A bridge crack measurement method, device and equipment based on laser array

Abstract

The present invention provides a bridge crack measurement method, device and equipment based on laser array. The method adopts a drone equipped with a camera and a laser transmitter to shoot damage, and quickly collects damage data on the bridge surface through the laser array, establishes the spatial coordinates of the laser projection point, and then combines the microelement method to realize the conversion of the mapping relationship between the curved surface and the straight surface to obtain the initial image, and uses the deep learning segmentation algorithm to obtain the target binary image of the initial image. Finally, the ratio calculation of crack pixels to all pixels and the Z-type step skeleton line method are used to accurately obtain the area, length and width of the actual crack. The method solves the problem of fast and accurate measurement of curved surface cracks when collecting image data at an inclination angle, and can realize the crack measurement of the plane and curved surface of the bridge. At the same time, it adopts drone non-contact measurement, which can be widely used in measurement projects in other fields. Moreover, the present invention can be realized by only using drones equipped with cameras and laser sensors, and the overall cost is low.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge measurement, and in particular to a bridge crack measurement method, device and equipment based on a laser array. Background Art

[0002] Traditional bridge crack detection methods mainly rely on manual detection, which can achieve effective detection in environments where the difference between background and cracks is obvious. However, when the background is involved or the detection conditions are lacking, the efficiency of crack detection will be significantly reduced. In addition, when measuring structures with potential safety hazards such as buildings, bridges or tunnels, the safety of surveyors may also be threatened. Therefore, manual detection has great limitations and cannot detect surface cracks of bridges in real time, accurately and quickly, thus affecting the maintenance of various bridge facilities. In order to improve detection efficiency and reduce limitations, semi-automatic crack detection vehicles have emerged. However, the use of semi-automatic crack detection vehicles for field measurements can only be carried out in simple environments such as roads or bridges. It is impossible to detect special environments such as tunnel walls or high bridge piers. In addition, crack detection vehicles are expensive and the detection cost is too high.

[0003] Therefore, a new technology of using drones to measure cracks in bridges has emerged in the industry, solving the problems of high cost and single detection environment of crack detection vehicles. However, when using drones to measure cracks in bridges, it is difficult to know the actual size of the damage from the damage images captured by its camera, especially the damage of bridge piers and arches. Since the main physical model of bridge piers and arches is a cylinder, the image data can only show its plane projection state and cannot reflect its three-dimensional damage properties. In addition, it is difficult to obtain a positive vertical shooting angle for damage image acquisition, and most of the images collected are images of damage at an inclined angle. There is generally a large deflection error in the analysis and calculation of the actual size of the damage from the image. The huge error problem in the data caused will lead to the misjudgment of the development trend of bridge damage by the evaluators, which will pose a serious threat to traffic safety and people's lives and property.

[0004] In summary, in the existing bridge crack measurement technology, there are problems such as difficulty in determining the damage size and errors caused by shooting angles, which leads to low accuracy of the detection results. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a bridge crack measurement method, device and equipment based on a laser array to solve the technical problems in the prior art that the damage size is difficult to determine and the errors caused by the shooting angle.

[0006] A bridge crack measurement method based on a laser array, the method comprising: S1 setting a drone equipped with a camera to fly around a bridge at a close distance, the drone being provided with five laser sensors; S2 using the camera to take an image of a damaged part of the bridge, and at the same time, the five laser sensors all emit lasers from point O to the damaged part of the bridge to form five projection points, and the five projection points constitute an N-shaped laser array; S3 establishing the coordinates of the five projection points based on the N-shaped laser array with point O as the origin; S4 substituting the coordinates into a preset curve function to obtain a target curve function; S5 using a differential element method to calculate the target curve function. The coordinates are converted to obtain pixel coordinates, and the pixel coordinates are supplemented by linear interpolation to obtain an initial image; S6 calculates the left and right heights of the initial image according to the N-shaped laser array; S7 processes the initial image using a segmentation algorithm based on deep learning to obtain a target binary image; S8 obtains a bridge crack pixel set in the target binary image and a full pixel set of the target binary image based on the target binary image, and calculates the area of ​​the bridge crack according to the pixel coordinates and the left and right heights; S9 calculates the target length of the bridge crack according to the Z-type step skeleton line method, and obtains the target width of the bridge crack.

