A coal mine filling mining surface settlement reduction measurement method and system

Abstract

The application relates to the technical field of mine surface subsidence measurement, in particular to a coal mine filling mining surface settlement reduction measurement method and system, which comprises the following steps: acquiring SAR images and leveling measurement data collected in a coal mine area each time; obtaining a subsidence calibration index of each measurement point according to the spatial distribution of all measurement points and measurement points in the neighborhood of the measurement points in the SAR images and the observation elevation difference in the leveling measurement data; calibrating the SAR images according to the subsidence calibration index to obtain coal mine surface elevation data; and obtaining a surface subsidence measurement result according to the difference between adjacent coal mine surface elevation data collected at different times, so that the measurement precision of the surface subsidence is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of surface subsidence measurement of mining area, and particularly relates to a coal mine filling mining surface settlement reduction measurement method and system. BACKGROUND

[0002] The mining subsidence in the process of coal mining can cause a large amount of farmland damage and soil quality decline in the mining area. Surface settlement reduction, also known as ground subsidence, refers to the phenomenon that the soil, rock and other solid materials in the surface or ground of a region are reduced in volume or weight due to natural factors or human activities, thereby causing the elevation of the surface or ground to be lowered. Therefore, when coal mining is carried out, the hollow area is filled after mining, which reduces the rate of surface subsidence to a certain extent. In order to avoid the influence of surface subsidence on people's production and life, it is necessary to measure the surface subsidence in the process of coal mining.

[0003] In the method of measuring surface subsidence, InSAR satellite-borne synthetic aperture radar interferometry is to perform differential processing on two SAR images, obtain elevation and other related information according to the interference phase, and then complete the measurement of surface subsidence. When the coal mine is mined, the surface will gradually become empty with the progress of mining, the stress originally borne by the coal will gradually be borne by the rock and soil layer around the coal, the stress of the rock and soil layer will gradually concentrate, and the surface subsidence rate will accelerate. When the local surface subsidence deformation amplitude is large, the precision of the SAR image data is poor, and the precision of the coal mine surface subsidence measurement by using adjacent two SAR image data is low. SUMMARY

[0004] In order to solve the above technical problems, the purpose of the present application is to provide a coal mine filling mining surface settlement reduction measurement method and system, and the technical scheme adopted is as follows:

[0005] In the first aspect, the present application provides a coal mine filling mining surface settlement reduction measurement method, which comprises the following steps:

[0006] S1, acquiring SAR images and leveling measurement data collected each time in the coal mine area, wherein the leveling measurement data includes the observed elevation of all measurement points;

[0007] S2, obtaining the settlement calibration index of each measurement point according to the spatial distribution of all measurement points in the SAR image and the observed elevation difference in the leveling measurement data, specifically comprising:

[0008] S21, dividing the near neighbor points of each measurement point according to the distribution of the measurement points in the SAR image, and obtaining the local settlement index according to the observed elevation difference between the measurement points and their near neighbor points;

[0009] S22, the strong settlement aggregation index of the measurement point is obtained based on the numerical distribution of the local settlement index of all neighboring points of the measurement point;

[0010] S23, the nearest neighbor with the largest strong settlement aggregation index among all the nearest neighbor points of each measurement point is recorded as the local settlement center point; the settlement calibration index of each measurement point is obtained based on the differences in strong settlement aggregation index and observation elevation between all the nearest neighbor points and the local settlement center point.

[0011] S3. The SAR image is calibrated according to the settlement calibration index of all measurement points to obtain the surface elevation data of the coal mine; the surface settlement measurement results are obtained based on the difference between the surface elevation data of the coal mine with adjacent acquisition times.

