Static load stack settlement measurement method based on machine vision and intelligent terminal thereof

Abstract

The present application relates to pile foundation detection technical field, especially to a kind of static load heaped load settlement measurement method and its intelligent terminal based on machine vision.The method comprises the following steps: S1, control hydraulic loading device to set load Q;S2, according to the set time interval, control each camera to collect the image of corresponding marker simultaneously;S3, each image collected is processed, and image information is obtained, and the settlement change of each marker is obtained by processing image information through camera mathematical model;S4, according to the settlement change of each marker, the settlement deformation of pile under different time points under the load Q is calculated by interpolation;S5, repeat step S1 to step S4 until termination loading condition is reached, the settlement deformation of pile under each time node under each load Q is derived, and the final settlement under different load Q Q-s curve is formed.The present application does not need to build reference beam, improves the measurement accuracy, and can realize continuous measurement of settlement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pile foundation detection, and particularly relates to a static load heaped settlement measurement method based on machine vision and an intelligent terminal thereof. BACKGROUND

[0002] At present, in the construction process of various projects, the static load heaped test is an indispensable test link for evaluating the bearing capacity of the foundation, that is, by gradually applying vertical pressure on the top of the pile, the settlement of the top of the pile over time is observed to determine the vertical compressive ultimate bearing capacity of the pile. In the test process, the slow maintenance load Q method is usually used, and the specific method is to divide the load Q into several levels and add it to the pile according to certain requirements. Before the pile subsides to a certain relative stability standard, the load Q of this level is maintained unchanged. When the stability standard is reached, the next level of load Q is added. When the specified test termination condition is reached, the loading is terminated. Then, the load is unloaded to zero in stages. The test needs to measure the settlement deformation of the pile foundation under different loads Q, and then analyze the ultimate bearing capacity of the foundation. At present, the measurement of foundation settlement mainly relies on manual reading of dial gauges at fixed time intervals. The manual demand is large, and the staff needs to handle multiple tasks at the same time. This not only requires a large amount of manual work, but also easily reduces the detection efficiency due to distraction, thereby affecting the accuracy of the measurement. As the pile load Q continues to load, the pressure on the support platform gradually increases, and manual reading becomes more and more dangerous. If the pile quality is not good, the support platform will collapse, causing casualties. While some digital dial gauges are used with automatic acquisition modules, a reference beam still needs to be built, and the process is complicated. SUMMARY

[0003] The technical problem to be solved by the present application is that the prior art needs to build a reference beam, the process is complicated, and the measurement process relies on a large amount of manual operation, which requires a large amount of manual work and has low measurement efficiency and poor precision. The present application provides a static load heaped settlement measurement method based on machine vision, which does not need to build a reference beam, improves the measurement accuracy, and can realize continuous measurement of the settlement.

[0004] The technical solution adopted by the present application to solve the technical problem is: a static load heaped settlement measurement method based on machine vision, a static load heaped test platform is installed on the pile, and the static load heaped test platform comprises:

[0005] A support frame is arranged directly above the pile, and a heaped object is arranged above the support frame.

[0006] A hydraulic loading device is arranged on the top of the pile and passes through the gap of the support frame.

[0007] A lower pad is arranged at the bottom of the hydraulic loading device, and a marker is arranged at each of the four corners of the top end of the lower pad.

[0008] An upper pad is set on the top of the hydraulic loading device. The center of the upper pad coincides with the center of the lower pad. A camera is provided at each of the four corners of the top of the upper pad, and each camera corresponds to a marker.

[0009] The method includes the following steps:

[0010] S1, control the hydraulic loading device to the set load Q;

[0011] S2 controls each camera to simultaneously acquire images of the corresponding landmark according to a set time interval;

[0012] S3, perform image processing on each acquired image to obtain image information, process the image information through the camera mathematical model, and obtain the sedimentation change of each marker;

[0013] S4. Based on the settlement change of each marker, the settlement deformation of the foundation pile at different time points under the load Q is calculated by interpolation.

[0014] S5. Repeat steps S1 to S4 until the loading termination condition is met. Derive the settlement deformation of the foundation piles at each time node under each load Q and form the settlement curve Qs under different loads Q.

