Mine tunnel deformation monitoring method and system

Abstract

The present invention relates to the field of deformation monitoring technology, and specifically to a mine tunnel deformation monitoring method and system. The present invention analyzes the number of intersections contained in each monitoring area in a three-dimensional model of a mine tunnel, determines the tolerance of the area where updated point cloud data is missing in each monitoring area during the model update process, and screens and obtains successfully updated point cloud data; then, for the successfully obtained updated point cloud data, by analyzing the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained last time, the effectiveness of the updated point cloud data is determined; then, based on the correlation between the point cloud data of the monitoring area and the adjacent monitoring areas, it is determined whether the collection position of the monitoring area needs to be offset during the three-dimensional model update process, as well as the corresponding offset direction and offset degree; and then new updated three-dimensional point cloud data is collected to obtain more accurate monitoring results when the mine tunnel deformation is subsequently monitored.

Description

Technical Field

[0001] The present invention relates to the technical field of deformation monitoring, and in particular to a mine tunnel deformation monitoring method and system. Background Art

[0002] Mine tunnel deformation monitoring is currently shifting from traditional manual inspections to new, high-precision, real-time technologies. Traditional monitoring methods suffer from low data collection efficiency, insufficient analysis capabilities, and inadequate real-time response mechanisms, making them inadequate for modern mine safety. Emerging technologies such as GNSS, laser ranging, 3D laser scanning, fiber optic sensing, and the Internet of Things are gaining widespread application in mine tunnel deformation monitoring. These technologies offer advantages such as high precision, real-time performance, and a high degree of automation, providing valuable insights into mine safety.

[0003] Currently, when monitoring mine tunnel deformation based on emerging technologies, multiple radars are typically installed within the tunnel to obtain high-precision monitoring data. Due to the complex geological conditions within the mine, advancing mining often generates large amounts of dust and fog, and existing monitoring points are often damaged. Consequently, blind spots often occur in these complex geological sections. This can lead to anomalies in the subsequent update of the tunnel's 3D model, resulting in anomalies in the deformation monitoring of the tunnel. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a mine tunnel deformation monitoring method and system.

[0005] According to a first aspect of an embodiment of the present invention, a method for monitoring deformation of a mine tunnel is provided, wherein the technical solution adopted is as follows:

[0006] Collect point cloud data from multiple monitoring areas in the mine tunnel to obtain the overall three-dimensional model of the mine tunnel;

[0007] Based on the overall three-dimensional model, counting the number of intersections included in the monitoring area and determining the tolerance of missing areas in the monitoring area;

[0008] Recollecting updated point cloud data of the monitoring area, obtaining the number and location of missing areas in the monitoring area, and determining whether the updated point cloud data of the monitoring area is successfully acquired based on the tolerance;

[0009] If yes, analyzing the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area;

[0010] Based on the effectiveness, analyzing the correlation between the updated point cloud data of the monitoring area and adjacent monitoring areas, and determining whether the updated point cloud data of the monitoring area is effective;

[0011] If not, then based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained in the last monitoring, analyze the offset of the acquisition position of the monitoring area, offset the acquisition position, and re-acquire new updated point cloud data until it is valid;

[0012] Based on the valid updated point cloud data, the overall three-dimensional model update is completed.

[0013] In some embodiments of the present invention, recollecting updated point cloud data of the monitoring area, obtaining the number and positions of missing areas of the monitoring area, and determining whether the updated point cloud data of the monitoring area is successfully acquired in combination with the tolerance includes:

[0014] Recollecting updated point cloud data of the monitoring area at the last monitoring position;

[0015] Gridding the updated point cloud data, and using a connected domain labeling algorithm to obtain the number and locations of missing areas in the monitoring area;

[0016] Determining whether the number of missing regions is less than the tolerance;

[0017] If not, the acquisition of updated point cloud data of the monitoring area fails;

[0018] If yes, determining whether the missing area is at an intersection based on the missing location;

[0019] If yes, the acquisition of updated point cloud data of the monitoring area fails;

[0020] If not, the updated point cloud data of the monitoring area is acquired successfully.

[0021] In some embodiments of the present invention, if the acquisition of the updated point cloud data of the monitoring area fails, the updated point cloud data of the monitoring area is re-collected.

[0022] In some embodiments of the present invention, analyzing the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area includes:

[0023] Calculating the quantitative difference between the updated point cloud data of the monitoring area and the point cloud data obtained during the last monitoring to obtain a quantitative difference degree;

[0024] Obtaining matching point cloud data and unmatched point cloud data between the updated point cloud data of the monitoring area and the point cloud data obtained during the last monitoring, and analyzing the degree of discreteness of the unmatched point cloud data to obtain a matching degree between the updated point cloud data of the monitoring area and the point cloud data obtained during the last monitoring;

[0025] The validity degree of the updated point cloud data of the monitoring area is obtained according to the degree of quantity difference and the degree of matching.