[0012] In the above scheme, by analyzing the local subsidence characteristics of the gradual concentration of stress in the rock and soil layers of the coal mine surface, the nearest points of each measurement point are identified. Based on the difference in observed elevation between each measurement point and its nearest points, a local subsidence index is constructed to reflect the degree of local ground subsidence, thus more accurately evaluating the possibility of local surface subsidence. Further analysis is conducted on the impact of surface subsidence deformation in cavity areas on SAR image accuracy. Based on the local subsidence index, a strong subsidence clustering index is constructed to determine the degree of membership of each measurement point in areas prone to subsidence. Further analysis is conducted on the similar subsidence characteristics among measurement points within surface subsidence areas. Firstly, the nearest points of each measurement point are selected... The local subsidence center point is used as the measurement point. Based on the strong subsidence clustering index and the difference in observation elevation between other nearby points and the local subsidence center point, the subsidence calibration index of each measurement point is obtained. The trend change of the subsidence degree describes the subsidence characteristics of the coal mine surface. The more obvious the subsidence characteristics, the greater the impact on the accuracy of SAR image at the measurement point, and the larger the calibration range should be. Finally, the surface subsidence measurement results of each acquisition relative to the previous acquisition time are obtained based on the calibrated coal mine surface elevation data. This solves the problem that the large local surface subsidence deformation amplitude when measuring surface subsidence in two adjacent measurements makes the SAR image data inaccurate, and improves the measurement accuracy of surface subsidence.

[0013] Furthermore, the step of dividing the nearest points of each measurement point according to the distribution of measurement points in the SAR image includes:

[0014] In SAR imagery, obtain the Euclidean distance between each measurement point and other measurement points; then, select the first preset number of other measurement points in the sequence formed by arranging all other measurement points in ascending order according to the Euclidean distance as the nearest neighbors of each measurement point.

[0015] In the above scheme, the distribution of measurement points in SAR images is analyzed to identify the nearest points of each measurement point, so as to facilitate further analysis of the subsidence of the surface at each measurement point.

[0016] Furthermore, obtaining the local settlement index based on the difference in observed elevation between the measurement point and its nearest neighbor points includes:

[0017] Obtain the absolute value of the difference between the observed elevations of any two nearest neighbor points of each measurement point; the local settlement index at each measurement point is positively correlated with the absolute value of the difference and negatively correlated with the Euclidean distance between any two nearest neighbor points of each measurement point.

[0018] In the above scheme, the local subsidence index reflects the difference in observed elevation between neighboring points of the measurement point, and evaluates the possibility of local surface subsidence at the measurement point in a relatively accurate manner.

[0019] Furthermore, obtaining the strong settlement aggregation index of the measurement point based on the numerical distribution of the local settlement indices of all neighboring points includes:

[0020] All measurement points are clustered based on the local settlement index to obtain each cluster;

[0021] Obtain the set of nearest neighbors for each measurement point; based on the membership of the nearest neighbors in the set of nearest neighbors to each cluster, obtain the settlement weight of each nearest neighbor.

[0022] Obtain the maximum value of the local settlement index of all nearest points in the nearest point set; calculate the difference between the maximum value and the local settlement index of each nearest point; the strong settlement aggregation index of each measurement point is positively correlated with the settlement weight of all nearest points of the measurement point, and negatively correlated with the difference of all nearest points of the measurement point.

[0023] In the above scheme, a strong settlement clustering index of the measurement point is constructed by using the settlement weights of all neighboring points of the measurement point and the magnitude of the local settlement index. Compared with direct clustering, by analyzing the similarity of settlement characteristics, the degree of membership of the measurement point to the coal mine settlement area can be determined more accurately.

[0024] Furthermore, obtaining the settlement weights of each nearest neighbor point includes:

[0025] The nearest neighbor points that belong to the same cluster as each of the nearest neighbor points in the set of nearest neighbor points are denoted as the co-cluster points of each nearest neighbor point; the ratio of the number of all co-cluster points of each nearest neighbor point to the number of nearest neighbor points contained in the set of nearest neighbor points is denoted as the category proportion of each nearest neighbor point; the local settlement index of all co-cluster points of each nearest neighbor point and the category proportion are both positively correlated with the settlement weight of each nearest neighbor point.

[0026] In the above scheme, the settlement weight is calculated based on the degree of membership of the nearest neighbor points to each cluster, so that the settlement weight of the measurement points with more nearest neighbors belonging to the coal mine settlement area is larger, which increases the distinguishability of the settlement situation of measurement points belonging to different categories.

