A method, device, medium and equipment for monitoring displacement of a slope surface

Abstract

The application provides a kind of slope surface displacement monitoring method, device, medium and equipment, comprising: collecting multiple image data of target slope;By analyzing multiple image data, determine multiple target targets and get the area code and pixel coordinates corresponding to the target target;Get the spatial coordinates of the four corner points of each target target in multiple target targets and the centroid coordinates of each target target;Respectively for each target target in multiple target targets, according to the spatial coordinates and centroid coordinates of the target target, the diagonal vector of each target target is calculated, the direction vector of the plane of each target target is obtained, and the direction vector has 3 direction degrees of freedom;The target slope is monitored by pixel coordinates, spatial coordinates, centroid coordinates and direction vector, and the monitoring result is obtained;Compared with the prior art, 3 displacement degrees of freedom + 3 direction degrees of freedom monitoring of the target slope is realized, and the precision of slope monitoring is improved.

Description

Technical Field

[0001] The present invention relates to the field of slope monitoring, and in particular to a method, device, medium and equipment for monitoring slope surface displacement. Background Art

[0002] Slope surface displacement monitoring is a hot topic in slope monitoring research. By monitoring the slope surface displacement, understanding the real-time changes of the slope surface displacement and issuing early warnings for exceeding the limit, the losses caused by slope geological disasters can be reduced or avoided to a certain extent.

[0003] Common methods for monitoring slope surface displacement include geodetic surveying, GPS, and machine vision. Traditional monitoring methods often focus on a single monitoring indicator, resulting in heavy workload, high monitoring costs, and an inability to reflect the overall slope condition, making them difficult to apply to large-scale mountain surveys.

[0004] Existing, commonly used machine vision monitoring methods are mostly based on monocular vision, which makes it difficult to provide accurate three-dimensional slope surface displacement information. Furthermore, most current solutions rely on simple targets with a single marker, which can lead to mismatching in field environments and weak anti-interference capabilities. Furthermore, these solutions only monitor the three displacement degrees of freedom of the marker, making it difficult to monitor all three directional degrees of freedom, resulting in low accuracy in slope surface displacement monitoring. Summary of the Invention

[0005] The present invention provides a slope surface displacement monitoring method, device, medium and equipment, the purpose of which is to solve the defect of low slope surface displacement monitoring accuracy in the above-mentioned prior art.

[0006] In order to achieve the above object, the present invention provides a slope surface displacement monitoring method, comprising:

[0007] Step 1: collecting multiple image data of the target slope, wherein the multiple image data are obtained by collecting images of the target slope from different angles;

[0008] Step 2: Analyze multiple image data to determine multiple target locations and obtain region codes and pixel coordinates corresponding to the target locations;

[0009] Step 3, obtaining the spatial coordinates of the four corner points of each target and the center of mass coordinates of each target;

[0010] Step 4: for each target among the multiple targets, calculate the diagonal vector of the target according to the spatial coordinates and the center of mass coordinates of the target to obtain the direction vector of the target plane, where the direction vector has three directional degrees of freedom;

[0011] Step 5, monitoring the target slope through pixel coordinates, spatial coordinates, centroid coordinates and direction vectors to obtain monitoring results.

[0012] Further, step 2 comprises:

[0013] According to the adaptive threshold segmentation algorithm, the multiple image data are binarized and the contours thereof are extracted to obtain multiple suspected slope displacement monitoring targets;

[0014] The suspected slope displacement monitoring targets are processed one by one, and the region codes thereof are analyzed.

[0015] It is judged whether the region codes of the suspected slope displacement monitoring targets are legal.

[0016] If the region codes of the suspected slope displacement monitoring targets are legal, the suspected slope displacement monitoring targets are determined as target targets.

[0017] The four corner points of the target targets are sub-pixel optimized to obtain the region codes and pixel coordinates corresponding to the target targets.