[0015] Furthermore, specifically, image processing is performed on each image to obtain image information, including:

[0016] 3.1 Obtain the marker template image T;

[0017] 3.2. Perform grayscale conversion on the acquired image to obtain a grayscale image containing the marker. I G ;

[0018] 3.3, the grayscale image I G Gaussian blurring is performed to obtain the image to be matched. I ;

[0019] 3.4, The image to be matched is based on a multi-resolution hierarchical matching method. I Perform template matching with the marker template image T to obtain the matching position of the marker. x, y );

[0020] 3.5, based on the matching position of the marker ( x, y Using the template image T as a base, expand the image by N pixels in four directions to the image to be matched. I Cropping is performed to obtain the ROI image containing the marker. I roi ;

[0021] 3.6, the ROI image I roi The feature point recognition is performed to obtain pixel coordinates of upper, lower, left and right four end points of the marker;

[0022] 3.7, the marker image information is calculated based on the pixel coordinates of the upper, lower, left and right four end points of the marker l pix , w pix , l pix The pixel length of the marker on the image is w pix The pixel width of the marker on the image is

[0023] Further, specifically, the settlement change amount of each marker is obtained by processing the image information through a camera mathematical model, which specifically includes:

[0024] The actual parameters of the marker are obtained l , w and The marker image information l pix , w pix ;

[0025] The actual parameters of the marker are obtained l , w and The marker image information l pix , w pix The settlement amount of each side of the marker is calculated ;

[0026] The settlement amount of each side of the marker is calculated The settlement change amount of the marker is calculated by a bilateral interpolation method .

[0027] Further, specifically, in the step S4, a correction algorithm is further provided to correct the settlement deformation amount of the foundation pile.

[0028] Further, specifically, in the step S4, the correction algorithm specifically includes:

[0029] The settlement amount of each side of the marker is calculated The settlement gradient of the center point of the marker is calculated The calculation formula is:

[0030]

[0031] Among them, The settlement amount of each side of the marker is calculated, respectively ;

[0032] According to the center point of the marker 9, the settlement gradient is determined The settlement correction value is calculated , and the calculation formula is:

[0033]

[0034] Wherein, K is the warping distance of the bottom plate;

[0035] The corrected settlement deformation of the foundation pile is calculated, and the calculation formula is:

[0036] .

[0037] Further, specifically, the template matching between the to-be-matched image I and the marker (9) template image T based on the multi-resolution hierarchical matching method specifically includes the following steps:

[0038] S3.41, a multi-resolution image sequence of the to-be-matched image I and the marker template image T is respectively created, the multi-resolution image sequence of the to-be-matched image I is I 0, I 1, I 2, … , I n , and the multi-resolution image sequence of the template image T is T 0, T 1, T 2, … , T n ; wherein the first layer of the multi-resolution image sequence is the image with the lowest resolution, and the n-th layer is the original high-resolution image;

[0039] S3.42, template matching is performed between the first layer I 0 of the multi-resolution sequence of the to-be-matched image I and the first layer of the multi-resolution sequence of the template image, to obtain a first matching region;

[0040] S3.43, the matching region of the previous step is projected onto the next layer of the to-be-matched image, and the next layer of the multi-resolution sequence of the template image T1 is matched in the projection region of the to-be-matched image, to obtain a new matching region;

[0041] S3.44, step S3.43 is repeated until the to-be-matched image Ithe end of the calculation of the n-th layer of the multi-resolution sequence of the to-be-matched image I the best matching position found in the n-th layer of the multi-resolution sequence of the to-be-matched image x t ,y t the best matching position found in the n-th layer of the multi-resolution sequence of the to-be-matched image x t ,y t the matching position of the marker x, y .

[0042] Further, specifically, the termination loading condition comprises:

[0043] (1) under the current level load Q, when the settlement amount of the top of the pile is greater than 5 times the settlement amount under the previous level load Q, and the total settlement amount of the top of the pile exceeds 40mm;

[0044] (2) under the current level load Q, the settlement amount of the top of the pile is greater than 2 times the settlement amount under the previous level load Q, and has not reached the relative stability standard for 24h;

[0045] (3) the maximum load Q has been reached and the settlement of the top of the pile reaches the stability standard.

[0046] Further, specifically, the settlement amount of each side of the marker The calculation formula is:

[0047]

[0048] Wherein, is the settlement amount of the length direction side, is the settlement amount of the width direction side.

[0049] Further, specifically, the camera and the corresponding marker are relatively arranged.