[0027] Furthermore, obtaining the settlement calibration index for each measurement point includes:

[0028] Sort all nearest neighbor points in ascending order of their Euclidean distance from the local settlement center to obtain the settlement center cluster sequence for each measurement point;

[0029] Based on the differences in strong settlement aggregation index and observed elevation between each nearest neighbor point in the settlement center aggregation sequence and the local settlement center point, the local settlement trend sequence and elevation trend sequence are obtained respectively.

[0030] The correlation between the local settlement trend sequence and the elevation trend sequence is calculated; the correlation and the strong settlement aggregation index of each measurement point are positively correlated with the settlement calibration index of each measurement point.

[0031] In the above scheme, the correlation between the settlement change trend and the elevation change trend of each neighboring point in the settlement center cluster sequence of the measurement point is used to obtain the degree of synchronization between the settlement trend and the elevation trend, which reflects the accuracy of the data at the measurement point; combined with the strong settlement clustering index, the settlement calibration index is obtained, and the extent to which the fitted elevation at each measurement point needs to be calibrated is comprehensively evaluated.

[0032] Furthermore, obtaining the local subsidence trend sequence and elevation trend sequence includes:

[0033] The sequence formed by the difference between the strong settlement aggregation index of each nearest neighbor point and the local settlement center in the settlement center aggregation sequence is denoted as the local settlement trend sequence; the sequence formed by the absolute value of the difference between the observed elevation of each nearest neighbor point and the local settlement center in the settlement center aggregation sequence is denoted as the elevation trend sequence.

[0034] In the above scheme, the spatial distribution of all neighboring points is sorted to make the calculated local subsidence trend sequence and elevation trend sequence more consistent with the local terrain.

[0035] Furthermore, the step of calibrating the SAR image based on the settlement calibration index of all measurement points to obtain coal mine surface elevation data specifically includes:

[0036] The SAR images collected in the coal mining area were converted into a DEM digital elevation model to obtain the fitted elevation of each measurement point.

[0037] The fitted elevation, observed elevation, and settlement calibration index of all measurement points are used as inputs to a CNN convolutional neural network to calibrate the fitted elevation of each measurement point, thus obtaining the calibrated elevation of each measurement point. Based on the calibrated elevations of all measurement points, the SAR images acquired in each iteration are registered and corrected to obtain the coal mine surface elevation data.

[0038] In the above scheme, the SAR images acquired in each acquisition are registered and corrected according to the settlement calibration index, which eliminates the impact of large local surface settlement deformation on the accuracy of SAR image data.

[0039] Furthermore, the step of obtaining the surface subsidence measurement results based on the differences between coal mine surface elevation data collected at adjacent times includes:

[0040] Differential processing was performed on the surface elevation data of all adjacent coal mines to obtain the surface subsidence measurement results of each coal mine acquisition relative to the previous acquisition time.

[0041] In the above scheme, the surface settlement measurement results of each collection relative to the previous collection time are obtained based on the calibrated coal mine surface elevation data, which effectively improves the measurement accuracy of surface settlement.

[0042] Secondly, embodiments of this application also provide a surface subsidence measurement system for coal mine backfilling mining, comprising:

[0043] The settlement measurement parameter extraction module is used to acquire SAR images and leveling data collected each time in the coal mining area. The leveling data includes the observed elevation of all measurement points.

[0044] The settlement measurement correction module is used to obtain the settlement calibration index for each measurement point based on the spatial distribution of all measurement points in the SAR image and the differences in observed elevations from the leveling data. The specific implementation is as follows:

[0045] Based on the distribution of measurement points in SAR images, the nearest points of each measurement point are identified; the local subsidence index is obtained based on the difference in observed elevation between the measurement point and its nearest points.

[0046] The strong settlement aggregation index of the measurement point is obtained by the numerical distribution of the local settlement index of all neighboring points of the measurement point.

[0047] The nearest neighbor with the largest strong settlement aggregation index among all the nearest neighbor points of each measurement point is recorded as the local settlement center point; the settlement calibration index of each measurement point is obtained based on the differences in strong settlement aggregation index and observation elevation between all the nearest neighbor points and the local settlement center point.