[0007] In one embodiment, the drone in step S1 maintains a safety distance of at least 0.3 meters from the bridge during flight.

[0008] In one of the embodiments, the step S3 is specifically as follows: according to the N-shaped laser array, obtaining the distance data between the five projection points and the point O, and the angle data of the five laser sensors; setting a spatial coordinate system with the point O as the origin, and establishing the coordinates of the five projection points according to the distance data and the angle data.

[0009] In one embodiment, the preset curve function in step S4 is specifically:

[0010] x 2 +k1xy+k2y 2 +k3x+k4y+k5=0

[0011] Among them: k1, k2, k3, k4, k5 are the coefficients of the curve equation.

[0012] In one embodiment, the formula used by the differential element method in step S5 is specifically:

[0013]

[0014] Where n is the pixel length of the arc FM projected on the x-axis, Δx iis the length of the i-th unit pixel difference of the curve function on the x-axis, Δy i is the curve function on the y-axis with Δx i The corresponding pixel difference length of the i-th unit, l n is the pixel arc length of arc FM, and the required pixel point coordinates are obtained from the pixel arc length.

[0015] In one embodiment, the step S9 is specifically:

[0016] S901 determines the edge of the bridge crack, divides the edge into a Q edge and a P edge by using normal vector estimation, and takes a point on the Q edge. And search for the point closest to Q0 on the P side Calculate the Euclidean distance I(Q0,P0) and obtain the midpoint coordinates I0 between the two points. The calculation formula is as follows:

[0017]

[0018]

[0019] S902 sets ΔI as the step size, and Q0 searches for two points on the edge of P whose distance is equal to the step size. The calculation formula of ΔI is:

[0020] ΔI i =δI(Q i-1 ,P i-1 )

[0021] Among them, δ is the step size factor;

[0022] S903 Search distance P on Q edge 1+ , P 1- The two nearest points Calculate its Euclidean distance I(Q 1± ,P 1± ), and obtain the midpoint coordinates between the two points I 1± , the calculation formula is as follows:

[0023]

[0024]

[0025] S904 uses the same formula as in step S902 and step S903, Q 1+ , Q 1- By searching for two points on the edge of P whose distance is equal to the step size Search for distance P on Q edge 2+ , P 2- The two nearest points Calculate its Euclidean distance I(Q2± ,P 2± ), and obtain the midpoint coordinates between the two points I 2± ;

[0026] S905 repeats step S904, and calculates all I(Q i± ,P i± ), I i± , and connect all midpoints I i± A connecting line I is formed, wherein the connecting line I is the skeleton line of the bridge crack, and the length of the skeleton line is the target length of the bridge crack;

[0027] S906 Calculate the average width of the bridge crack The specific formula is as follows:

[0028]

[0029] Among them, I(Q i± ,P i± ) is the Euclidean distance between the i-th P and Q, and the average width That is the target width of the bridge crack.

[0030] In one of the embodiments, after step S9, the method further includes: regularly measuring the cracks of the bridge and drawing a crack trend graph; and analyzing and predicting the change trend of the cracks of the bridge according to the crack trend graph.