[0018] Further, step 3 comprises:

[0019] For each of the multiple target targets, a spatial coordinate relationship equation is established and solved through a conversion matrix and pixel coordinates according to the principle of triangulation to obtain the spatial coordinates of the four corner points of the target target, wherein the conversion matrix is calculated according to a camera intrinsic matrix and a camera extrinsic matrix.

[0020] The spatial coordinates of the four corner points of the target target are mean calculated to obtain the centroid coordinates of the multiple target targets.

[0021] Further, step 4 comprises:

[0022] For each of the multiple target targets, the coordinate value of the target target in the total station coordinate system is measured.

[0023] For each of the multiple target targets, the centroid coordinates of the target target in the total station coordinate system and in the camera coordinate system are calculated.

[0024] For each of the multiple target targets, two diagonal vectors of the target target are calculated according to the spatial coordinates of the four corner points of the target target.

[0025] For each of the multiple target targets, the two diagonal vectors of the target target are cross-multiplied to obtain a cross-multiplication result, and the cross-multiplication result is divided by the modulus of the two diagonal vectors to obtain a direction vector of the target target plane, the direction vector having three degrees of freedom.

[0026] Further, the plurality of image data of the target slope is collected by the binocular camera.

[0027] Before step 1, further comprising:

[0028] Calibrating the camera intrinsic matrix, the distortion coefficient and the camera extrinsic matrix of the binocular camera, the camera intrinsic matrix is:

[0029]

[0030] Wherein, K is the camera intrinsic matrix, f x is the focal length in x direction, f y is the focal length in y direction, dX is the physical length corresponding to one pixel in x direction, dY is the physical length corresponding to one pixel in y direction, u0 is the center of the x direction, v0 is the center of the y direction, and θ is the angle deviation generated when the imaging plate is assembled;

[0031] The distortion coefficient is:

[0032]

[0033] Wherein, x', y' represents the normalized image coordinates of real imaging, x, y is the normalized image coordinates of real imaging, r is the distance from the pixel point to the imaging center point, k1, k2, k3 are the radial distortion coefficients of each order, and p1, p2 are the tangential distortion coefficients of each order;

[0034] Further, the camera extrinsic matrix calibration method comprises:

[0035] Converting the image data under the world coordinate system into the camera coordinate system;

[0036] Processing and optimizing the centroid coordinates of the target target under the total station coordinate system and under the camera coordinate system to obtain the rotation matrix from the camera coordinate system to the total station coordinate system;

[0037] Performing centroid transformation on the rotation matrix to obtain the displacement vector;

[0038] Using the rotation matrix and the displacement vector, the image data in the left camera is re-projected onto the imaging plane of the right camera, and the re-projection error is calculated;

[0039] The camera extrinsic matrix of the left camera and the right camera is calculated through the re-projection error.

[0040] Further, after step 5, comprising:

[0041] Cumulatively calculating the monitoring results of each target target to determine whether the slope surface displacement exceeds the pre-set warning value;

[0042] When the displacement of the slope surface exceeds the early warning value, an alarm is given.

[0043] The application further provides a slope surface displacement monitoring device for realizing the slope surface displacement monitoring method.

[0044] The collecting module is configured to collect a plurality of image data of the target slope, the plurality of image data being obtained by image collection of the target slope from different angles.

[0045] The analyzing module is configured to determine a plurality of target markers and obtain region codes and pixel coordinates corresponding to the target markers by analyzing the plurality of image data.

[0046] The first calculating module is configured to obtain spatial coordinates of four corner points of each target marker in the plurality of target markers and a centroid coordinate of each target marker.

[0047] The second calculating module is configured to calculate a diagonal vector of each target marker in the plurality of target markers according to the spatial coordinates and the centroid coordinate of the target marker, and obtain a direction vector of a target marker plane of the target marker, the direction vector having three degrees of freedom.

[0048] The monitoring module is configured to monitor the target slope by using the pixel coordinates, the spatial coordinates, the centroid coordinates and the direction vector, and obtain a monitoring result.

[0049] The application further provides a computer readable storage medium having a computer program stored thereon, when the computer program is executed, the computer program is used to realize the slope surface displacement monitoring method.

