A mining area subsidence monitoring method and system based on image data

Abstract

The present invention relates to the technical field of mine safety monitoring, and more specifically, to a mine settlement monitoring method and system based on image data, including: collecting mine cave images in the mine area, detecting the edges of the mine cave images, and obtaining edge images; screening out high-definition images and images to be enhanced according to the clarity of the edge images, wherein the clarity reflects the pixel values ​​and position change degree of the pixel points at the same position of two adjacent frames of edge images. The present invention calculates the supplementary coefficient of each pixel point using the features between the high-definition image and the image to be enhanced, enhances the image to be enhanced, so that a complete mine cave edge can be obtained in the image to be enhanced, and finally calculates the moment when the mine cave edge undergoes a sudden change, and according to the difference and change time between the area of ​​the mine cave edge at the sudden change moment and the area of ​​the mine cave edge at the current moment, the settlement coefficient of the mine area can be accurately calculated to monitor the safety of the mine cave.

Description

Technical Field

[0001] The present invention relates to the technical field of mine safety monitoring, and more specifically, to a mine settlement monitoring method and system based on image data. Background Art

[0002] Mining work will form empty layers below the ground. After the empty layers are formed, the support force of the ground is reduced, which will cause ground subsidence or collapse. Ground subsidence is a long-term accumulation process. After accumulating to a certain extent, it will suddenly collapse in a very short time. Therefore, buildings around the mining area need to monitor their subsidence in real time to facilitate early prediction of danger and early evacuation. Mining area subsidence monitoring is an important part of ensuring the safe operation of the mining area. Especially during the mining process, due to changes in the surface structure, the mining area may experience varying degrees of ground subsidence. Traditional mining area subsidence monitoring methods mainly rely on ground measurement, GPS monitoring, geological surveys and other means.

[0003] In the related technology, for example, the Chinese patent application document with publication number CN119064930A discloses a method and system for predicting dynamic subsidence basins in mining areas based on a combined time function. According to the surface deformation results of the mining area, a combined Usher time function model is used to predict a first surface subsidence result; the first surface subsidence result is the surface subsidence amount of the measured area at any time in the future; according to the first surface subsidence result, an IPIM model is used to predict a second surface subsidence result; the second surface subsidence result is the surface subsidence amount of the entire basin in the measured area. This patent application document can improve the accuracy of prediction of dynamic subsidence basins in mining areas.

[0004] At present, in the process of underground mining, smoke and dust are present in the process of excavation, worker movement, mine car movement and other factors, making the mine image unclear in the real-time monitoring screen. Therefore, when using image processing methods to monitor the subsidence of the mining area, it is impossible to accurately obtain the degree of subsidence in the mining area. Summary of the invention

[0005] The present invention provides a mining area subsidence monitoring method and system based on image data, aiming to solve the problem in the related art that the mining area subsidence degree cannot be accurately obtained when the mining area subsidence is monitored by using image processing method due to unclear mine cave images in real-time monitoring pictures.

[0006] In the first aspect, the present invention provides a mining area settlement monitoring method based on image data, including collecting mine cave images in the mining area, detecting the edges of the mine cave images to obtain edge images; screening out high-definition images and images to be enhanced according to the clarity of the edge images, wherein the clarity reflects the pixel values ​​and position change degrees of pixel points at the same position of two adjacent frames of edge images, and enhancing the images to be enhanced, thereby calculating the settlement coefficient of the mining area, and monitoring the safety of the mining area based on the size of the settlement coefficient of the mining area, wherein the settlement coefficient is positively correlated with the degree of deformation of the mine cave edge in the edge image; enhancing the images to be enhanced , including: obtaining the high-definition image closest to the image to be enhanced, and subtracting the pixel values ​​of the edge of the high-definition image from the edge of the image to be enhanced to obtain a difference sequence, clustering the difference sequence to obtain multiple clusters, and dividing them into clear clusters and unclear clusters according to the mean value of the difference in each cluster; taking any pixel point on the boundary of the clear cluster in the image to be enhanced as the starting point, selecting the pixel point corresponding to the maximum value of the supplementary coefficient belonging to the unclear cluster in its neighborhood for supplementation; wherein the supplementary coefficient of the pixel point and the similarity between the pixel point in the neighborhood range of the pixel point and the pixel point at the same position of the high-definition image are positively correlated. By adopting this method, the quality of the mining area image can be significantly improved through the difference enhancement technology between the high-definition image and the image to be enhanced, thereby improving the accuracy of the calculation of the sedimentation coefficient of the mining area and providing more accurate data support for the safety monitoring of the mining area. In the process of image enhancement, the supplementary coefficient of the pixel point is calculated through cluster analysis and pixel similarity, which ensures the accuracy and stability of the enhancement effect and avoids excessive noise interference.