[0031] A bridge crack measuring device based on laser array, comprising a drone building module, a bridge shooting module, a coordinate building module, a function building module, an image building module, a height calculation module, an image processing module, a crack area calculation module and a crack length and width calculation module, wherein: the drone building module is used to set a drone equipped with a camera to fly around the bridge at a close distance, and the drone is provided with five laser sensors; the bridge shooting module is used to use the camera to take images of the damaged part of the bridge, and at the same time, the five laser sensors all emit lasers from point O to the damaged part of the bridge to form five projection points, and the five projection points constitute an N-shaped laser array; the coordinate building module is used to establish the coordinates of the five projection points according to the N-shaped laser array with point O as the origin; the function building module is used to substitute the coordinates into a preset curve function The target curve function is obtained from the number; the image construction module is used to use the differential element method to perform coordinate conversion on the target curve function to obtain pixel coordinates, and linear interpolation is performed on the pixel coordinates to obtain the initial image; the height calculation module is used to calculate the left and right heights of the initial image according to the N-shaped laser array; the image processing module is used to process the initial image using a segmentation algorithm based on deep learning to obtain a target binary image; the crack area calculation module is used to obtain the bridge crack pixel set in the target binary image and the entire pixel set of the target binary image based on the target binary image, and calculate the area of ​​the bridge crack according to the pixel coordinates and the left and right heights; the crack length and width calculation module is used to calculate the target length of the bridge crack according to the Z-type stepping skeleton line method, and obtain the target width of the bridge crack.

[0032] In one of the embodiments, the device also includes a crack analysis module, which includes a trend drawing unit and a trend analysis unit, wherein: the trend drawing unit is used to regularly measure the bridge cracks and draw a crack trend graph; the trend analysis unit is used to analyze and predict the changing trend of the bridge cracks based on the crack trend graph.

[0033] A device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of a bridge crack measurement method based on a laser array described in each of the above embodiments are implemented.

[0034] From the above-mentioned technology, a bridge crack measurement method, device and equipment based on laser array, it can be seen that the beneficial technical effects of the present invention are as follows:

[0035] The present invention adopts a drone equipped with a camera and a laser transmitter to shoot damage, and quickly collects damage data on the bridge surface through a laser array, establishes the spatial coordinates of the laser projection point, and combines the microelement method to realize the conversion of the mapping relationship between the curved surface and the straight surface to obtain the initial image, and uses the deep learning segmentation algorithm to obtain the target binary image of the initial image. Finally, based on the target binary image, the ratio of crack pixels to all pixels and the Z-type step skeleton line method are used to accurately obtain the area, length and width of the actual crack. This method solves the problem of fast and accurate measurement of curved surface cracks when collecting image data at an inclination angle, and can realize the crack measurement of the plane and curved surface of the bridge. At the same time, it adopts drone non-contact measurement, which can be widely used in measurement projects in other fields. Moreover, the present invention can be realized by only using drones equipped with cameras and laser sensors, and the overall cost is low. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the prior art description. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0037] Figure 1 A flow chart of a bridge crack measurement method based on laser dot matrix provided in an embodiment of the present invention;

[0038] Figure 2 A schematic diagram of a bridge crack measurement method based on a laser array provided by an embodiment of the present invention working on a bridge pier;

[0039] Figure 3 A schematic diagram of a bridge crack measurement method based on a laser array provided by an embodiment of the present invention working on a bridge arch;

[0040] Figure 4 A schematic diagram of an N-shaped laser array provided in an embodiment of the present invention;

[0041] Figure 5 for Figure 2 Front view of the measuring section of the middle crack;

[0042] Figure 6 for Figure 2 Left view of the measurement section of the middle crack;

[0043] Figure 7 for Figure 2 Top view of the measured part of the middle crack;

[0044] Figure 8 is a principle diagram of coordinate conversion of a target curve function provided by an embodiment of the present invention;

[0045] Fig. 9 is a schematic diagram of a Z-type step skeleton line method provided by an embodiment of the present invention;

[0046] Fig.10 A structural block diagram of a bridge crack measurement device based on a laser array provided in an embodiment of the present invention;

[0047] Fig.11 It is a diagram of the internal structure of the device provided by the embodiment of the present invention.