[0050] An intelligent terminal comprising a memory and a processor, the memory storing a computer program capable of being loaded and executed by the processor to perform the machine vision-based static load pile-up settlement measurement method as described above.

[0051] The machine vision-based static load pile-up settlement measurement method has the advantages that the four-direction settlement of the bottom plate is obtained by processing and calculating the continuous images of the four directions of the bottom plate, and the pile settlement deformation is calculated, the reference beam is not needed to be built, the arrangement is convenient, the automatic high-precision settlement measurement of the static load test is realized, the measurement precision is high, and the continuous measurement of the settlement can be realized. BRIEF DESCRIPTION OF DRAWINGS

[0052] The application will be further described below with reference to the accompanying drawings and examples.

[0053] Figure 1 is a structural schematic diagram of a static load stacking detection test platform according to the most preferred embodiment of the application.

[0054] Figure 2 is a settlement measurement flowchart of a static load stacking detection test platform according to the most preferred embodiment of the application.

[0055] Figure 3 is a structural schematic diagram of a lower base plate according to the most preferred embodiment of the application.

[0056] Figure 4 is a structural schematic diagram of a camera according to the most preferred embodiment of the application. Figure 3 is an enlarged schematic diagram of position H in the structural schematic diagram of the camera according to the most preferred embodiment of the application.

[0057] Figure 5 is a settlement amount calculation schematic diagram of a length direction edge according to the most preferred embodiment of the application, (a) is a geometric schematic diagram of the position relationship between a camera and a length direction edge of a marker before testing; (b) is a geometric schematic diagram of the position relationship between the camera and the length direction edge of the marker during settlement.

[0058] Figure 6 is a settlement amount calculation schematic diagram of a width direction edge according to the most preferred embodiment of the application, (a) is a geometric schematic diagram of the position relationship between a camera and a width direction edge of a marker before testing; (b) is a geometric schematic diagram of the position relationship between the camera and the width direction edge of the marker during settlement.

[0059] Figure 7 is a geometric calculation schematic diagram of a warping distance of a lower base plate according to the most preferred embodiment of the application.

[0060] Figure 8 is a geometric calculation schematic diagram of warping of a lower base plate according to the most preferred embodiment of the application.

[0061] Figure 9 is an enlarged schematic diagram of position R in the structural schematic diagram of the camera according to the most preferred embodiment of the application. Figure 8

[0062] Fig. 1, pile; 2, support frame; 3, stacking object; 4, hydraulic loading device; 6, upper base plate; 7, camera; 8, lower base plate; 9, marker. DETAILED DESCRIPTION

[0063] The application will now be further described in detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams and only schematically illustrate the basic structure of the application, and thus only show the components related to the application.

[0064] ​In the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like are 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 therefore cannot be understood as indicating or implying 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 features defined as "first" and "second" can be explicitly or implicitly included one or more features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specified.

[0065] In the description of the present application, it should be noted that, unless otherwise specified and limited, the terms "mounting", "connection", "connection" should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication between the two elements inside. 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.

[0066] Embodiment 1

[0067] The embodiment of the present application provides a static load stacking settlement measurement method based on machine vision, as shown in the figure, first, a static load stacking test platform is installed on the foundation pile 1, wherein the static load stacking test platform comprises: Figures 1-4

[0068] The support frame 2 is arranged directly above the foundation pile 1, and the upper portion of the support frame 2 is provided with a stacking object 3;

[0069] The hydraulic loading device 4 is arranged on the top of the foundation pile 1 and passes through the gap of the support frame 2;

[0070] The lower pad plate 8 is arranged at the bottom of the hydraulic loading device 4, and the top end of the lower pad plate 8 is provided with a marker 9 at each of the four corners;

[0071] The upper pad plate 6 is arranged at the top of the hydraulic loading device 4, the center of the upper pad plate 6 coincides with the center of the lower pad plate 8, and the upper pad plate 6 is provided with a camera 7 at each of the four corners of the top end. Each camera 7 corresponds to a marker 9. In order to facilitate image acquisition of the camera 7, the camera 7 and the corresponding marker are arranged oppositely.

[0072] The slow maintenance load Q method is used to control the working of the static load stacking test platform, and the static load stacking settlement is measured, the method comprising the following steps:​

[0073] S1, control the hydraulic loading device 4 to the set load Q.