[0007] Furthermore, the settlement coefficient of the mining area is calculated. , the calculation formula is: ; In the formula, , Respectively Time, The number of pixels in the connected domain of the mine edge in the target image at the moment, For the Moment and The interval between moments, The preset reaction time.

[0008] Furthermore, the safety monitoring of the mining area is carried out based on the size of the settlement coefficient of the mining area, including: if the settlement coefficient of the mining area is greater than the settlement alarm threshold, an alarm prompt is issued; if the settlement coefficient of the mining area is less than or equal to the settlement alarm threshold, the work continues. When the settlement coefficient of the mining area is greater than the set settlement alarm threshold, the system will trigger an alarm to remind the mining area management personnel to take preventive measures in time. Through real-time monitoring of settlement, early warning can be issued before the settlement reaches a dangerous level, and potential safety risks can be identified in advance. This helps to avoid large-scale mining accidents and reduce casualties and property losses.

[0009] Further, The method for determining the time includes: calculating the sedimentation mutation coefficient of each target image, wherein the target image includes a high-definition image and an enhanced image to be enhanced; selecting the acquisition time of the target image corresponding to the maximum sedimentation mutation coefficient as the first time.

[0010] Furthermore, the sedimentation mutation coefficient of each target image is calculated, and the calculation formula is: ; In the formula, For the The sedimentation mutation coefficient of the target image at time , For the The number of pixels in the connected domain of the mine edge in the target image at the moment, Indicates taking an integer. is the duration of abnormal settlement between the jth moment and the ith moment, represents the hyperbolic tangent function, By analyzing the ratio of the number of pixels in the connected domain at the edge of the mine in the target image and combining the scaling of the time difference by the tanh function, the nodes with significant settlement mutations in time can be accurately captured, which is convenient for the subsequent calculation of the settlement coefficient of the mining area.

[0011] Furthermore, the clusters are divided into clear clusters and unclear clusters according to the mean value of the difference within each cluster, including: if the mean value of the cluster is greater than the first threshold, the cluster is an unclear cluster; when the mean value of the cluster is less than or equal to the first threshold, the cluster is a clear cluster. By calculating the mean value of the difference within each cluster, the internal consistency of the image or data point can be effectively measured. If the mean value of the difference within a cluster is large, it means that the image quality within the cluster is poor, and there may be more noise or blurred areas, indicating that the cluster is an unclear cluster. The cluster with a smaller mean value means that the image quality within the cluster is better and the image details are clearer, so it is judged as a clear cluster.

[0012] Furthermore, pixels The supplementary coefficient The calculation formula is: ; In the formula, is the pixel point in the image to be enhanced A vector of pixel values ​​of pixels in the neighborhood that belong to non-clear clusters. For high-resolution images and acquisition is a vector composed of the pixel values ​​of the pixels at the same position. is the pixel point in the image to be enhanced A vector of pixel values ​​belonging to clear clusters in the neighborhood. For high-resolution images and acquisition is a vector composed of the pixel values ​​of the pixels at the same position. Similarity calculation. According to the similarity between the pixel points near the edge of the mine in the image to be enhanced and the high-definition image, the supplement coefficient of the pixel point can be accurately calculated. The larger the supplement coefficient, the more the pixel point belongs to the feature of the clear edge, and the pixel point is used for supplementation.