[0048] Reference numerals:

[0049] 1- Drone, 2- Bridge pier, 3- Bridge arch. DETAILED DESCRIPTION

[0050] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0051] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0052] In one embodiment, Figure 1 As shown, a bridge crack measurement method based on laser dot matrix is ​​provided, comprising the following steps:

[0053] S1 sets a drone equipped with a camera to fly around the bridge at a close distance, and the drone is equipped with five laser sensors; S2 uses a camera to capture images of the damaged part of the bridge, and at the same time, five laser sensors all emit lasers from point O to the damaged part of the bridge to form five projection points, and the five projection points constitute an N-shaped laser array; S3 establishes the coordinates of the five projection points based on the N-shaped laser array with point O as the origin; S4 substitutes the coordinates into the preset curve function to obtain the target curve function; S5 uses the differential element method to convert the coordinates of the target curve function to obtain the pixel point coordinates, and linearly interpolates and supplements the pixel point coordinates to obtain the initial image; S6 calculates the left and right heights of the initial image based on the N-shaped laser array; S7 uses a segmentation algorithm based on deep learning to process the initial image to obtain a target binary image; S8 obtains the bridge crack pixel set and the entire pixel set in the target binary image based on the target binary image, and calculates the area of ​​the bridge crack based on the pixel point coordinates and the left and right heights; S9 calculates the target length of the bridge crack based on the Z-type step skeleton line method, and obtains the target width of the bridge crack.

[0054] Specifically, a rotating flat plate is arranged directly in front of the UAV, and a camera is mounted on the rotating platform. The mounted camera is a high-definition distortion-free camera, which is used to photograph the surface damage of the bridge. At the same time, five laser sensors are arranged on the rotating platform of the UAV, and the five laser sensors are arranged directly above the high-definition distortion-free camera. The laser sensor is used to emit laser. While the camera is used to capture images of the damaged part of the bridge, the five laser sensors all emit lasers from point O. The laser reaches the damaged part of the bridge to form five projection points. The five projection points constitute an N-shaped laser array. Then, according to the N-shaped laser array, the coordinates of the five projection points are established with point O as the origin.

[0055] However, the cracks on the bridge are divided into two types due to the different bridge surfaces: piers and arches. The schematic diagram of the inspection work for piers and arches is as follows: Figure 2-Figure 3 As shown, Figure 2 This is a schematic diagram of the pier crack detection work, in which a rotating platform is set on the drone, on which is mounted a high-definition distortion-free camera (i.e. camera) and an array of five laser sensors. The five laser sensors all emit lasers from point O, and the lasers reach the bridge surface, forming five projection points A, B, C, D, and E in an N-shaped distribution. There are five projection points and crack damage in the field of view of the camera. Set point O as the origin to establish a spatial rectangular coordinate system, with the z-axis along the opposite direction of the plumb line (vertically upward), the y-axis in the xOy plane and pointing straight ahead along the center line of the camera's field of view, and the x-axis is perpendicular to the yOz plane. Figure 3 This is a schematic diagram of the bridge arch crack detection work. Similarly, point O is set as the origin to establish a spatial rectangular coordinate system. The y-axis is along the opposite direction of the plumb line (vertically upward), the z-axis is in the horizontal plane passing through point O and along the opposite direction of the camera's field of view center line, and the x-axis is perpendicular to the yOz plane.