[0026] In some embodiments of the present invention, analyzing the correlation between the updated point cloud data of the monitoring area and adjacent monitoring areas based on the effectiveness, and determining whether the updated point cloud data of the monitoring area is effective, includes:

[0027] Marking the monitoring area that has point cloud data overlapping with the monitoring area as the monitoring area to be matched;

[0028] Analyze the change relationship between the updated point cloud data of the monitoring area to be matched and the point cloud data obtained from the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area to be matched;

[0029] Analyzing the size correlation between the effectiveness of the monitoring area and the monitoring area to be matched, and determining whether the updated point cloud data of the monitoring area is initially invalid;

[0030] If not, the updated point cloud data of the monitoring area is valid;

[0031] If yes, obtaining the number of matching point clouds between the monitoring area and the monitoring area to be matched based on the point cloud data acquired during the measurement of the monitoring area;

[0032] and obtaining the number of updated matching point clouds between the monitoring area and the monitoring area to be matched based on the updated point cloud data obtained during the measurement of the monitoring area to be matched;

[0033] Analyzing the difference between the number of the matched point cloud and the number of the updated matched point cloud to determine whether the updated point cloud data of the monitoring area is valid;

[0034] If yes, the updated point cloud data of the monitoring area is valid;

[0035] If not, the updated point cloud data of the monitoring area is invalid.

[0036] In some embodiments of the present invention, analyzing the correlation between the effectiveness of the monitoring area and the monitoring area to be matched, and determining whether the updated point cloud data of the monitoring area is preliminarily invalid, includes:

[0037] Setting a validity threshold to determine whether the validity levels corresponding to the monitoring area and the monitoring area to be matched are both less than the validity threshold;

[0038] If yes, the updated point cloud data of the monitoring area is valid;

[0039] If not, the updated point cloud data of the monitoring area is preliminarily invalid.

[0040] In some embodiments of the present invention, analyzing the difference between the number of the matching point cloud and the number of the updated matching point cloud to determine whether the updated point cloud data of the monitoring area is valid includes:

[0041] Setting a difference threshold to determine whether the difference between the number of matching point clouds and the number of updated matching point clouds is less than the difference threshold;

[0042] If yes, the updated point cloud data of the monitoring area is valid;

[0043] If not, the updated point cloud data of the monitoring area is invalid.

[0044] In some embodiments of the present invention, analyzing the offset of the acquisition position of the monitoring area based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained in the last monitoring, and offsetting the acquisition position includes:

[0045] Calculate the average value of the position coordinates of the matching point cloud data in the point cloud data of the monitoring area during the last measurement process, and record it as the first feature position;

[0046] Calculate the mean value of the position coordinates of the updated point cloud data of the monitoring area and record it as the second characteristic position;

[0047] Calculating the difference between the first characteristic position and the second characteristic position, and combining the preset displacement of the acquisition position to obtain an offset distance;

[0048] Based on the acquisition position, the acquisition center line is obtained;

[0049] Recording a feature position corresponding to a maximum value of the number of matching point clouds and the number of updated matching point clouds as an offset feature coordinate;

[0050] The direction of the offset feature coordinates relative to the current acquisition position is used as the offset direction;

[0051] The acquisition position is offset on the acquisition center line according to the offset distance and the offset direction.

[0052] According to a second aspect of an embodiment of the present invention, a mine tunnel deformation monitoring system is provided, comprising: a memory and a processor, wherein:

[0053] The memory is used to store program code;

[0054] The processor is configured to read the program code stored in the memory and execute the method described in the first aspect of the embodiment of the present invention.

[0055] In some embodiments of the present invention, the processor includes:

[0056] The 3D model building module is used to collect point cloud data from multiple monitoring areas in the mine tunnel to obtain the overall 3D model of the mine tunnel;

[0057] a tolerance analysis module, configured to count the number of intersections included in the monitoring area based on the overall three-dimensional model and determine the tolerance of missing areas in the monitoring area;

[0058] An updated point cloud data acquisition module is used to re-collect updated point cloud data of the monitoring area, obtain the number and location of missing areas in the monitoring area, and determine whether the updated point cloud data of the monitoring area is successfully acquired in combination with the tolerance;

[0059] An updated point cloud data validity judgment module is configured to, when the updated point cloud data is successfully acquired, analyze the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the previous monitoring to obtain the validity of the updated point cloud data of the monitoring area; and based on the validity, analyze the correlation between the updated point cloud data of the monitoring area and the updated point cloud data of adjacent monitoring areas to determine whether the updated point cloud data of the monitoring area is valid;

[0060] An updated point cloud data re-collection module is configured to, when the updated point cloud data is invalid, analyze the offset of the collection position of the monitoring area based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained from the last monitoring, and offset the collection position; and re-collect new updated point cloud data after the collection position is offset until it is valid;

[0061] The three-dimensional model updating module is used to complete the overall three-dimensional model updating based on the valid updated point cloud data.

[0062] Compared with the existing technology, the mine tunnel deformation monitoring method and system provided by the present invention have the following beneficial effects:

[0063] The present invention determines the tolerance for missing areas of updated point cloud data in each monitoring area during the update process by analyzing the degree of intersections contained in each monitoring area in the three-dimensional model of the mine tunnel; re-collects the updated point cloud data of the monitoring area to obtain the number and location of missing areas in the monitoring area, and uses the tolerance to determine whether the updated point cloud data of the monitoring area has been successfully acquired; determines the updated point cloud data that failed to be acquired based on the tolerance, and directly re-collects it, saving subsequent computing resources. If the updated point cloud data is successfully acquired, the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the previous monitoring is analyzed to obtain the validity of the updated point cloud data of the monitoring area; then, based on the validity, the correlation between the updated point cloud data of the monitoring area and the adjacent monitoring areas is analyzed to determine whether the updated point cloud data of the monitoring area is valid, thereby eliminating the impact of the mining machine operation on the validity of the acquired updated point cloud data. For updated point cloud data that fails to be acquired or is invalid, the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained in the last monitoring are used to analyze the offset of the acquisition position of the monitoring area, offset the acquisition position, and re-acquire more accurate new updated point cloud data. In the subsequent monitoring of mine tunnel deformation, more accurate monitoring results are obtained. Through the present invention, accurate and reliable mine tunnel deformation information can be obtained, and maintenance personnel can then promptly maintain dangerous areas of the mine tunnel, ensuring the progress of mine tunnel excavation work and the safety of personnel in the mine tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] 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.