[0074] S2, according to the set time interval, control each camera 7 to collect the image of the corresponding marker 9 at the same time; the set time includes:

[0075] 1. After each level of load Q is applied, measure the settlement every 5, 15, 30, 45, and 60 minutes, and then measure the settlement every 30 minutes.

[0076] 2. When the settlement of the pile 1 is relatively stable, the settlement in each hour does not exceed 0.1 mm, and appears twice continuously (from the 30th minute after the application of the graded load Q, the settlement observation value is calculated every 30 minutes for 1.5 hours continuously three times).

[0077] 3. When unloading, each level of load Q is maintained for 1 hour, and the settlement is measured at 15, 30, and 60 minutes.

[0078] S3, image processing is performed on each collected image to obtain image information, and the image information is processed through a camera mathematical model to obtain the settlement change of each marker 9.

[0079] S4, according to the settlement change of each marker 9, the settlement deformation of the pile at different time points under the load Q is calculated by interpolation.

[0080] S5, repeat steps S1 to S4 until the termination loading condition is reached, derive the settlement deformation of the pile at each time node under each level of load Q, and form the settlement Q-s curve under different loads Q.

[0081] It should be noted that the termination loading condition includes:

[0082] (1) Under the current level of load Q, the settlement of the top of the pile 1 is greater than 5 times the settlement under the previous level of load Q, and the total settlement of the top of the pile 1 exceeds 40 mm.

[0083] (2) Under the current level of load Q, the settlement of the top of the pile 1 is greater than 2 times the settlement under the previous level of load Q, and has not reached the relative stability standard after 24 hours.

[0084] (3) The maximum load Q has been reached and the settlement of the top of the pile 1 has reached the stability standard.

[0085] In the embodiment, the image processing of each image to obtain image information specifically includes:

[0086] 3.1, obtaining a marker 9 template image T.

[0087] 3.2, the collected image is converted to grayscale to obtain a grayscale image containing the marker 9 I G .

[0088] 3.3, the grayscale image I G is subjected to Gaussian blur processing to obtain a to-be-matched image I .

[0089] 3.4, the to-be-matched image I and the marker 9 template image T are template matched based on a multi-resolution hierarchical matching method to obtain the matching position of the marker 9 x, y ); through the multi-resolution hierarchical matching method, the calculation amount is greatly reduced from low resolution to high resolution, and the scale change on the image caused by the change of the marker 9 can be better adapted, and further, the specific steps include:

[0090] S3.41, a multi-resolution image sequence of the to-be-matched image I and the marker 9 template image T is created respectively, the multi-resolution image sequence of the to-be-matched image I is I 0, I 1, I 2, … , I n , and the multi-resolution image sequence of the template image T is T 0, T 1, T 2, … , T n ; wherein the first layer of the multi-resolution image sequence is the image with the lowest resolution, n the last layer is the original high-resolution image; further, the to-be-matched image I and the marker 9 template image T are processed through continuous downsampling, for example, the image size is reduced by half every layer, to obtain the multi-resolution image sequence of the to-be-matched image I and the marker 9 template image T.

[0091] S3.42, the first layer I 0 of the multi-resolution sequence of the to-be-matched image I and the first layer of the multi-resolution sequence of the template image are template matched to obtain the first matching area.

[0092] S3.43, project the matching region of the last step to the next layer to-be-matched image, and match in the projected region of the next layer template image T1 on the to-be-matched image to obtain a new matching region; it should be noted that due to the resolution being doubled, this matching position will correspond to the projected region of the next layer. In this projected region instead of the entire image, more accurate template matching is performed again to refine the matching result layer by layer. For example, if the best match is found at the (x, y) of the nth layer, then in the nth 1 layer, only the projected region centered at (2x, 2y) needs to be searched instead of the entire image.

[0093] S3.44, repeat step S3.43 until the calculation of the nth layer of the multi-resolution sequence of the to-be-matched image I is completed, and finally the best matching position found in the nth layer of the multi-resolution sequence of the to-be-matched image I is the matching position of the marker 9 x t ,y t . x t , y t . x, y .

[0094] 3.5, based on the matching position of the marker 9 x, y and the marker 9 template image T, expand N pixels in four directions to the to-be-matched image I , crop the to-be-matched image I to obtain a ROI image I containing the marker 9 roi .