[0013] Furthermore, high-definition images and images to be enhanced are screened out, including: if the clarity of the edge image is greater than the high-definition threshold, the edge image frame is a high-definition image; if the clarity of the edge image is less than or equal to the high-definition threshold and greater than the low-definition threshold, the edge image frame is an image to be enhanced; if the clarity of the edge image is less than or equal to the low-definition threshold, the edge image frame is a low-definition image, and the edge images belonging to the low-definition images are eliminated. Low-definition edge images may interfere with subsequent image enhancement, recognition or processing tasks because they contain more noise or unclear features. Eliminating these low-definition images and preventing them from entering further processing procedures can help improve processing efficiency, reduce waste of computing resources, and ensure that subsequent operations are more efficient.

[0014] In a second aspect, a mine subsidence monitoring system based on image data is also provided, comprising a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to implement any of the mine subsidence monitoring methods based on image data described above.

[0015] Beneficial effect: The supplementary coefficient of each pixel point is calculated by utilizing the features between the high-definition image and the image to be enhanced, and the image to be enhanced is enhanced so that the complete edge of the mine can be obtained in the image to be enhanced. Finally, the moment when the edge of the mine changes suddenly is calculated using the high-definition image and the enhanced image to be enhanced. According to the difference between the area of ​​the edge of the mine at the moment of mutation and the area of ​​the edge of the mine at the current moment and the change time, the settlement coefficient of the mining area can be calculated, thereby accurately judging the safety of the mine. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The following detailed description is read with reference to the accompanying drawings, which illustrate several embodiments of the present invention in an exemplary and non-limiting manner, and in which like or corresponding reference numerals represent like or corresponding parts, wherein:

[0017] Figure 1 is a flow chart schematically illustrating a method of calculating a settlement coefficient of a mining area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0019] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0020] like Figure 1 As shown, S101: collecting mine cave images in the mining area.

[0021] In one embodiment, a mine cave in a mining area is photographed by a photographing device to obtain a mine cave image, wherein the photographing device is a high-definition camera. After obtaining the mine cave image, the mine cave image is preprocessed to obtain an edge image. The preprocessing includes: converting the mine cave image into a grayscale image, marking the pixel points belonging to the mine cave area in the grayscale image, and then performing edge detection on the mine cave area to obtain an edge image of the mine cave. Among them, the edge detection can use a Canny detection algorithm.

[0022] It should be noted that the reason for obtaining the image of the edge of the mine is that when the mine sinks, the edge of the mine will undergo relative deformation, for example, the upper and lower parts of the mine will deform, or the left and right sides of the mine will deform. Subsequently, it can be determined whether the mining area has sunk based on the degree of deformation between the edge of the mine at the initial moment and the edge of the mine at the current moment.

[0023] In one embodiment, during the mining process in a mining area, due to the continuous progress of various mining activities, especially the influence of various factors such as excavation operations, movement of workers, and movement of mine cars, the internal environment of the mine cave often undergoes complex changes. These activities not only generate noise and vibration, but are also accompanied by the generation of a large amount of smoke and dust, which seriously affects the air quality and visibility on site. Due to the interference of smoke and dust, the images of the mine cave taken by the camera often become blurred, seriously affecting the image quality. In a high-concentration smoke and dust environment, the edges of the mine cave in the image are difficult to identify, and the details and feature points are blurred, resulting in the failure of traditional edge detection, target recognition, and image analysis technologies. Therefore, it is necessary to further identify and enhance the edges according to this method.

[0024] S102: Obtain the clarity of each edge image.

[0025] In one embodiment, the edge image is divided into windows to obtain multiple windows of equal size, the purpose of which is to find the distribution characteristics of the pixel points in each window area, so that the edge of the mine on the edge image can be extracted more accurately. The clarity of each frame edge image is judged according to the distribution characteristics of the mine edge. Specifically, the clarity of the edge image of the latter frame is calculated according to the edge characteristics of the edge images of the previous and next frames, wherein the clarity of the edge in each frame image is negatively correlated with the pixel value and position change degree of the pixel point at the same position of the edge image of the previous frame, and the greater the change degree, the lower the clarity of the edge image of the frame, and vice versa.

[0026] It should be noted that when calculating the clarity of the edge image using the edge features of the two frames of edge images, the two frames of images need to be calculated sequentially along the timeline. And the edge in the initial edge image is completely clear, which means that the personnel can clearly see the complete edge of the mine on the edge image. So far, the edge image is taken as the first frame of the timeline, and the clarity of the subsequent edge images is calculated.