[0065] Figure 1 A schematic diagram of the basic process of a mine tunnel deformation monitoring method provided by one embodiment of the present invention;

[0066] Figure 2 A schematic diagram of an overall three-dimensional model of a mine tunnel provided by one embodiment of the present invention;

[0067] Figure 3 A centerline schematic diagram provided by one embodiment of the present invention;

[0068] Figure 4 A schematic diagram of displaying deformation information on a three-dimensional model of a mine tunnel provided by one embodiment of the present invention;

[0069] Figure 5A schematic diagram of the basic components of a mine tunnel deformation monitoring system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0070] 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 a mine tunnel deformation monitoring method and system according to the present invention, including its specific implementation, structure, features, and effectiveness. 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.

[0071] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. Terms such as "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a circuit structure, article, or device comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such article or device. In the absence of further limitations, the phrase "comprising a ..." to define an element does not preclude the presence of other identical elements in the article or device comprising the element.

[0072] The following describes in detail a method for monitoring deformation of a mine tunnel provided by the present invention with reference to the accompanying drawings.

[0073] See also Figure 1 , which shows the basic process of a mine tunnel deformation monitoring method provided by an embodiment of the present invention.

[0074] like Figure 1 As shown, an embodiment of the present invention provides a mine tunnel deformation monitoring method, which specifically includes:

[0075] S100: Collect point cloud data of multiple monitoring areas in the mine tunnel to obtain an overall three-dimensional model of the mine tunnel.

[0076] During mine tunneling, there are multiple potential hazards, such as roof collapse during drilling and pumice removal, and roof collapse during blasting. Therefore, to ensure operational safety and the safety of both mine workers and personnel, a portable radar monitoring system is used to monitor the tunnels.

[0077] A portable radar monitoring system was installed approximately 20 cm from the monitoring area. After system assembly, wiring, and power-up, it was activated to collect 3D data of the mine tunnels within the monitoring area. This acquisition took approximately 20 to 30 minutes to obtain point cloud data for the monitored area. The acquired point cloud data was then pre-processed using denoising and resampling to remove debris such as pipes, cables, and human figures within the mine tunnels. This resulted in a point cloud dataset that only retained the tunnel walls.

[0078] After completing data acquisition in one monitoring area, the portable radar monitoring system's acquisition location is recorded. The portable radar monitoring system is then moved to another monitoring area and the acquisition continues as described above. In particular, to ensure the integrity of subsequent mine tunnel modeling, the point cloud data acquired in different monitoring areas will have a certain degree of overlap.

[0079] After the acquisition is completed, the point cloud data of all monitoring areas will be input into the 3D model construction system for modeling, and then the 3D model of each monitoring area in the mine tunnel and the overall 3D model of the mine tunnel will be obtained, such as Figure 2 shown.

[0080] After a complete mine tunnel model is obtained according to the above steps, the mine tunnel model needs to be updated in the subsequent mine tunnel deformation monitoring process, which specifically includes steps S200 to S600.

[0081] S200: Based on the overall three-dimensional model, count the number of intersections included in the monitoring area and determine the tolerance of missing areas in the monitoring area.

[0082] Since the geological sections in mines are relatively complex, a large amount of fog and dust is usually generated after the mining face advances. Therefore, monitoring blind spots often appear in complex geological sections, resulting in missing areas in data collection. Different mine conditions have different tolerances for missing areas. Therefore, first, based on the overall three-dimensional model, the number of intersections included in the monitoring area is counted to determine the tolerance for missing areas in the monitoring area. The specific implementation method is as follows:

[0083] In the overall 3D model of the mine tunnel, the number of intersections in each monitoring area is counted, and the coordinates of each intersection are recorded (expressed by the coordinates of the center point of the intersection). The more intersections a monitoring area contains, the more complex the monitoring area is. Therefore, in order to more accurately render the detailed features when building the 3D model later, the fewer missing areas in the monitoring area should be during the data collection process. The calculation formula for the tolerance of missing areas in a monitoring area is:

[0084]

[0085] Where, Indicates the Tolerance for missing areas in each monitoring area; Indicates the The number of intersections included in each monitoring area; Indicates the total number of monitoring areas during the last measurement; Indicates the total number of intersections included in all monitoring areas during the last measurement; Indicates the threshold for the number of missing regions (the value can be 3, based on experience).

[0086] Indicates the The relative number of intersections contained in a monitoring area relative to the total number of intersections contained in all monitoring areas. The larger the value, the more intersections the monitoring area contains, and the more complex the corresponding monitoring area is. Therefore, in order to more accurately render the detailed features when building the 3D model later, the corresponding missing area of ​​the monitoring area should be less during the data collection process. The tolerance of missing areas in the monitoring area is obtained by weighting the number threshold of the confirmed areas.