[0095] 3.6, perform feature point recognition on the ROI image I roi to obtain the pixel coordinates of the upper, lower, left and right four end points of the marker 9;

[0096] 3.7, calculate the image information of the marker 9 l pix , w pix , l pix as the pixel length of the marker 9 on the image w pix as the pixel width of the marker 9 on the image.

[0097] In the embodiment, the four end points of the marker 9, i.e. the upper, lower, left and right end points, are A, B, C and D respectively. The settlement change amount of each marker 9 is obtained by processing the image information through the camera mathematical model, and specifically includes:

[0098] The actual parameters of the marker 9 are obtained l , w and The image information of the marker 9 is obtained l pix , w pix ;

[0099] As shown in Figures 5-6 , the camera 7 parameters are known, and the actual parameters of the marker 9 l , w and the image information of the marker 9 l pix , w pix are used to calculate the AD side settlement amount and the CD side settlement amount respectively, and then the settlement amount of each side of the marker 9 is calculated , and the calculation formula is:

[0100] Specifically, the AD side settlement amount , i.e. the settlement amount of the length direction side, is known, as shown in Figure 5 (a), the distance Z from the camera 7 to the marker 9 before measurement is known, and can be obtained according to , wherein is the focal length of the camera, p is the conversion factor, is the length pixel of the marker 9 in the image before measurement, .

[0101] As shown in Figure 5 (b), the distance from the camera 7 to the marker 9 after settlement measurement is known, and the settlement amount of the length direction side is calculated according to the calculation formula:

[0102] .

[0103] The CD side settlement amount , i.e. the settlement amount of the width direction side, is known, as shown in Figure 6 (a), and can be obtained according to , wherein is the width pixel of the marker 9 in the image before measurement, .

[0104] As shown in Figure 6(b) the distance of the camera 7 to the marker 9 after the settlement measurement , the settlement of the lengthwise edge The calculation formula is:

[0105] .

[0106] According to the settlement of the four edges of the marker 9 , the settlement change of the marker 9 is calculated by the double edge interpolation method .

[0107] In the embodiment, since the mat plate will produce a small amount of warping under the action of force, the marker will also be warped and deformed on the warped and deformed mat plate, in order to avoid a slight difference between the measured settlement deformation and the actual deformation of the center of the pile 1, a correction algorithm is further provided in step S4 to correct the settlement deformation of the pile.

[0108] In the present embodiment, the warping plane is characterized by a single curved surface formed by four non-coplanar points through a hyperbolic paraboloid.

[0109] In the canonical coordinate system of analytic geometry (the vertex is at the origin of the coordinate origin), the equation expression of the hyperbolic paraboloid is:

[0110] ;

[0111] In actual measurement of the warping of an object, the general form expression containing cross terms is:

[0112] ;

[0113] If there are four points: , the coefficients can be quickly derived as:

[0114] ;

[0115] ;

[0116] ;

[0117] .

[0118] The xy coordinates of the four points are converted into a normalized coordinate system, which specifically includes:

[0119] 1. Set the coordinate system and boundary conditions, first, define a rectangular region containing the four corner points, and establish a local Cartesian coordinate system and a normalized parameter coordinate system .

[0120] Let the four points be in The projections on the plane form a rectangle, and their settlement amounts are The mapping relationship between the local Cartesian coordinate system and the normalized parameter coordinate system of the four points is shown in the following table:

[0121] Table 1 Mapping relationship table between local Cartesian coordinate system and normalized parameter coordinate system

[0122]

[0123] It should be noted that, , .

[0124] 2. Establish the coordinate mapping relationship, for simplicity, use the normalized coordinates , map to [0, 1], and map to [0, 1].

[0125] According to , the is obtained;

[0126] According to , the is obtained.

[0127] 3. Use the standard equation to solve the coefficient , four points and their corresponding settlement are substituted into the general form expression:

[0128] ;

[0129] Point A is substituted to obtain:

[0130] ;

[0131] Point B is substituted to obtain:

[0132]

[0133] The is obtained;

[0134] Point C is substituted to obtain:

[0135]

[0136] The is obtained;

[0137] Point D is substituted to obtain:

[0138] ;

[0139] Substitute:

[0140]

[0141]

[0142] .

[0143] 4. From the equation to the equation, the expression of and the coordinate mapping relationship Substitute back to the original standard equation:

[0144]

[0145] .