[0050] The application further provides a slope surface displacement monitoring device, comprising a memory and a processor.

[0051] The memory is configured to store the computer program and intermediate data during program processing.

[0052] The processor is configured to execute the computer program to realize the slope surface displacement monitoring method.

[0053] The above-mentioned scheme of the application has the following advantages:

[0054] Compared with the prior art, the present application realizes 3 displacement degrees of freedom + 3 direction degrees of freedom monitoring of the target slope by collecting image data of the target slope from different angles, analyzing the image data to determine the target target, calculating the pixel coordinates of the target target, the spatial coordinates of the four corner points, the center of mass coordinates and the direction vector with three direction degrees of freedom to monitor the slope, realizes 3 displacement degrees of freedom + 3 direction degrees of freedom monitoring of the target slope, improves single marker monitoring to multi-marker monitoring, improves two-dimensional monitoring to three-dimensional monitoring, fully reflects the change of the overall and local displacement of the slope in space and time, and improves the precision of the slope monitoring.

[0055] Other benefits of the present application will be described in detail in the subsequent specific embodiment part. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 The flowchart of the present application is shown in the figure;

[0057] Figure 2 The schematic diagram of the target of the embodiment of the present application is shown in the figure;

[0058] Figure 3 The schematic diagram of the target of the embodiment of the present application is shown in the figure;

[0059] Figure 4 The slope surface displacement precision verification diagram of the embodiment of the present application is shown in the figure;

[0060] Figure 5 The three-axis displacement long-time monitoring result diagram of the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0061] In order to make the technical problems, technical solutions and advantages of the present application clearer, specific embodiments will be described in detail below with reference to the drawings and specific embodiments. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0062] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0063] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be a locking connection, or a detachable connection, or an integral connection; it can be a mechanical connection, or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0064] In addition, the technical features involved in the different embodiments of the application described below can be combined with each other as long as there is no conflict.

[0065] The present application aims at the existing problems, and provides a slope surface displacement monitoring method, device, medium and equipment, which monitors the displacement of a certain slope surface in Guizhou through a binocular camera.

[0066] As Figure 1 shown, the embodiment of the present application provides a slope surface displacement monitoring method, comprising:

[0067] Step 1, collecting a plurality of image data of a target slope, the plurality of image data being obtained by a binocular camera for image collection of the target slope from different angles;

[0068] Step 2, determining a plurality of target markers and obtaining the area code and pixel coordinates corresponding to the target markers by analyzing the plurality of image data;

[0069] Step 3, obtaining the spatial coordinates of four corner points of each target marker in the plurality of target markers and the centroid coordinates of each target marker;

[0070] Step 4, for each target marker in the plurality of target markers, calculating the diagonal vector of the target marker according to the spatial coordinates and centroid coordinates of the target marker, obtaining the direction vector of the target marker plane, and the direction vector has three degrees of freedom;

[0071] Step 5, monitoring the target slope through pixel coordinates, spatial coordinates, centroid coordinates and direction vectors to obtain monitoring results.

[0072] Specifically, before step 1, it further includes:

[0073] On the target slope, 10 monitoring points are selected and numbered as 0-9 to erect 10 coded targets, a binocular camera is erected at a reasonable position in front of the target slope, which needs to ensure that the binocular camera can observe the 10 coded targets at the same time; the binocular camera is connected with the control device through the IP interface, the socket method is used for data transmission, the control instruction of the binocular camera is preset in the control device, so that the binocular camera focuses between the 10 monitoring points, thereby realizing the observation of the displacement of the monitoring points.

[0074] Specifically, the camera intrinsic matrix, distortion coefficient and camera extrinsic matrix of the binocular camera need to be calibrated, the camera intrinsic matrix represents the mapping relationship from the real world three-dimensional coordinates to the camera coordinate system, and the camera intrinsic matrix is:

[0075]

[0076] Wherein, K is the camera intrinsic matrix, f x is the focal length in x direction, f y is the focal length in y direction, dX is the physical length corresponding to one pixel in x direction, dY is the physical length corresponding to one pixel in y direction, u0 is the center of x direction, v0 is the center of y direction, and θ is the angle deviation generated when the photosensitive plate is assembled.