[0048] The surface subsidence measurement module is used to calibrate SAR images based on the subsidence calibration index of all measurement points to obtain coal mine surface elevation data; and to obtain surface subsidence measurement results based on the differences between coal mine surface elevation data acquired at adjacent times. Attached Figure Description

[0049] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A flowchart illustrating the steps of a method for measuring surface subsidence reduction during coal mine backfilling mining, as provided in one embodiment of this application;

[0051] Figure 2 Block diagram of a surface subsidence measurement system for coal mine backfilling;

[0052] Figure 3 This is a flowchart for obtaining the settlement calibration index. Detailed Implementation

[0053] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. In the absence of conflict, the embodiments of this application and the technical features in the embodiments can be combined with each other. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0054] Unless otherwise defined, 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 application pertains.

[0055] The following description, in conjunction with the accompanying drawings, details the specific scheme of the surface subsidence measurement method and system for coal mine backfilling mining provided in this application.

[0056] Please see Figure 1 The diagram illustrates a flowchart of a method for measuring surface subsidence reduction in coal mine backfilling mining according to an embodiment of this application. The method includes the following steps:

[0057] Step S1: Obtain SAR images and leveling data collected each time in the coal mine area, where the leveling data includes the observed elevation of all measurement points.

[0058] Measurement points for settlement monitoring are evenly distributed in the coal mining area. SAR images and leveling data of the mining area are collected at equal intervals T using the settlement measurement parameter extraction module of the coal mine backfilling mining surface settlement reduction measurement system. The leveling data includes the observed elevation of all measurement points measured using a first-order level instrument. T is 10 days and the number of data collection times is N=30.

[0059] Because environmental interference and other factors may cause missing values ​​or other anomalies in the collected leveling data during the data acquisition process, this application uses a regression filling method to fill in the missing values ​​in the leveling data in order to improve the quality and usability of the leveling data and facilitate more accurate analysis in subsequent steps. The regression filling method is a known technique, and its specific process will not be described in detail here.

[0060] Follow the steps described above to complete the preprocessing of the leveling data obtained from N collections in the mining area.

[0061] Step S2: Based on the spatial distribution of all measurement points in the SAR image and the difference in observed elevation in the leveling data, obtain the settlement calibration index for each measurement point.

[0062] This application analyzes the processing of the settlement measurement correction module in a surface settlement measurement system for coal mine backfilling mining. During underground coal mining activities, the destruction of underground rock masses, coal seam cavities caused by mining, and the construction of underground engineering projects all lead to a gradual concentration of stress in the surface rock and soil layers, resulting in an accelerated surface settlement rate. Therefore, in the collected leveling data, the greater the elevation difference between closely spaced measurement points, the greater the degree of ground settlement.

[0063] Step S21: Based on the distribution of measurement points in the SAR image, divide the nearest points of each measurement point; obtain the local subsidence index based on the difference in observed elevation between the measurement point and its nearest points.

[0064] Based on the above analysis, a local subsidence index is constructed to reflect the degree of local ground subsidence, including:

[0065] Obtain the Euclidean distance between each measurement point and other measurement points in the SAR image; the first M measurement points in the sequence formed by arranging all measurement points in ascending order according to the Euclidean distance are taken as the nearest neighbors of each measurement point; in this embodiment, M=15, which is a manually set value, and the implementer can set the number of nearest neighbors according to the specific situation.

[0066] Obtain the absolute value of the difference between the observed elevations of any two nearest neighbor points of each measurement point; the local settlement index at each measurement point is positively correlated with the absolute value of the difference and negatively correlated with the Euclidean distance between any two nearest neighbor points of each measurement point.

[0067] It should be noted that a positive correlation indicates a unidirectional relationship in which the dependent variable increases as the independent variable increases and decreases as the independent variable decreases. A positive correlation can specifically be a multiplicative relationship, an additive relationship, or a normalized result, etc. A negative correlation indicates a unidirectional relationship in which the dependent variable decreases as the independent variable increases and increases as the independent variable decreases. A negative correlation can specifically be a division relationship, a subtraction relationship, etc., which is determined by the actual application and is not subject to any special restrictions in this application.