[0087] S300: re-collecting updated point cloud data of the monitoring area, obtaining the number and location of missing areas in the monitoring area, and determining whether the updated point cloud data of the monitoring area is successfully acquired based on the tolerance.

[0088] Re-collect the updated point cloud data of the monitoring area, obtain the number and location of missing areas in the monitoring area, and determine whether the updated point cloud data of the monitoring area is successfully acquired based on the tolerance. Further steps include:

[0089] First, the updated point cloud data of the monitoring area is collected again at the last monitoring location. The point cloud data of the mine tunnel is collected again at the last monitoring location, and the point cloud data is preprocessed to obtain the first Updated point cloud data for each monitoring area.

[0090] Then, the updated point cloud data is gridded and the connected domain labeling algorithm is used to obtain the number and location of missing areas in the monitoring area. The updated point cloud data of each monitoring area is gridded (existing technology), and the connected domain labeling algorithm is used to identify the missing areas in the updated point cloud data, thereby obtaining the first The number of missing areas and missing locations of the updated point cloud data of each monitoring area; since the updated point cloud data has been gridded, the number of missing areas is the number of missing grids, recorded as , the missing position is the coordinate of the center point of the corresponding grid.

[0091] Then, determine whether the number of missing areas is less than the tolerance; if not, the acquisition of updated point cloud data for the monitoring area fails; if yes, based on the missing location, determine whether the missing area is at an intersection; if yes, the acquisition of updated point cloud data for the monitoring area fails; if not, the acquisition of updated point cloud data for the monitoring area succeeds.

[0092] Specifically, For example, when the monitoring area The number of missing areas in the updated point cloud data of the monitoring area Greater than The tolerance of missing areas in each monitoring area is , indicating that at this time The missing area of ​​the updated point cloud data obtained for a monitoring area is too large, and the updated point cloud data of the monitoring area needs to be obtained again.

[0093] When The number of missing areas in the updated point cloud data of the monitoring area Less than or equal to The tolerance of missing areas in each monitoring area is Since the missing area should be avoided as much as possible in the intersection area, it is necessary to further consider whether the missing area is in the key intersection area. More specifically: calculate the The distance between the position coordinates of each missing area in the monitoring area and the coordinates of all intersections is recorded as the minimum value of the distance as the degree of each missing area in the key intersection area. Then the average value of the degree of all missing areas in the key intersection area is recorded as the first The degree value of the missing area in the monitoring area in the critical cross-path area is recorded as X. Preset degree value threshold , the specific value can be 0.5m (based on experience). When The missing area of ​​the updated point cloud data obtained in the monitoring area is too biased towards the cross-path area, indicating that the The updated point cloud data of the monitoring area failed to be acquired, and the updated point cloud data of the monitoring area needs to be collected again. The acquisition of the updated point cloud number of the monitoring area is successful. At this time, it is necessary to quantify the acquired The method specifically includes steps S400 and S500 to determine whether the updated point cloud data of each monitoring area is valid.

[0094] S400: Analyze the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area.

[0095] Analyze the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area, further including:

[0096] First, calculate the quantitative difference between the updated point cloud data of the monitoring area and the point cloud data obtained in the last monitoring to obtain the degree of quantitative difference. The difference between the updated point cloud data of a monitoring area and the point cloud data measured last time in the monitoring area indicates the degree of quantitative difference, which is recorded as ;in The larger the value is, the more data the updated point cloud data has. This means that the updated point cloud data has more data than the last measurement, and thus more detailed features can be displayed when the 3D model is subsequently constructed. Therefore, the degree of quantitative difference is obtained, and then the following steps are included: judging the initial validity of the updated point cloud data in the monitoring area based on the degree of quantitative difference. Specifically, the preset quantitative difference threshold , The value can be 200 (based on experience). The difference is less than or equal to When The number of updated point cloud data collected in each monitoring area does not meet the requirements, that is, The updated point cloud data of the monitoring area is invalid, so the The acquisition position of the portable radar monitoring system in each monitoring area is offset, and new updated point cloud data is collected again. The specific offset method is the same as step S600 and will not be repeated here. Greater than the preset quantity difference threshold , then judge the The updated point cloud data of each monitoring area is initially valid. The matching degree between the updated point cloud data of a monitoring area and the last measured point cloud data of the monitoring area.

[0097] Then, the matching point cloud data and unmatched point cloud data of the updated point cloud data of the monitoring area and the point cloud data obtained from the last monitoring are obtained, and the discrete degree of the unmatched point cloud data is analyzed to obtain the matching degree of the updated point cloud data of the monitoring area and the point cloud data obtained from the last monitoring.

[0098] Since the tunnel boring machine in the mine tunnel will generate a lot of dust during the operation when acquiring the updated point cloud data, it is necessary to determine whether the acquired updated point cloud data has any serious loss.

[0099] Specifically, the ICP algorithm is used to calculate the The updated point cloud data of each monitoring area matches the point cloud data of the last measurement in the monitoring area, and the number of matching point clouds is recorded as , the number of unmatched point cloud data is recorded as (divided into The updated point cloud data of the monitoring area does not match the point cloud dataset There is no matching point cloud dataset in the last measured point cloud data of the monitoring area ).