[0077] The camera distortion coefficient represents the distortion degree of the real imaging model compared with the ideal imaging model, and the distortion coefficient is:

[0078]

[0079] Wherein, x', y' represents the normalized image coordinates of real imaging, x, y represents the normalized image coordinates of real imaging, r represents the distance from the pixel point to the imaging center point, k1, k2, k3 represents the radial distortion coefficient of each order, p1, p2 represents the tangential distortion coefficient of each order; the distortion coefficient vector is usually represented in the form of D=[k1, k2, p1, p2, k3], and D is the distortion coefficient vector.

[0080] In the embodiment of the present application, the method for solving the camera intrinsic matrix and the camera distortion coefficient adopts Zhang Zhengyou calibration method, which is a camera calibration based on 2D plane target. After obtaining an image of the calibration board, the pixel coordinates (u, v) of each corner point can be obtained by using the corresponding image detection algorithm. The world coordinate system is fixed on the checkerboard, so the physical coordinates of any point on the checkerboard are equal to 0. Since the world coordinate system of the calibration board is artificially defined in advance, the size of each grid on the calibration board is known, and the physical coordinates (x, y, z = 0) of each corner point in the world coordinate system can be calculated.

[0081] The camera calibration is performed by using the information: the pixel coordinates (u, v) of each corner point and the physical coordinates (x, y, z = 0) of each corner point in the world coordinate system, to obtain the camera intrinsic matrix and the distortion parameter.

[0082] The camera extrinsic matrix includes a 3x3 rotation matrix R and a 3x1 displacement vector y, and the camera extrinsic matrix represents the spatial relative pose relationship between the left camera and the right camera of the binocular camera. The camera extrinsic matrix is calibrated based on the method of minimizing the re-projection error, including:

[0083] Taking the P point in the image data as an example, the P point in the world coordinate system is converted into the P point in the camera coordinate system through the conversion matrix, and the specific algorithm is as follows:

[0084] P i =[R i t i ]P w (3)

[0085] Wherein, P w is the coordinate of the P point in the world coordinate system, P i is the coordinate of the P point in the left camera coordinate system, R i is the rotation matrix of the world coordinate system to the left camera coordinate system, and t i is the translation matrix of the world coordinate system to the left camera coordinate system.

[0086] The image data in the left camera is re-projected onto the imaging plane of the right camera by using the rotation matrix R and the displacement vector t, and the specific algorithm is as follows:

[0087]

[0088] Wherein, is the re-projection point of the left camera in the right camera, P l is the coordinate of the P point in the left camera coordinate system, R is the rotation matrix of the left camera coordinate system to the left camera coordinate system, T is the translation matrix of the world coordinate system to the left camera coordinate system, and K ris an intrinsic matrix of the right camera.

[0089] The re-projection error is calculated, and the camera extrinsic matrix of the left camera and the right camera is optimized by minimizing the re-projection error, wherein the calculation formula of the re-projection error is as follows:

[0090]

[0091] Wherein, err is the re-projection error, is a re-projection point of the left camera point in the right camera, p i is an imaging point in the right camera.

[0092] Specifically, step 2 includes:

[0093] According to the adaptive threshold segmentation algorithm, the image data is binarized and the contour is extracted, and a plurality of suspected slope displacement monitoring targets are determined, and the specific steps are as follows:

[0094] Firstly, the adaptive threshold segmentation algorithm is used to process the original image collected on site to obtain a binary image;

[0095] Secondly, the binary image is extracted to obtain a complete closed contour;

[0096] Finally, all the suspected rectangular contour regions are retained, and all the suspected rectangular contour regions are determined as suspected slope displacement monitoring targets;

[0097] The suspected slope displacement monitoring targets are processed one by one, and the region codes thereof are analyzed;