[0027] In one embodiment, calculating the clarity of each frame edge image includes: calculating the matching degree of the edge images of the previous and next two frames, and the calculation formula is: ; In the formula, is the matching degree of the edge images of the previous and next two frames, is the nth pixel position of the mine edge in the edge image of the previous frame, is the nth pixel position of the edge of the mine in the next frame edge image, is the pixel value of the nth pixel point at the edge of the mine in the edge image of the previous frame, is the pixel value of the nth pixel point at the edge of the mine in the next frame edge image, is the number of all pixels on the edge of the mine hole in the edge image of the previous frame. Then the edge of the mine hole in the edge image of the previous frame is traversed on the image of the next frame, and a matching degree is obtained each time. The maximum matching degree is selected as the clarity of the edge image of the next frame.

[0028] S103: Screening out high-definition images and images to be enhanced.

[0029] In one embodiment, after obtaining the clarity of each frame edge image, high-definition images and images to be enhanced are screened out according to the clarity of each frame edge image. Specifically, if the clarity of the frame edge image is greater than the high-definition threshold, the frame edge image is a high-definition image; if the clarity of the frame edge image is less than or equal to the high-definition threshold and greater than the low-definition threshold, the frame edge image is an image to be enhanced; if the clarity of the frame edge image is less than or equal to the low-definition threshold, the frame edge image is a low-definition image, and the edge images belonging to the low-definition images are eliminated. At this point, the image frames that belong to the images to be enhanced in sequence along the timeline are obtained.

[0030] S104: Enhance the image to be enhanced.

[0031] In one embodiment, since the image to be enhanced is limited by its clarity, there may be unclear or incomplete areas in the image to be enhanced. The reason is that the dust and smoke in the mining area will block the edge of the mine cave, resulting in the shooting equipment being unable to obtain a clear image of the edge of the mine cave. Therefore, it is necessary to enhance the unclear areas in the image to be enhanced so that the complete edge of the mine cave can be obtained in the image to be enhanced.

[0032] In one embodiment, a method for enhancing an image to be enhanced includes: selecting a high-definition edge image that is closest to the current image to be enhanced and before it is taken, and subtracting the pixel values ​​of the high-definition edge image from the pixel values ​​of the current image to be enhanced to obtain a difference sequence, and performing ordered sample clustering on the difference sequence to obtain multiple clusters on the current image to be enhanced, and the multiple clusters are continuous, and calculating the mean of the difference in the cluster as the clarity of the cluster, when the clarity of the cluster is greater than a first threshold, the cluster is an unclear cluster; when the clarity of the cluster is less than or equal to the first threshold, the cluster is a clear cluster, wherein the experience of the first threshold is 5, and the pixel points of the unclear cluster are supplemented. Then, taking any pixel point on the boundary of the clear cluster as the starting point, select the pixel point corresponding to the maximum value of the supplement coefficient in the unclear cluster in its neighborhood to supplement, until all the pixel points on the edge of the mine are supplemented on the current image to be enhanced, and a new complete image edge on the image to be enhanced is obtained, thereby completing the enhancement.

[0033] In one embodiment, the supplementation process is as follows: starting from any pixel point on the boundary of the clear cluster cluster, the pixel points in the unclear cluster cluster are continuously explored and supplemented. Specifically, the pixel point corresponding to the starting point is selected, the supplementation coefficient of each pixel point in the unclear cluster cluster in its neighborhood is calculated, and the pixel point corresponding to the maximum value of the supplementation coefficient in its neighborhood is selected as the supplementation point, which is supplemented to the clear cluster cluster. Then, starting from the supplementation point, the pixel point corresponding to the maximum value of the supplementation coefficient in the neighborhood of the supplementation point is selected as the next supplementation point. The above steps are repeated until the maximum supplementation coefficient of each pixel point in the neighborhood of a certain supplementation point is less than the second threshold value, and the supplementation is stopped. The experience of the second threshold value is 0.5. In other embodiments, the experience value of the second threshold value can be 0.6 or 0.55, etc., which can be adjusted according to the specific implementation situation. If the edge of the mine cave is not continuous at this time, the nearest pixel points of two adjacent clear cluster clusters are connected to complete the continuous supplementation of the edge of the mine cave. At this point, the enhancement of the image to be enhanced is completed, and the enhanced image to be enhanced and the high-definition image are used as target images for calculating the settlement coefficient of the subsequent mining area.