[0146] 5. Organize into normalized coordinate bilinear interpolation form, recombine the expansion, and organize the bilinear interpolation form based on normalized coordinates .

[0147] According to , the coefficients are extracted respectively:

[0148] ;

[0149] All terms are aggregated:

[0150] ;

[0151] Using factorization:

[0152] ;

[0153] ;

[0154] ;

[0155] .

[0156] The ABCD bilinear surface parameter equation of the marker 9 in the normalized coordinate is:

[0157] .

[0158] The derivative of the gradient in any direction (tangent vector, with respect to any direction ) is ​​​​​

[0159]

[0160] wherein,

[0161] ;

[0162]

[0163] After substitution, the settlement amount of the four edges of the marker 9 is The calculation formula of the settlement gradient of the center point of the marker 9 is:

[0164]

[0165] wherein, respectively, the settlement amount of the four edges, the center point of the marker 9 , the gradient direction points to the center of the pile, and the vector is (a), as indicated by the arrow. Figure 4

[0166] As shown in FIG. 9, the settlement correction value Figures 7-9 is calculated according to the settlement gradient of the center point of the marker 9 , and the calculation formula is:

[0167]

[0168] wherein, K is the warping distance of the bottom plate 8, and the calculation formula is: , L is the side length of the bottom plate 8, q , and d is the distance from the center of the marker 9 to the two edges.

[0169] The corrected settlement deformation amount of the pile is calculated, and the calculation formula is:

[0170] .

[0171] By means of the subtle changes of the key points of the marker 9 in the image, the axial deformation gradient of the center point of the marker 9 is calculated, the gradient change is used for interpolation, the settlement deformation amount of the pile is obtained, the error of the settlement measurement result caused by the space deformation of the bottom plate due to stress and self-extensibility is eliminated, the accurate settlement deformation amount of the pile is obtained, and the measurement accuracy is further improved.

[0172] ​​​The application embodiment provides a static load pile load settlement measurement method based on machine vision. Four continuous image processing calculations of the four directions of the bottom plate 8 are performed to obtain the settlement of the four directions of the bottom plate, and then the settlement deformation of the pile is calculated. The application does not need to build a reference beam, is convenient to arrange, realizes automatic high-precision measurement of the settlement of the static load test, has high measurement precision, and can realize continuous measurement of the settlement.

[0173] Embodiment 2

[0174] Based on the same inventive concept, the application embodiment further provides an intelligent terminal including a memory and a processor, and the memory stores a computer program capable of being loaded and executed by the processor to perform a static load pile load settlement measurement method based on machine vision.

[0175] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0176] Based on the above ideal embodiments according to the application, through the above description, relevant personnel can make various changes and modifications without deviating from the technical concept of the application. The technical scope of the application is not limited to the content in the specification, and must be determined according to the scope of claims.

Claims

1. A method for measuring static load settlement based on machine vision, wherein a static load testing platform is installed on a foundation pile (1), characterized in that, The static load testing platform includes: A support frame (2) is set directly above the foundation pile (1), and a load (3) is set above the support frame (2). A hydraulic loading device (4) is installed on the top of the foundation pile (1) and passes through the gap of the support frame (2); The lower pad (8) is located at the bottom of the hydraulic loading device (4), and each of the four corners of the top of the lower pad (8) is provided with a marker (9). The upper pad (6) is set on the top of the hydraulic loading device (4). The center of the upper pad (6) coincides with the center of the lower pad (8). Each of the four corners of the top of the upper pad (6) is provided with a camera (7), and each camera (7) corresponds to a marker (9). The method includes the following steps: S1, control the hydraulic loading device (4) to the set load Q; S2, according to the set time interval, control each camera (7) to simultaneously acquire the image of the corresponding marker (9); S3 performs image processing on each acquired image to obtain image information. l pix , w pix ,in, l pix For the pixel length of the marker (9) in the image, w pix To determine the pixel width of each marker (9) in the image, the image information is processed using a camera mathematical model to obtain the sedimentation change of each marker (9). Specifically, it includes: Obtain the actual parameters of marker (9) l , w and Image information of marker (9) l pix , w pix ; Given the camera (7) parameters, based on the actual parameters of the marker (9) l , w and Image information of markers (9) l pix , w pix Calculate the settlement of each side of marker (9). ; Based on the settlement of the four sides of the marker (9) The settlement change of marker (9) was calculated by bilateral interpolation. ; S4, based on the settlement change of each marker (9), the settlement deformation of the foundation pile at different time points under the load Q is calculated by interpolation; S5. Repeat steps S1 to S4 until the loading termination condition is met. Derive the settlement deformation of the foundation piles at each time node under each load Q and form the settlement Qs curve under different loads Q. The system also includes a correction algorithm to correct the settlement deformation of the foundation piles, specifically including: Settlement based on the four sides of marker (9) Calculate the settlement gradient at the center point of marker (9) The calculation formula is: in, These represent the settlement amounts on the four sides. ; Based on the settlement gradient at the center point of marker (9) Calculate settlement correction value The calculation formula is: Wherein, K is the warpage distance of the lower pad (8); The formula for calculating the corrected pile settlement deformation is as follows: 。 2. The method for measuring static load settlement based on machine vision as described in claim 1, characterized in that, Image processing is performed on each image to obtain image information, specifically including: 3.1 Obtain the template image T of the marker (9); 3.

