all warning area possibility indicators and the area impact indicator is normalized and used as the warning judgment value; When the warning judgment value is greater than the preset warning threshold, it is determined to perform warning processing; otherwise, it is determined not to perform warning processing.

8. An intelligent early warning system for coal mine safety accidents, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images