[0100] Unmatched point cloud dataset The mean Euclidean distance between all updated point cloud data is recorded as the unmatched point cloud dataset The degree of discreteness. Similarly, the unmatched point cloud dataset The mean Euclidean distance between all point cloud data is recorded as the unmatched point cloud dataset The degree of discreteness of the point cloud dataset will not be matched. The degree of dispersion and unmatched point cloud dataset The mean of the dispersion degree is recorded as The degree of dispersion of unmatched point cloud data in a monitoring area is expressed as . The larger the value, the higher the degree of dispersion of the unmatched point cloud data, which means that the unmatched point cloud data does not show obvious aggregation, and it is less likely that missing areas will appear in the future. The more matching point cloud data the updated point cloud data obtained in a monitoring area has with the point cloud data obtained in the last measurement of the monitoring area, and the higher the discrete degree of the unmatched point cloud data, the higher the corresponding The higher the matching degree between the updated point cloud data obtained in the monitoring area and the point cloud data obtained in the last measurement of the monitoring area, the better the result. The formula for calculating the matching degree between the updated point cloud data obtained in a monitoring area and the point cloud data obtained in the last measurement of the monitoring area is:

[0101]

[0102] Where, Indicates the The matching degree between the updated point cloud data obtained in each monitoring area and the point cloud data obtained in the last measurement of the monitoring area; Indicates the The updated point cloud data of each monitoring area matches the point cloud data obtained by the last measurement of the monitoring area; Indicates the The updated point cloud data of a monitoring area does not match the point cloud data obtained by the last measurement of the monitoring area; Indicates the discrete degree of unmatched point cloud data; Represents the normalization function.

[0103] Indicates the The proportion of matching point cloud data of the updated point cloud data obtained in the monitoring area in all point cloud data. The larger the value, the higher the The more updated point cloud data obtained in a monitoring area matches the point cloud data obtained in the last measurement of the monitoring area, the greater the number of matching point cloud data; The larger the value, the higher the degree of dispersion of the unmatched point cloud data, which means that the unmatched point cloud data does not show obvious aggregation, and thus it is less likely that missing areas will appear in the future; The larger the value, the The higher the matching degree between the updated point cloud data obtained in a monitoring area and the point cloud data obtained in the last measurement of the monitoring area, the better the matching degree will be.

[0104] Finally, according to the degree of quantity difference and matching, the effectiveness of the updated point cloud data in the monitoring area is obtained. The matching degree between the updated point cloud data obtained in each monitoring area and the point cloud data obtained in the last measurement of the monitoring area and Combine to determine the The effectiveness of the updated point cloud data obtained in each monitoring area. The corresponding mathematical formula is:

[0105]

[0106] Where, Indicates the The effectiveness of the updated point cloud data obtained in each monitoring area; Indicates the The matching degree between the updated point cloud data obtained in each monitoring area and the point cloud data obtained in the last measurement of the monitoring area; Indicates the The quantitative difference between the updated point cloud data of a monitoring area and the point cloud data obtained by the last measurement of the monitoring area, that is, the degree of quantitative difference; Indicates the preset quantity difference threshold, that is, the minimum threshold of the quantity difference; Represents the normalization function.

[0107] The larger the value, the The more updated point cloud data obtained in each monitoring area can show the detailed characteristics of the mine tunnel, the more effective the updated point cloud data obtained in the monitoring area is; the matching degree The larger the value, the The greater the reliability of the updated point cloud data obtained in a monitoring area, the greater the effectiveness of the updated point cloud data obtained in the monitoring area.

[0108] S500: Based on the effectiveness, analyzing the correlation between the updated point cloud data of the monitoring area and the adjacent monitoring areas, and determining whether the updated point cloud data of the monitoring area is effective.

[0109] The above steps calculated The effectiveness of the updated point cloud data of each monitoring area is determined by the correlation with the historical point cloud data at the same location. When using the portable radar monitoring system to build a 3D model of the mine tunnel, in order to ensure full coverage of the tunnel, there is usually a certain overlap between the point cloud data obtained from adjacent monitoring areas. When the staff is using the portable radar monitoring system to collect data, the mining machine is also working synchronously, so there is a certain overlap between the point cloud data obtained from the next monitoring area. After the measurement of the monitoring area is completed, the mining machine Mining is carried out in each monitoring area, which affects the accuracy of the updated point cloud data collected subsequently.

[0110] Based on the above analysis, in an embodiment of the present invention, based on the degree of effectiveness, the correlation between the updated point cloud data of the monitoring area and the adjacent monitoring areas is analyzed to determine whether the updated point cloud data of the monitoring area is valid. Further including:

[0111] First, the monitoring area that overlaps with the monitoring area with point cloud data is marked as the monitoring area to be matched. monitoring areas, and the remaining monitoring areas have monitoring areas and There is overlap of point cloud data in the monitoring area. The monitoring area is recorded as The monitoring areas to be matched in each monitoring area.

[0112] Then, the relationship between the updated point cloud data of the monitoring area to be matched and the point cloud data obtained in the last monitoring is analyzed to obtain the effectiveness of the updated point cloud data of the monitoring area to be matched. The effectiveness of the updated point cloud data obtained in each monitoring area is obtained. The effectiveness of the updated point cloud data of the monitoring area to be matched , I will not go into details here.