[0098] It is judged whether the region code of the suspected slope displacement monitoring target is legal;

[0099] If the region code of the suspected slope displacement monitoring target is legal, the suspected slope displacement monitoring target is determined as a target target;

[0100] The four corner points of the target target are sub-pixel optimized to obtain the region code and pixel coordinates corresponding to the target target, and the specific steps are as follows:

[0101] Firstly, all the suspected slope displacement monitoring targets are perspective transformed to obtain a square image;

[0102] Secondly, the square image is subjected to Ossu thresholding processing to obtain a separated binary image;

[0103] Then, the binary image is regionally divided to obtain n×n small regions;

[0104] Then, the number of black pixels and white pixels in each small area is counted, and the absolute majority pixel quantity is taken as the encoding of the area, where black is 0 and white is 1, as shown in Figure 2 ;

[0105] Next, it is checked whether the encoding of the area is legal, the obtained area encoding is checked, the area with illegal encoding is discarded, the legal encoding area is retained, and the legal encoding area is determined as the target target;

[0106] Finally, the four corner points of the target target are sub-pixel optimized to obtain the sub-pixel level pixel coordinates (u i ,v i ) of the four corner points and the corresponding encoding.

[0107] Specifically, step 3 includes:

[0108] First, match the same number of target targets of the left camera and the right camera, sample 20 times in succession, take the mean value, reduce the risk of accidental error existing in single sampling, and improve the stability of the monitoring result.

[0109] Since multiple target target images need to be obtained through the angle of view of the camera, the camera coordinate system is unstable, and the camera coordinate system is calibrated by obtaining the coordinates of the target target in the world coordinate system. Therefore, according to the camera intrinsic matrix (K, D) and the extrinsic matrix (R, t), the conversion matrix M of the binocular camera world coordinate system to the camera pixel coordinate is calculated, and the conversion matrix M represents the conversion relationship from the world coordinate system to the pixel coordinate system. The conversion expression is as follows:

[0110]

[0111] Where X, Y, Z are the spatial coordinates of the target target in the world coordinate system, u i , v i are the pixel coordinates of the target target in the left camera, and M is the conversion matrix from the world coordinate system to the left camera pixel coordinate.

[0112] The conversion matrix M is calculated through the camera intrinsic matrix (K, D) and the extrinsic matrix (R, t), and the calculation formula is as follows:

[0113]

[0114] Where f x is the x-direction focal length, f y is the y-direction focal length, x0 is the x-direction photosensitive center coordinate, y0 is the y-direction photosensitive center coordinate, s is the error term caused by the assembly deviation of the photosensitive plate, R ij is the element of the i-th row and j-th column of the rotation matrix, and T 0iis the i-th element of the displacement vector, M is the conversion matrix from the world coordinate system coordinates to the camera pixel coordinates, m ij is the i-th row j-th column element of the conversion matrix.

[0115] Then, for each target target in the plurality of target targets, a spatial coordinate relationship equation is established according to the principle of triangulation as shown in formula (1) and is solved by the conversion matrix M and the pixel coordinates to obtain the spatial coordinates of the four corner points of the target target. Figure 3

[0116] Specifically, the spatial coordinate relationship equation is as follows:

[0117]

[0118] wherein, is the i-th row j-th column element of the k-th conversion matrix, X, Y, Z are the spatial coordinates of the target target in the world coordinate system, u1, v1 are the pixel coordinates of the target target in the left camera, and u2, v2 are the pixel coordinates of the target target in the right camera.

[0119] Finally, the mean value of the spatial coordinates of the four corner points of the target target is calculated to obtain the centroid coordinates of the plurality of target targets.