2. Perform grayscale conversion on the acquired image to obtain a grayscale image containing the marker (9). I G ; 3.3, the grayscale image I G Gaussian blurring is performed to obtain the image to be matched. I ; 3.4, The image to be matched is based on a multi-resolution hierarchical matching method. I Template matching is performed with the template image T of the marker (9) to obtain the matching position of the marker (9). x,y ); 3.5, with the matching position of the marker (9) x,y Using the template image T of the marker (9) as the base, expand the image by N pixels in four directions to the image to be matched. I Cropping is performed to obtain an ROI image containing the marker (9). I roi ; 3.6, For the ROI image I roi Feature point recognition is performed to obtain the pixel coordinates of the top, bottom, left and right endpoints of the marker (9); 3.7 Calculate the image information of the marker (9) based on the pixel coordinates of the four endpoints of the marker (9) (top, bottom, left, and right). l pix , w pix .

3. The method for measuring static load settlement based on machine vision as described in claim 2, characterized in that, The image to be matched is based on a multi-resolution hierarchical matching method. I The template matching with the template image T of the marker (9) specifically includes the following steps: S3.41, Create the images to be matched respectively. I The multi-resolution image sequence of the template image T of the marker (9), the image to be matched I Multi-resolution image sequences are I 0, I 1, I 2, … , I n The multi-resolution image sequence of the template image T is T 0, T 1, T 2, … , T n In this multi-resolution image sequence, the first layer consists of the images with the lowest resolution. n The layer is the original high-resolution image; S3.42, the image to be matched I The first layer of multi-resolution sequences I 0 and the first layer of the multi-resolution sequence of the template image Perform template matching to obtain the first matching region; S3.43, Project the matching region from the previous step onto the next layer of the image to be matched, and use the template image T1 of the next layer multi-resolution sequence to perform matching within the projection region on the image to be matched to obtain a new matching region; S3.44, Repeat step S3.43 until the image to be matched is obtained. I The calculation of the nth layer of the multi-resolution sequence ends, and finally, in the image to be matched... I The best matching position found in the nth layer of the multi-resolution sequence ( x t ,y t The best matching position ( x t ,y t ) is the matching position of marker (9) x,y ).

4. The method for measuring static load settlement based on machine vision as described in claim 1, characterized in that, The conditions for terminating loading include: (1) Under the current level of load Q, when the settlement at the top of the pile (1) is greater than 5 times the settlement under the previous level of load Q, and the total settlement at the top of the pile (1) exceeds 40 mm; (2) Under the current level of load Q, the settlement at the top of the pile (1) is more than twice the settlement under the previous level of load Q, and has not yet reached the relative stability standard after 24 hours. (3) The maximum load Q has been reached and the settlement at the top of the pile (1) has reached the stability standard.

5. The method for measuring static load settlement based on machine vision as described in claim 1, characterized in that, The settlement of each side of the marker (9) The calculation formula is: in, This represents the settlement along the length of the side. This represents the settlement along the width direction.

6. The method for measuring static load settlement based on machine vision as described in claim 1, characterized in that, The camera (7) and the corresponding marker are positioned opposite each other.

7. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 6, which is a machine vision-based static load settlement measurement method.

Images