[0113] Then, analyze the size correlation of the effectiveness of the monitoring area and the monitoring area to be matched, and judge whether the updated point cloud data of the monitoring area is initially invalid; if not, the updated point cloud data of the monitoring area is valid; if yes, based on the point cloud data obtained during the measurement of the monitoring area, obtain the number of matching point clouds between the monitoring area and the monitoring area to be matched; and based on the updated point cloud data obtained during the measurement of the monitoring area to be matched, obtain the number of updated matching point clouds between the monitoring area and the monitoring area to be matched; analyze the degree of difference between the number of matching point clouds and the number of updated matching point clouds, and judge whether the updated point cloud data of the monitoring area is valid; if yes, the updated point cloud data of the monitoring area is valid; if not, the updated point cloud data of the monitoring area is invalid. The specific implementation method is:

[0114] Setting the validity threshold , The value can be 0.5 (based on experience), to determine whether the effectiveness of the monitoring area and the monitoring area to be matched are both less than the effectiveness threshold; if so, that is, ,and , indicating the monitoring areas and The validity values ​​of the updated point cloud data of the monitoring areas to be matched are relatively small, which means that the monitoring areas and The distribution of point cloud data of the two monitoring areas may change due to the mining of the mining machine, so the first The updated point cloud data of the monitoring area is valid; if not, when The value and The value is not less than Or both are greater than When , the updated point cloud data of the monitoring area is initially invalid, and then further judgment is made as to whether the updated point cloud data of the monitoring area that is initially invalid is valid. After the measurement of the first monitoring area is completed, the staff will go to the next monitoring area for measurement, and the next monitoring area is the The monitoring area to be matched in the monitoring area is set as monitoring areas to be matched. After the measurement of the monitoring area to be matched is completed, the ICP algorithm is used to calculate the The updated point cloud data of the monitoring area to be matched and the The number of updated matching point clouds of the updated point cloud data of the monitoring area is recorded as ; The last measurement process is summarized in monitoring areas and The number of matching point clouds in the monitoring area to be matched is recorded as . Set the difference threshold , The value can be 0.2 (based on experience) to determine the number of matching point clouds Update the number of matching point clouds Is the difference less than the difference threshold? If so, , then it means the The effectiveness of the updated point cloud data of the monitoring area is good, and the updated point cloud data of the monitoring area is effective; if not, that is, , then it means the The effectiveness of the updated point cloud data of each monitoring area is poor, and the updated point cloud data of the monitoring area is invalid. It is necessary to offset the collection position and re-collect new updated point cloud data. The specific method is step S600.

[0115] S600: Based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained in the last monitoring, the acquisition position offset of the monitoring area is analyzed, the acquisition position is offset, and new updated point cloud data is re-collected until it is valid.

[0116] After steps S400 and S500, invalid data in the updated point cloud data is obtained. It is necessary to offset the acquisition position and then re-acquire new updated point cloud data. Therefore, based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained from the last monitoring, the acquisition position offset of the monitoring area is analyzed, the acquisition position is offset, and new updated point cloud data is re-acquired to complete the overall 3D model update. The specific implementation method is as follows:

[0117] First, calculate the number of The mean value of the position coordinates of the matching point cloud data in the point cloud data of the monitoring area is recorded as the first feature position; and the The average of the updated point cloud data position coordinates of the monitoring area is recorded as the second feature position; the difference between the first feature position and the second feature position is calculated and recorded as the offset distance , combined with the preset displacement of the acquisition position (Assume that the acquisition position deviates by y meters each time, the corresponding offset distance is ), acquisition position preset displacement The value can be 10 meters (based on experience), so the distance the collection position needs to be offset is , that is, the offset distance .

[0118] Then, since the distance between the acquisition position and the edge of the roadway remains unchanged during the offset process, its movement is relative motion. Therefore, based on the acquisition position, the acquisition center line is obtained, such as Figure 3 As shown, the acquisition position is offset on the center line.

[0119] Then, the number of matching point clouds will be And update the number of matching point clouds The feature position corresponding to the maximum value of is recorded as the offset feature coordinate, that is, if the number of matching point clouds is updated Relative to the number of matching point clouds If it is larger, the The average value of the updated point cloud data position coordinates of the monitoring area (the second feature position) is recorded as the offset feature coordinate; if the number of matching point clouds Relative to updating the number of matching point clouds If it is larger, the The mean value of the position coordinates of the point cloud data of each monitoring area (the first feature position) is recorded as the offset feature coordinate.

[0120] The offset direction is determined by the direction of the offset feature coordinate relative to the last acquisition position. Specifically, the offset feature coordinate is projected onto a horizontal plane, and the centerline is divided into two parts, front and back, by the current acquisition position. If the offset feature coordinate projection position is in front of the centerline relative to the original acquisition position, the corresponding offset direction is forward along the centerline; otherwise, the offset direction is backward.

[0121] Then, the acquisition position is offset on the acquisition center line according to the offset distance and offset direction.

[0122] Finally, the new updated point cloud data is collected again to complete the overall 3D model update. Specifically, after the migration is completed, the first The new updated point cloud data of the monitoring area is obtained, and then the method of step S200 to step S500 is continued to determine the obtained The validity of the new updated point cloud data of each monitoring area is determined until the new updated point cloud data is valid. Collection of updated point cloud data for each monitoring area.

[0123] S700: Complete the overall 3D model update based on valid updated point cloud data.

[0124] Based on the valid updated point cloud data, the overall 3D model is updated. The valid updated point cloud data includes the valid updated point cloud data obtained in step S500 and the new valid updated point cloud data obtained in step S600. When the updated point cloud data of all monitoring areas are collected, the overall 3D model of the mine tunnel is updated. After the update is completed, the deformation monitoring system automatically superimposes the deformation information on the 3D model for display, such as Figure 4The figure below shows deformation information displayed on a 3D model of a mine tunnel. Different colors represent different deformation values. When the deformation speed or value in the monitoring area exceeds the set warning threshold, the system issues an alarm, and maintenance personnel perform maintenance on the dangerous areas of the mine tunnel based on the alert.