[0120] Specifically, step 4 comprises:

[0121] For each target target in the plurality of target targets, the conversion matrix after the camera coordinate system is converted into the total station coordinate system is calculated by using the pixel coordinates of the target target in the camera coordinate system and the coordinates of the target target measured by the total station, and the specific calculation is as follows:

[0122] First, the coordinate values of the target target at different positions in the total station coordinate system are measured;

[0123] The centroid coordinates of the target target in the total station coordinate system and in the camera coordinate system are calculated, and the centroid operation is performed on the centroid coordinates of the two groups of point clouds to obtain the centroid-removed point cloud coordinate values, and the calculation formula is as follows:

[0124]

[0125] Secondly, the least square method is used to optimize the mean square error function of the two groups of point clouds, and the optimization result is taken as the rotation matrix R * from the camera coordinate system to the total station coordinate system, and the specific optimization equation is as follows:

[0126]

[0127] According to the rotation matrix R * , the displacement vector t is obtained by centroid transformation: ​

[0128] t = p 1 -Rp 2 (12)

[0129] wherein p 1 , p 2 is the centroid coordinate of the first and second group of point clouds, is the spatial coordinate of the i-th point of the first and second group of point clouds, R is the rotation matrix from the left camera coordinate system to the left camera coordinate system, is the de-centroid spatial coordinate of the i-th point of the first and second group of point clouds.

[0130] Using the rotation matrix R * , the coordinates of the target target in the camera coordinate system are converted into the coordinates in the total station coordinate system.

[0131] Finally, according to the spatial coordinates of the four corner points of the plurality of target targets, the two diagonal vectors of each target target are calculated as:

[0132] n1 = p1-p3 (13)

[0133] n2 = p2-p4 (14)

[0134] wherein n1, n2 are diagonal vectors, p1 is the first corner point, p2 is the second corner point, p3 is the third corner point, and p4 is the fourth corner point.

[0135] For each target target in the plurality of target targets, the two diagonal vectors of the target target are cross-multiplied and divided by the modulus of the two diagonal vectors to obtain the direction vector of the target target plane, and the direction vector has three direction degrees of freedom, and the calculation formula is as follows:

[0136]

[0137] wherein n is the direction vector, and n1, n2 are the diagonal vectors.

[0138] Specifically, step 5 comprises: obtaining the 6 degrees of freedom monitoring data results of 3 displacement directions + 3 inclination directions by obtaining the pixel coordinates of the target target, the spatial coordinates of the four corner points, the centroid coordinates, and combining the direction vector of the target target.

[0139] The horizontal direction displacement accuracy obtained by the above method is verified as follows Figure 4 (a) shown, the vertical direction displacement accuracy is verified as follows Figure 4 (b) shown, the z displacement long-time monitoring result obtained in the example is as Figure 5 (a) shown, the x displacement long-time monitoring result is as Figure 5 (b) shown, the y displacement long-time monitoring result is as Figure 5(c) shown.

[0140] Specifically, after step 5, include:

[0141] The monitoring results of multiple target targets are transmitted to the back-end server in real time, and an early warning value is set in the back-end server. The 6-DOF monitoring results of 10 monitoring points are recorded by the back-end server. The monitoring results of each target target are used to accumulate the surface displacement of the slope to determine whether the surface displacement of the slope exceeds the early warning value. When the surface displacement of the slope exceeds the early warning value, an alarm is issued.

[0142] An embodiment of the present invention further provides a slope surface displacement monitoring device, which is used to implement the above-mentioned slope surface displacement monitoring method, including:

[0143] An acquisition module is used to acquire multiple image data of the target slope, wherein the multiple image data are obtained by acquiring images of the target slope from different angles;

[0144] An analysis module is used to determine multiple target locations by analyzing multiple image data and obtain region codes and pixel coordinates corresponding to the target locations;

[0145] A first calculation module is used to obtain the spatial coordinates of four corner points of each target and the coordinates of the center of mass of each target;

[0146] The second calculation module is used to calculate the diagonal vector of each target target among the multiple targets according to the spatial coordinates and the center of mass coordinates of the target target, and obtain the direction vector of the target target plane, where the direction vector has three directional degrees of freedom;

[0147] The monitoring module is used to monitor the target slope through pixel coordinates, space coordinates, centroid coordinates and direction vectors to obtain monitoring results.