[0068] It should be noted that, to ensure the calculation results of the degree of repulsion are meaningful, when the negative correlation is a division relationship, a parameter tuning factor greater than 0 needs to be added to the denominator to prevent the denominator from being zero. The value of the parameter tuning factor is set by the implementer according to the actual situation, and this application does not impose any special restrictions.

[0069] As an embodiment of this application, the ratio of the absolute value of the difference to the Euclidean distance between any two nearest neighbors of each measurement point is calculated; the local settlement index at each measurement point is the sum of the ratios contained in all the nearest neighbors of each measurement point.

[0070] In coal mining areas, the greater the difference in observed elevation between neighboring points with closer Euclidean distances, the greater the underground mining volume and the more significant the surface subsidence, thus resulting in a larger calculated local subsidence index.

[0071] Step S22: Obtain the strong settlement aggregation index of the measurement point based on the numerical distribution of the local settlement indices of all neighboring points of the measurement point.

[0072] In coal mining areas, the surface above areas with cavities created by underground mining is more prone to subsidence and deformation. Using SAR imagery data of these areas in subsequent subsidence measurements may lead to inaccurate readings. Because these areas are more susceptible to subsidence, the more concentrated the distribution of measurement points with higher local subsidence indices in the leveling data, the more likely these areas are to be prone to subsidence. Based on this analysis, this application constructs a strong subsidence clustering index based on local subsidence indices to reflect the degree to which each measurement point in the leveling data of the coal mining area belongs to a subsidence-prone region.

[0073] The local settlement indices of all measurement points are used as input to the K-means clustering algorithm to cluster all measurement points. In this embodiment, the necessary parameter, the number of clusters K, is set to 3. The output is each cluster. The mean of the local settlement index of all measurement points in each cluster is calculated. All clusters are sorted in descending order of the mean. The sorted clusters are then denoted as the coal mine settlement zone, the coal mine settlement buffer zone, and the coal mine stable zone, respectively. The specific implementation process of the K-means clustering algorithm is a well-known technique and will not be described in detail here.

[0074] It should be noted that the K-means clustering algorithm is only one algorithm for clustering measurement points provided in this embodiment. Under the premise that the purpose of clustering can be achieved, implementers may also use other feasible clustering algorithms in the prior art as other implementation methods, and this application does not limit them.

[0075] Obtain the set of nearest neighbors consisting of all nearest neighbors of each measurement point; denot the nearest neighbors in the set that belong to the same cluster as each nearest neighbor as the co-cluster points of each nearest neighbor; denote the ratio of the number of all co-cluster points of each nearest neighbor to the number of nearest neighbors contained in the set of nearest neighbors as the category proportion of each nearest neighbor; the local settlement index of all co-cluster points of each nearest neighbor and the category proportion are both positively correlated with the settlement weight of each nearest neighbor.

[0076] In this embodiment, the mean of the local settlement index of all clustered points of each neighboring point is calculated and denoted as the first mean; the settlement weight of each neighboring point is the product of the category proportion and the first mean.

[0077] In the nearest neighbor set, the more nearest neighbors that belong to the same cluster as each nearest neighbor, and the larger the mean of the local subsidence index, the steeper the location of the data point, and the more likely surface subsidence will occur. Therefore, the calculated subsidence weight is larger, which increases the difference between the subsidence of measurement points belonging to different categories and improves the distinguishability.

[0078] Obtain the maximum value of the local settlement index of all nearest points in the nearest point set; calculate the difference between the maximum value and the local settlement index of each nearest point; the strong settlement aggregation index of each measurement point is positively correlated with the settlement weight of all nearest points of the measurement point, and negatively correlated with the difference of all nearest points of the measurement point.

[0079] In this embodiment, the ratio of the settlement weight of each nearest neighbor point to the difference is calculated and denoted as the first ratio; the strong settlement agglomeration index of each measurement point is the sum of the first ratios of all the nearest neighbors of each measurement point.

[0080] The smaller the difference between the local subsidence index and the maximum local subsidence index of each nearest point in the set of nearest points of the measurement point, the more uneven the overall location of the nearest point is, and the more likely it is to belong to the coal mine subsidence area. Therefore, the calculated strong subsidence agglomeration index is larger.