[0125] Based on the same inventive concept as the above method, this embodiment also provides a mine tunnel deformation monitoring system.

[0126] See also Figure 5 , which shows the basic composition of a mine tunnel deformation monitoring system provided by an embodiment of the present invention.

[0127] like Figure 5 As shown, a mine tunnel deformation monitoring system includes: a memory 10 and a processor 20, wherein:

[0128] Memory 10, for storing program code;

[0129] The processor 20 is used to read the program code stored in the memory 10, and execute the point cloud data of multiple monitoring areas in the mine tunnel to obtain an overall three-dimensional model of the mine tunnel; based on the overall three-dimensional model, the number of intersections contained in the monitoring area is counted, and the tolerance of the missing area of ​​the monitoring area is determined; the updated point cloud data of the monitoring area is re-collected to obtain the number and missing positions of the missing areas of the monitoring area, and combined with the tolerance, it is judged whether the updated point cloud data of the monitoring area is successfully acquired; if so, the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained from the last monitoring is analyzed to obtain the effectiveness of the updated point cloud data of the monitoring area; based on the effectiveness, the correlation between the updated point cloud data of the monitoring area and the adjacent monitoring areas is analyzed to judge whether the updated point cloud data of the monitoring area is valid; if not, based on the position coordinates of the updated point cloud data of the monitoring area and the point cloud data obtained from the last monitoring, the acquisition position offset of the monitoring area is analyzed, the acquisition position is offset, and new updated point cloud data is re-collected until it is valid; based on the valid updated point cloud data, the overall three-dimensional model is updated.

[0130] Furthermore, the processor 20 includes: a 3D model construction module 21, a tolerance analysis module 22, an updated point cloud data acquisition module 23, an updated point cloud data validity judgment module 24, an updated point cloud data re-acquisition module 25 and a 3D model update module 26. Among them:

[0131] A three-dimensional model building module 21 is used to collect point cloud data of multiple monitoring areas in the mine tunnel to obtain an overall three-dimensional model of the mine tunnel;

[0132] Tolerance analysis module 22, for counting the number of intersections included in the monitoring area based on the overall three-dimensional model and determining the tolerance of missing areas in the monitoring area;

[0133] The updated point cloud data acquisition module 23 is used to re-collect the updated point cloud data of the monitoring area, obtain the number and location of missing areas in the monitoring area, and determine whether the updated point cloud data of the monitoring area is successfully acquired based on the tolerance;

[0134] The updated point cloud data validity judgment module 24 is used to analyze the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the previous monitoring when the updated point cloud data is successfully obtained, so as to obtain the validity of the updated point cloud data of the monitoring area; and based on the validity, analyze the correlation between the updated point cloud data of the monitoring area and the adjacent monitoring areas to determine whether the updated point cloud data of the monitoring area is valid;

[0135] The updated point cloud data re-collection module 25 is configured to analyze the offset of the collection position of the monitoring area based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained in the last monitoring, and offset the collection position when the updated point cloud data is invalid; and re-collect new updated point cloud data after the collection position is offset until it is valid;

[0136] The 3D model updating module 26 is used to complete the update of the entire 3D model based on the valid updated point cloud data.

[0137] 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.

[0138] 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. A mine tunnel deformation monitoring method, characterized in that: The method comprises: Collect point cloud data from multiple monitoring areas in the mine tunnel to obtain the overall three-dimensional model of the mine tunnel; Based on the overall three-dimensional model, counting the number of intersections included in the monitoring area and determining the tolerance of missing areas in the monitoring area; Recollecting updated point cloud data of the monitoring area, obtaining the number and location of missing areas in the monitoring area, and determining whether the updated point cloud data of the monitoring area is successfully acquired based on the tolerance; If yes, analyzing the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area; Based on the effectiveness, analyzing the correlation between the updated point cloud data of the monitoring area and adjacent monitoring areas, and determining whether the updated point cloud data of the monitoring area is effective; If not, then based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained in the last monitoring, analyze the offset of the acquisition position of the monitoring area, offset the acquisition position, and re-acquire new updated point cloud data until it is valid; Based on the valid updated point cloud data, the overall three-dimensional model is updated; Analyzing the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area includes: Calculating the quantitative difference between the updated point cloud data of the monitoring area and the point cloud data obtained during the last monitoring to obtain a quantitative difference degree; Obtaining matching point cloud data and unmatched point cloud data between the updated point cloud data of the monitoring area and the point cloud data obtained during the last monitoring, and analyzing the degree of discreteness of the unmatched point cloud data to obtain a matching degree between the updated point cloud data of the monitoring area and the point cloud data obtained during the last monitoring; Obtaining the validity of the updated point cloud data of the monitoring area according to the degree of difference in quantity and the degree of matching; Analyzing the correlation between the updated point cloud data of the monitoring area and the adjacent monitoring areas, and determining whether the updated point cloud data of the monitoring area is valid, including: Marking the monitoring area that has point cloud data overlapping with the monitoring area as the monitoring area to be matched; Analyze the change relationship between the updated point cloud data of the monitoring area to be matched and the point cloud data obtained from the last monitoring to obtain the effectiveness of the updated point cloud data of the monitoring area to be matched; Analyzing the size correlation between the effectiveness of the monitoring area and the monitoring area to be matched, and determining whether the updated point cloud data of the monitoring area is initially invalid; If not, the updated point cloud data of the monitoring area is valid; If yes, obtaining the number of matching point clouds between the monitoring area and the monitoring area to be matched based on the point cloud data acquired during the measurement of the monitoring area; and obtaining the number of updated matching point clouds between the monitoring area and the monitoring area to be matched based on the updated point cloud data obtained during the measurement of the monitoring area to be matched; Analyzing the difference between the number of the matched point cloud and the number of the updated matched point cloud to determine whether the updated point cloud data of the monitoring area is valid; If yes, the updated point cloud data of the monitoring area is valid; If not, the updated point cloud data of the monitoring area is invalid.