[0034] Specifically, the pixel points in the neighborhood of the pixel point corresponding to the starting point belong to the unclear cluster For example, to calculate the pixel points The supplementary coefficient , the calculation formula is: ,in, is the pixel point in the image to be enhanced A vector of pixel values ​​of pixels in the neighborhood that belong to non-clear clusters. For high-resolution images and acquisition is a vector composed of the pixel values ​​of the pixels at the same position. is the pixel point in the image to be enhanced A vector of pixel values ​​belonging to clear clusters in the neighborhood. For high-resolution images and acquisition is a vector composed of the pixel values ​​of the pixels at the same position. For similarity calculation, the similarity can be calculated using cosine similarity or Euclidean distance. The supplementary coefficient The larger the value, the more similar the edge of the mine hole in the image to be enhanced is to that in the high-definition image. The more it conforms to the characteristics of clear edges.

[0035] S105: Calculate the settlement coefficient of the mining area.

[0036] Specifically, the target images sorted along the time series are obtained, and the sedimentation mutation coefficient of each target image is calculated. , and then select the target image corresponding to the maximum value of the settlement mutation coefficient from each target image to calculate the settlement coefficient of the mining area. The calculation formula of the settlement mutation coefficient of each target image is: ; For the The sedimentation mutation coefficient of the target image at time , For the The number of pixels in the connected domain of the mine edge in the target image at the moment, Indicates taking an integer. For the Moment and The duration of abnormal subsidence between the times, represents the hyperbolic tangent function, Indicates the average value. If at a certain moment in the time series A mutation occurs, indicating that the target image at this moment is significantly different from the target images before and after. The sedimentation mutation coefficient of the target image at the moment will be larger. When the maximum value is taken, it means that the moment corresponding to the maximum value of the settlement mutation coefficient is the moment when the mine accelerates settlement.

[0037] In one embodiment, the target image corresponding to the maximum value of the sedimentation mutation coefficient is then screened out from each target image to calculate the sedimentation coefficient of the mining area. , the calculation formula is: ; In the formula, , Respectively Time, The number of pixels in the connected domain of the mine edge in the target image at the moment, is the jth moment and the The interval between moments, The preset response time is set by the user according to his / her own needs. The larger is, the lower is the sensitivity of w, and vice versa. is the standard normalization function. The area of ​​the mine edge in the target image at time The greater the difference between the edge of the mine hole in the target image at the moment, the greater the deformation degree of the edge of the mine hole, and the greater the settlement coefficient of the mining area. And when the settlement coefficient of the mining area is greater than the settlement alarm threshold, an alarm is issued to remind relevant personnel to handle it. Among them, the empirical value of the settlement alarm threshold is 0.5. In other embodiments, the empirical value of the settlement alarm threshold can be 0.6 or 0.55, etc., which can be adjusted according to the specific implementation situation.

[0038] Through the above steps, the supplementary coefficient of each pixel point is calculated using the features between the high-definition image and the image to be enhanced, and the image to be enhanced is enhanced so that the complete edge of the mine can be obtained in the image to be enhanced. Finally, the moment when the edge of the mine changes suddenly is calculated using the high-definition image and the enhanced image to be enhanced. According to the difference between the area of ​​the edge of the mine at the moment of mutation and the area of ​​the edge of the mine at the current moment and the change time, the settlement coefficient of the mining area can be calculated, thereby accurately judging the safety of the mine.

[0039] The present invention also provides a mining area subsidence monitoring system based on image data. The system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a mining area subsidence monitoring method based on image data according to the first aspect of the present invention is implemented.

[0040] The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, whose configuration and functions are known in the art and thus will not be described in detail here.

[0041] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions stored or otherwise maintained in such a computer-readable medium.