[0081] Step S23: Record the nearest neighbor with the largest strong settlement aggregation index among all the nearest neighbor points of each measurement point as the local settlement center point; obtain the settlement calibration index of each measurement point based on the differences in strong settlement aggregation index and observation elevation between all the nearest neighbor points and the local settlement center point.

[0082] During coal mining operations, the underground gradually becomes hollow, reducing the supporting force on the surface. The resulting surface subsidence areas should exhibit similar subsidence characteristics to their spatially adjacent areas. On the coal mine surface, the greater the subsidence occurring at adjacent data acquisition times, the greater the error in the surface subsidence measurement results obtained through differential processing of SAR image data from adjacent acquisition times. Therefore, when calibrating SAR image data, the calibration degree should be greater for areas on the coal mine surface more prone to subsidence to improve the accuracy and effectiveness of subsequent surface subsidence measurement results. Based on the degree of subsidence at each data point in the coal mine leveling data obtained through the above steps, a subsidence calibration index can be constructed based on a strong subsidence clustering index to reflect the calibration degree of each data point in the leveling data. The construction process of the subsidence calibration index is as follows:

[0083] The nearest neighbor with the largest strong settlement clustering index in the nearest neighbor set of each measurement point is designated as the local settlement center point. All nearest neighbor points in the nearest neighbor set are sorted in ascending order according to their Euclidean distance from the local settlement center point to obtain the settlement center clustering sequence for each measurement point. The sequence formed by the difference between the strong settlement clustering index of each nearest neighbor point in the settlement center clustering sequence and the local settlement center point is designated as the local settlement trend sequence. The sequence formed by the absolute value of the difference between the observed elevation of each nearest neighbor point in the settlement center clustering sequence and the local settlement center point is designated as the elevation trend sequence.

[0084] The correlation between the local settlement trend sequence and the elevation trend sequence is calculated; the correlation and the strong settlement aggregation index of each measurement point are positively correlated with the settlement calibration index of each measurement point.

[0085] In this embodiment, the settlement calibration index for each measurement point is the product of the strong settlement aggregation index of each measurement point and the correlation.

[0086] It should be noted that the correlation between two sequences can be calculated using methods such as Pearson correlation coefficient or cosine similarity, which should be determined by the actual application. This application does not impose any special restrictions.

[0087] In leveling data, the larger the strong settlement aggregation index of the measurement point, the more obvious the settlement degree. At the same time, the greater the correlation between the local settlement trend sequence and the elevation trend sequence, the more regular the settlement degree changes with the increase of spatial distance. The more consistent it is with the settlement characteristics of the coal mine surface, the greater the degree of calibration of the measurement point location should be. Therefore, the calculated settlement calibration index is larger.

[0088] A flowchart for obtaining the settlement calibration index, as shown below. Figure 3 As shown.

[0089] Step S3: calibrate the SAR image according to the settlement calibration index of all measurement points to obtain coal mine surface elevation data; obtain the surface settlement measurement results based on the difference between coal mine surface elevation data with adjacent acquisition times.

[0090] The settlement calibration index of each data point in the leveling data obtained through the above steps reflects the degree of correction for each data point. InSAR technology can then be used to convert the SAR image data collected in the coal mine area into a DEM (Digital Elevation Model), obtaining the fitted elevation for each measurement point. The fitted elevation, observed elevation, and settlement calibration index of all measurement points are used as input to a CNN (Convolutional Neural Network) to calibrate the fitted elevation of each measurement point. The Adam algorithm is used as the optimization algorithm, and the root mean square error (RMSE) is used as the loss function. The output is the calibrated elevation for each measurement point. Based on the calibrated elevations of all measurement points, the SAR images collected in each step are registered and corrected to obtain the coal mine surface elevation data. The conversion of SAR image data into a DEM using InSAR technology, the training process of the CNN, and the registration and correction methods are all well-known techniques, and the specific details are not elaborated here.

[0091] In the surface subsidence measurement module, differential processing is performed on the surface elevation data of all adjacent coal mines to obtain the surface subsidence measurement results of each coal mine acquisition relative to the previous acquisition time. The differential processing process is a well-known technique and will not be described in detail here.