2. The mine tunnel deformation monitoring method according to claim 1, characterized in that: Recollecting updated point cloud data of the monitoring area, obtaining the number and location of missing areas of the monitoring area, and determining whether the updated point cloud data of the monitoring area is successfully acquired in combination with the tolerance, including: Recollecting updated point cloud data of the monitoring area at the last monitoring position; Performing grid processing on the updated point cloud data, and using a connected domain labeling algorithm to obtain the number and positions of missing areas in the monitoring area; Determining whether the number of missing regions is less than the tolerance; If not, the acquisition of updated point cloud data of the monitoring area fails; If yes, determining whether the missing area is at an intersection based on the missing location; If yes, the acquisition of updated point cloud data of the monitoring area fails; If not, the updated point cloud data of the monitoring area is acquired successfully.

3. The mine tunnel deformation monitoring method according to claim 2, characterized in that: If the acquisition of the updated point cloud data of the monitoring area fails, the updated point cloud data of the monitoring area is re-collected.

4. The mine tunnel deformation monitoring method according to claim 1, characterized in that: Analyzing the correlation between the effectiveness of the monitoring area and the monitoring area to be matched, and determining whether the updated point cloud data of the monitoring area is preliminarily invalid, including: Setting a validity threshold, and determining whether the validity levels corresponding to the monitoring area and the monitoring area to be matched are both less than the validity threshold; If yes, the updated point cloud data of the monitoring area is valid; If not, the updated point cloud data of the monitoring area is preliminarily invalid.

5. The mine tunnel deformation monitoring method according to claim 1, characterized in that: Analyzing the difference between the number of the matched point cloud and the number of the updated matched point cloud to determine whether the updated point cloud data of the monitoring area is valid includes: Setting a difference threshold to determine whether the difference between the number of matching point clouds and the number of updated matching point clouds is less than the difference threshold; If yes, the updated point cloud data of the monitoring area is valid; If not, the updated point cloud data of the monitoring area is invalid.

6. The mine tunnel deformation monitoring method according to claim 1, characterized in that: Analyzing the offset of the acquisition position of the monitoring area based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained in the last monitoring, and offsetting the acquisition position, including: Calculate the average value of the position coordinates of the matching point cloud data in the point cloud data of the monitoring area during the last measurement process, and record it as the first feature position; Calculate the mean value of the position coordinates of the updated point cloud data of the monitoring area and record it as the second characteristic position; Calculating the difference between the first characteristic position and the second characteristic position, and combining the preset displacement of the acquisition position to obtain an offset distance; Based on the acquisition position, the acquisition center line is obtained; Recording a feature position corresponding to a maximum value of the number of matching point clouds and the number of updated matching point clouds as an offset feature coordinate; The direction of the offset feature coordinates relative to the current acquisition position is used as the offset direction; The acquisition position is offset on the acquisition center line according to the offset distance and the offset direction.

7. A mine tunnel deformation monitoring system, characterized in that: The system comprises: a memory and a processor, wherein: The memory is used to store program code; The processor is configured to read the program code stored in the memory and execute the method according to any one of claims 1 to 6.

8. The mine tunnel deformation monitoring system according to claim 7, characterized in that: The processor includes: The 3D model building module is used to collect point cloud data from multiple monitoring areas in the mine tunnel to obtain the overall 3D model of the mine tunnel; a tolerance analysis module, configured to count the number of intersections included in the monitoring area based on the overall three-dimensional model and determine the tolerance of missing areas in the monitoring area; An updated point cloud data acquisition module is used to re-collect updated point cloud data of the monitoring area, obtain the number and location of missing areas in the monitoring area, and determine whether the updated point cloud data of the monitoring area is successfully acquired in combination with the tolerance; An updated point cloud data validity judgment module is configured to, when the updated point cloud data is successfully acquired, analyze the change relationship between the updated point cloud data of the monitoring area and the point cloud data obtained in the previous monitoring to obtain the validity of the updated point cloud data of the monitoring area; and based on the validity, analyze the correlation between the updated point cloud data of the monitoring area and the updated point cloud data of adjacent monitoring areas to determine whether the updated point cloud data of the monitoring area is valid; An updated point cloud data re-collection module is configured to, when the updated point cloud data is invalid, analyze the offset of the collection position of the monitoring area based on the updated point cloud data of the monitoring area and the position coordinates of the point cloud data obtained from the last monitoring, and offset the collection position; and re-collect new updated point cloud data after the collection position is offset until it is valid; The three-dimensional model updating module is used to complete the overall three-dimensional model updating based on the valid updated point cloud data.

Images