[0148] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed, it is used to implement the above-mentioned slope surface displacement monitoring method.

[0149] An embodiment of the present invention further provides a slope surface displacement monitoring device, comprising a memory and a processor;

[0150] The memory is used to store computer programs and intermediate data during program processing;

[0151] The processor is used to execute the computer program to implement the above-mentioned slope surface displacement monitoring method.

[0152] Compared with the prior art, the embodiment of the present application realizes 3 displacement degrees of freedom + 3 direction degrees of freedom monitoring of the target slope by collecting image data of the target slope from different angles, analyzing the image data to determine the target, calculating the pixel coordinates of the target, the spatial coordinates of the four corner points, the center of mass coordinates and the direction vector with three direction degrees of freedom to monitor the slope, improves single-mark monitoring to multi-mark monitoring, improves two-dimensional monitoring to three-dimensional monitoring, fully reflects the changes of the overall and local displacement of the slope in space and time, improves the precision of the slope monitoring, and plays the progress of the machine vision algorithm, reduces the instrument cost and labor cost of per square meter of the slope surface displacement monitoring, and can realize the monitoring of the slope surface displacement of a larger area at a lower cost.

[0153] The above is the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should be considered as the protection scope of the present application.

Claims

1. A method for monitoring slope surface displacement, characterized in that: include: Step 1: collecting multiple image data of the target slope, wherein the multiple image data are obtained by collecting images of the target slope from different angles; Step 2, determining multiple target objects by analyzing the multiple image data and obtaining region codes and pixel coordinates corresponding to the target objects; Step 3, obtaining the spatial coordinates of the four corner points of each of the multiple target targets and the coordinates of the center of mass of each target target; Step 4, for each of the plurality of target targets, calculating the diagonal vector of the target target according to the spatial coordinates and the center of mass coordinates of the target target, to obtain a direction vector of each target target plane, wherein the direction vector has three directional degrees of freedom; Step 5: monitoring the target slope using the pixel coordinates, the spatial coordinates, the centroid coordinates, and the direction vector to obtain a monitoring result; Wherein, the step 3 comprises: For each of the plurality of target targets, a spatial coordinate relationship equation is established and solved using a transformation matrix and the pixel coordinates according to the triangulation principle to obtain the spatial coordinates of the four corner points of the target target, wherein the transformation matrix is ​​calculated based on the camera intrinsic parameter matrix and the camera extrinsic parameter matrix; Calculating the mean of the spatial coordinates of the four corner points of the target to obtain the center of mass coordinates of the target; The step 4 comprises: For each of the plurality of target targets, respectively, measuring the coordinate value of the target target in the total station coordinate system; For each of the plurality of target targets, respectively, calculating the centroid coordinates of the target target in the total station coordinate system and the camera coordinate system; For each of the plurality of target targets, respectively, calculate two diagonal vectors of the target target according to the spatial coordinates of the four corner points of the target target; For each of the multiple target targets, cross product is performed on the two diagonal vectors of the target target to obtain a cross product result, and the cross product result is divided by the modulus of the two diagonal vectors to obtain a direction vector of the target target plane, where the direction vector has three directional degrees of freedom.

2. The slope surface displacement monitoring method according to claim 1, characterized in that: The step 2 includes: Binarizing the plurality of image data and extracting their contours according to an adaptive threshold segmentation algorithm to obtain a plurality of suspected slope displacement monitoring targets; Processing the suspected slope displacement monitoring targets one by one and analyzing their regional codes; Determining whether the area coding of the suspected slope displacement monitoring target is legal; If the area code of the suspected slope displacement monitoring target is legal, determining the suspected slope displacement monitoring target as the target target; Sub-pixel optimization is performed on the four corner points of the target to obtain the region code and pixel coordinates corresponding to the target.