[0092] Based on the same inventive concept as the above method, this application also provides a surface subsidence measurement system for coal mine backfilling mining, such as... Figure 2 As shown, it includes the following modules:

[0093] The settlement measurement parameter extraction module is used to acquire SAR images and leveling data collected each time in the coal mining area. The leveling data includes the observed elevation of all measurement points.

[0094] The settlement measurement correction module is used to obtain the settlement calibration index for each measurement point based on the spatial distribution of all measurement points in the SAR image and the differences in observed elevations from the leveling data. The specific implementation is as follows:

[0095] Based on the distribution of measurement points in SAR images, the nearest points of each measurement point are identified; the local subsidence index is obtained based on the difference in observed elevation between the measurement point and its nearest points.

[0096] The strong settlement aggregation index of the measurement point is obtained by the numerical distribution of the local settlement index of all neighboring points of the measurement point.

[0097] The nearest neighbor with the largest strong settlement aggregation index among all the nearest neighbor points of each measurement point is recorded as the local settlement center point; the settlement calibration index of each measurement point is obtained based on the differences in strong settlement aggregation index and observation elevation between all the nearest neighbor points and the local settlement center point.

[0098] The surface subsidence measurement module is used to calibrate SAR images based on the subsidence calibration index of all measurement points to obtain coal mine surface elevation data; and to obtain surface subsidence measurement results based on the differences between coal mine surface elevation data acquired at adjacent times.

[0099] Through the above description of the embodiments in conjunction with the accompanying drawings, those skilled in the art will understand that, for the sake of convenience and brevity, the above division of functional modules is only used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0100] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of surface subsidence measurement for coal mine backfilling mining, characterized in that, The method comprises the following steps: S1, obtaining SAR images and leveling data of each acquisition of a coal mine area, wherein the leveling data comprises observed elevations of all measurement points; S2, obtaining a subsidence calibration index of each measurement point according to spatial distribution and observed elevation difference of all measurement points in the SAR images, and the specific implementation is as follows: S21, dividing the near neighbor points of each measurement point according to the distribution of the measurement points in the SAR images; obtaining a local subsidence index according to the observed elevation difference between the measurement points and their near neighbor points; S22, obtaining a strong subsidence aggregation index of the measurement point according to the numerical distribution of the local subsidence indexes of all near neighbor points of the measurement point; S23, taking the near neighbor point with the maximum strong subsidence aggregation index among all near neighbor points of each measurement point as a local subsidence center point; obtaining a subsidence calibration index of each measurement point according to the difference between the strong subsidence aggregation indexes and the observed elevations of all near neighbor points and the local subsidence center point; S3, calibrating the SAR images according to the subsidence calibration indexes of all measurement points to obtain coal mine surface elevation data; obtaining a surface subsidence measurement result according to the difference between the coal mine surface elevation data of adjacent acquisition times; The method comprises the following steps: According to the local subsidence indexes, the measurement points are clustered to obtain each cluster; The near neighbor point set composed of all near neighbor points of each measurement point is obtained; according to the membership of the near neighbor points in the near neighbor point set to each cluster, the subsidence weight of each near neighbor point is obtained; The maximum value of the local subsidence indexes of all near neighbor points in the near neighbor point set is obtained; the difference between the maximum value and the local subsidence indexes of each near neighbor point is calculated; the strong subsidence aggregation index of each measurement point is positively correlated with the subsidence weight of all near neighbor points of the measurement point, and is inversely correlated with the difference between the measurement point and all near neighbor points; The method comprises the following steps: The SAR images of each acquisition of the coal mine area are converted into DEM digital elevation models to obtain a fitting elevation of each measurement point; The fitting elevation, the observed elevation and the subsidence calibration index of all measurement points are taken as inputs of a CNN convolutional neural network to calibrate the fitting elevation of each measurement point, so as to obtain a calibrated elevation of each measurement point; the calibrated elevations of all measurement points are used to correct and register the SAR images of each acquisition, so as to obtain the coal mine surface elevation data.