[0042] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0043] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0044] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent application. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A mining area subsidence monitoring method based on image data, characterized in that: include: Collecting mine cave images in the mining area, detecting edges of the mine cave images, and obtaining edge images; According to the clarity of the edge image, a high-definition image and an image to be enhanced are screened out, wherein the clarity reflects the pixel value and position change degree of the pixel point at the same position of two adjacent frames of edge images, and the image to be enhanced is enhanced, and then the settlement coefficient of the mining area is calculated, and the mining area safety is monitored based on the settlement coefficient of the mining area, wherein the settlement coefficient is positively correlated with the deformation degree of the edge of the mine cave in the edge image; Sedimentation coefficient of the mining area The calculation formula is: ; , Respectively Time, The number of pixels in the connected domain of the mine edge in the target image at the moment, For the Moment and The interval between moments, To preset the reaction time, is the standard normalization function; The image to be enhanced is enhanced, including: obtaining a high-definition image that is closest in time to the image to be enhanced, and subtracting pixel values ​​of an edge of the high-definition image from an edge of the image to be enhanced to obtain a difference sequence, clustering the difference sequence to obtain a plurality of clusters, and dividing the difference sequence into clear clusters and unclear clusters according to the mean value of the difference in each cluster; Taking any pixel point on the boundary of the clear cluster in the image to be enhanced as the starting point, select the pixel point corresponding to the maximum value of the supplement coefficient of the unclear cluster in its neighborhood for supplementation; Pixels The supplementary coefficient The calculation formula is: ; is the pixel point in the image to be enhanced A vector of pixel values ​​of pixels in the neighborhood that belong to unclear clusters. For high-resolution images and acquisition is a vector composed of the pixel values ​​of the pixels at the same position. is the pixel point in the image to be enhanced A vector of pixel values ​​belonging to clear clusters in the neighborhood. For high-resolution images and acquisition is a vector composed of the pixel values ​​of the pixels at the same position. Calculate the similarity.

2. The method for monitoring mining area subsidence based on image data according to claim 1, characterized in that: Safety monitoring of the mining area based on the size of the settlement coefficient of the mining area includes: If the sedimentation coefficient of the mining area is greater than the sedimentation alarm threshold, an alarm is issued; If the settlement coefficient of the mining area is less than or equal to the settlement alarm threshold, continue working.

3. The method for monitoring mining subsidence based on image data according to claim 1, characterized in that: No. The method of determining the time includes: Calculating the sedimentation mutation coefficient of each target image, wherein the target image includes a high-definition image and an enhanced image to be enhanced; The acquisition time of the target image corresponding to the maximum value of the sedimentation mutation coefficient is selected as time.

4. The method for monitoring mining area subsidence based on image data according to claim 1, characterized in that: Calculate the sedimentation mutation coefficient of each target image, and the calculation formula is: ； In the formula, For the The sedimentation mutation coefficient of the target image at time , For the The number of pixels in the connected domain of the mine edge in the target image at the moment, Indicates taking an integer. is the duration of abnormal settlement between the jth moment and the ith moment, represents the hyperbolic tangent function, Indicates the average value.

5. The method for monitoring mining area subsidence based on image data according to claim 1, characterized in that: According to the mean value of the difference within each cluster, it is divided into clear clusters and unclear clusters, including: Calculate the mean of the differences in the clusters as the clarity of the clusters; If the clarity of the cluster is greater than a first threshold, the cluster is an unclear cluster; When the clarity of the cluster is less than or equal to the first threshold, the cluster is a clear cluster.

6. The method for monitoring mining area subsidence based on image data according to claim 1, characterized in that: Screen out high-definition images and images to be enhanced, including: If the clarity of the edge image is greater than the high-definition threshold, the edge image of this frame is a high-definition image; If the clarity of the edge image is less than or equal to the high-definition threshold and greater than the low-definition threshold, then the edge image of this frame is the image to be enhanced; If the clarity of the edge image is less than or equal to the low-definition threshold, the edge image of this frame is a low-definition image, and the edge image belonging to the low-definition image is eliminated.

7. A mining area subsidence monitoring system based on image data, comprising a processor and a memory, characterized in that: The memory stores a computer program, and the processor executes the computer program to implement the mining area subsidence monitoring method based on image data as described in any one of claims 1-6.

Images