3. The slope surface displacement monitoring method according to claim 1, characterized in that: The plurality of image data of the target slope are collected by a binocular camera; Before step 1, the method further includes: The camera intrinsic parameter matrix, distortion coefficient and camera extrinsic parameter matrix of the binocular camera are calibrated. The camera intrinsic parameter matrix is: ； in, is the camera intrinsic parameter matrix, for Directional focal length, for Directional focal length, for The physical length of one pixel in the direction, for The physical length of one pixel in the direction, for The coordinates of the photosensitive center in the direction, for The coordinates of the photosensitive center in the direction, Angular deviation caused when assembling the photographic plate; The distortion coefficient is: ； in, , Expressed as the normalized image coordinates of real imaging, is the normalized image coordinate of real imaging, is the distance from the pixel to the imaging center, , , are the radial distortion coefficients of each order, , are the tangential distortion coefficients of each order.

4. The slope surface displacement monitoring method according to claim 3, characterized in that: The camera extrinsic parameter matrix calibration method includes: Convert the image data in the world coordinate system to the camera coordinate system; Processing and optimizing the center of mass coordinates of the target in the total station coordinate system and the camera coordinate system to obtain a rotation matrix from the camera coordinate system to the total station coordinate system; Performing a centroid transformation on the rotation matrix to obtain a displacement vector; Reprojecting the image data from the left camera onto the imaging plane of the right camera using the rotation matrix and the displacement vector, and calculating the reprojection error; The camera extrinsic parameter matrices of the left camera and the right camera are calculated using the reprojection error.

5. The slope surface displacement monitoring method according to claim 4, characterized in that: After step 5, the method includes: Performing cumulative calculation of the slope surface displacement based on the monitoring results of each target, and determining whether the slope surface displacement exceeds a preset warning value; When the surface displacement of the slope exceeds the warning value, an alarm is issued.

6. A slope surface displacement monitoring device, used to implement the slope surface displacement monitoring method according to any one of claims 1 to 5, characterized in that: include: An acquisition module is used to acquire a plurality of image data of the target slope, wherein the plurality of image data are acquired by acquiring images of the target slope from different angles; An analysis module, configured to determine a plurality of target objects by analyzing the plurality of image data and obtain region codes and pixel coordinates corresponding to the target objects; A first calculation module is used to obtain the spatial coordinates of four corner points of each of the multiple target targets and the coordinates of the center of mass of each target target; a second calculation module, configured to calculate, for each of the plurality of target targets, a diagonal vector of the target target according to the spatial coordinates and the center of mass coordinates of the target target, to obtain a direction vector of the target target plane, wherein the direction vector has three directional degrees of freedom; A monitoring module, configured to monitor the target slope using the pixel coordinates, the spatial coordinates, the centroid coordinates, and the direction vector to obtain a monitoring result; The first calculation module is specifically configured to implement: For each of the plurality of target targets, a spatial coordinate relationship equation is established and solved using a transformation matrix and the pixel coordinates according to the triangulation principle to obtain the spatial coordinates of the four corner points of the target target, wherein the transformation matrix is ​​calculated based on the camera intrinsic parameter matrix and the camera extrinsic parameter matrix; Calculating the mean of the spatial coordinates of the four corner points of the target to obtain the center of mass coordinates of the target; The second calculation module is specifically used to implement: For each of the plurality of target targets, respectively, measuring the coordinate value of the target target in the total station coordinate system; For each of the plurality of target targets, respectively, calculating the centroid coordinates of the target target in the total station coordinate system and the camera coordinate system; For each of the plurality of target targets, respectively, calculate two diagonal vectors of the target target according to the spatial coordinates of the four corner points of the target target; For each of the multiple target targets, cross product is performed on the two diagonal vectors of the target target to obtain a cross product result, and the cross product result is divided by the modulus of the two diagonal vectors to obtain a direction vector of the target target plane, where the direction vector has three directional degrees of freedom.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the slope surface displacement monitoring method according to any one of claims 1 to 5 is implemented.

8. A slope surface displacement monitoring device, characterized in that: Including memory and processor; The memory is used to store computer programs and intermediate data during program processing; The processor is used to execute the computer program to implement the slope surface displacement monitoring method according to any one of claims 1 to 5.

Images