2. A method of surface settlement reduction measurement for coal mine filling mining as claimed in claim 1, characterized in that, The method comprises the following steps: The Euclidean distances between each measurement point and other measurement points in the SAR images are obtained; the first preset number of other measurement points in a sequence arranged in ascending order according to the Euclidean distances are taken as the near neighbor points of each measurement point.

3. A method of surface settlement reduction measurement for coal mine filling mining according to claim 1, characterized in that, The method comprises the following steps: The method comprises the following steps: An absolute value of a difference between observation elevations of any two neighboring points of each measuring point; the local subsidence index of each measuring point is positively correlated with the absolute value and inversely correlated with a Euclidean distance between any two neighboring points of each measuring point.

4. A method of surface settlement reduction measurement for coal mine filling mining according to claim 1, characterized in that, The obtaining of the subsidence weight of each neighboring point comprises: The neighboring points in the same cluster as each neighboring point in the neighboring point set are recorded as the same-cluster points of each neighboring point; a ratio of a number of all the same-cluster points of each neighboring point to a number of the neighboring points contained in the neighboring point set is recorded as a category proportion of each neighboring point; the local subsidence index of all the same-cluster points of each neighboring point and the category proportion are positively correlated with the subsidence weight of each neighboring point.

5. A method of surface settlement reduction measurement for coal mine filling mining according to claim 1, characterized in that, The obtaining of the subsidence calibration index of each measuring point comprises: The neighboring points are sorted in ascending order of the Euclidean distance between the local subsidence center point and each neighboring point, to obtain a subsidence center aggregation sequence of each measuring point; The local subsidence trend sequence and the elevation trend sequence are obtained according to the strong subsidence aggregation index and the observation elevation difference between each neighboring point and the local subsidence center point in the subsidence center aggregation sequence; The correlation between the local subsidence trend sequence and the elevation trend sequence is calculated; the correlation and the strong subsidence aggregation index of each measuring point are positively correlated with the subsidence calibration index of each measuring point.

6. A method of surface settlement reduction measurement for coal mine filling mining as claimed in claim 5, characterised in that, The obtaining of the local subsidence trend sequence and the elevation trend sequence comprises: A sequence formed by the difference between the strong subsidence aggregation index of each neighboring point and the local subsidence center point in the subsidence center aggregation sequence is recorded as the local subsidence trend sequence; a sequence formed by the absolute value of the difference between the observation elevation of each neighboring point and the local subsidence center point in the subsidence center aggregation sequence is recorded as the elevation trend sequence.

7. A method of surface settlement reduction measurement for coal mine filling mining according to claim 1, characterized in that, The obtaining of the surface subsidence measurement result according to the difference between the adjacent coal mine surface elevation data comprises: The adjacent coal mine surface elevation data are subjected to difference processing to obtain the surface subsidence measurement result of each collection of the coal mine relative to the last collection time.

8. A coal mine fill mining surface settlement measurement system implementing the method as claimed in claim 1, characterized by, The coal mine backfill mining surface subsidence measurement system comprises: A subsidence measurement parameter extraction module is configured to acquire SAR images and leveling measurement data of each collection of the coal mine region, wherein the leveling measurement data comprises observation elevations of all the measuring points; A subsidence measurement correction module is configured to obtain a subsidence calibration index of each measuring point according to the spatial distribution and the observation elevation difference of all the measuring points in the SAR images, and the subsidence calibration index is obtained as follows: The neighboring points of each measuring point are divided according to the distribution of the measuring points in the SAR images; the local subsidence index is obtained according to the observation elevation difference between the measuring points and the neighboring points; The strong subsidence aggregation index of each measuring point is obtained according to the numerical distribution of the local subsidence index of all the neighboring points of the measuring point; The neighboring point with the largest strong subsidence aggregation index among all the neighboring points of each measuring point is recorded as the local subsidence center point; the subsidence calibration index of each measuring point is obtained according to the strong subsidence aggregation index and the observation elevation difference between all the neighboring points and the local subsidence center point; The ground subsidence measurement module is used for calibrating the SAR image according to the subsidence calibration indexes of all measurement points, obtaining the coal mine ground elevation data, and obtaining the ground subsidence measurement result according to the difference between the adjacent coal mine ground elevation data in the collection time.

Images