[0056] like Figure 4 As shown in the figure, while the camera is capturing the damage image, the five laser sensors of the laser array emit lasers from point O to the damage site, forming an N-shaped distribution, and collecting the distance L from the five laser sensors to the damage site. OA , L OB , L OC , L OD , L OE ,by Figure 2 For example, Figure 5-Figure 7 As shown, Figure 5 yes Figure 2 The front view of the crack measurement part, in which there is a horizontal plane passing through point O and perpendicular to the plumb line, and the projection points of the five projection points A, B, C, D, and E along the opposite direction of the plumb line (vertically upward) on this horizontal plane are A′, B′, C′, D′, and E′. Figure 6 yes Figure 2Left view of the crack measurement part, where the line connecting point O and projection point A forms line segment OA, and the angle between line segment OA and the plumb line is set to α A ; Similarly, let the angle between line segment OB and the plumb line be α B , the angle between line segment OC and the plumb line is α C , the angle between line segment OD and the plumb line is α D , the angle between line segment OE and the plumb line is α E ; The length of line segment OA is L OA , the length of line segment OB is L OB 、The length of line segment OC is L OC , the length of line segment OD is L OD , the length of line segment OE is L OE . Figure 7 It is a top view of the crack measurement part, where the line connecting point O and projection point A′ forms line segment OA′, and the angle between line segment OA′ and y axis is set to θ A ; Similarly, let the angle between line segment OB′ and the y-axis be θ B , the angle between line segment OD′ and the y-axis is θ D , the angle between line segment OE′ and the y-axis is θ E ; The length of line segment OA′ is L OA sinα A , the length of line segment OA′ is L OB sinα B , the length of line segment OA′ is L OC sinα C , the length of line segment OA′ is L OD sinα D , the length of line segment OE′ is L OE sinα E . You can calculate the projection point matrix, that is, calculate the coordinates of the five projection points A′, B′, C′, D′, and E′, and construct the matrix on the left. The specific formula is as follows:

[0057]

[0058] Because the surface curve equation of the existing bridge piers and arches basically conforms to the characteristics of quadratic curves, the specific curve equation is set as follows:

[0059] x 2 +k1xy+k2y 2 +k3x+k4y+k5=0

[0060] Among them: k1, k2, k3, k4, k5 are the coefficients of the curve equation. Substituting the x and y coordinates of the five projection points A', B', C', D', E' into the curve function, we can get the target curve function. Then, we use the differential element method to convert the coordinates of the target curve function, as follows: Figure 8 As shown, first of all, the radius of the surface is known, and the actual width W of the collected image is calculated. The idea of ​​the differential element method is used to solve the coordinate conversion of the mapping relationship between the curve and the straight line. The mapping relationship is the relationship between the pixel length of the projection of the arc FM on the x-axis on the curve and the pixel arc length of the arc FM. The specific formula is as follows:

[0061]

[0062] Where n is the pixel length of the arc FM projected on the x-axis, Δx i is the length of the i-th unit pixel difference of the curve function on the x-axis, Δy i is the curve function on the y-axis with Δx i The corresponding pixel difference length of the i-th unit, l n is the pixel arc length of arc FM, and the pixel coordinates are obtained from the pixel arc length. n The conversion relationship of all x coordinates with the curve function is used to obtain an image with a width of W, and then the image data is supplemented by linear interpolation to obtain the converted image, which is the initial image. The width of the initial image is still the width W;

[0063] Secondly, calculate the left and right heights of the initial image. The calculation formula is as follows:

[0064]

[0065] Among them, H L , H R Are the left height and right height of the image, where the left height is the vertical height of the two projection points A and B, and the right height is the vertical height of the two projection points D and E. The two height data are calculated to reduce the error of left-right asymmetry of the converted image. The width of the converted image remains unchanged and is still the width W.

[0066] The converted image (i.e., the initial image) is processed using a trained deep learning-based segmentation algorithm. Existing deep learning-based segmentation algorithms include Vision Transformer, FCN, SegNet, U-Net, DilatedConvolutions, DeepLab (v1&v2), RefineNet, PSPNet, Large Kernel Matters, DeepLab v3, etc. The segmentation algorithm used here is one of these segmentation algorithms. Obtain the converted binary image (i.e., the target binary image). The width of the target binary image remains unchanged and is still the width W. Use the target binary image to calculate the crack area by pixels and proportions. The specific formula is as follows:

[0067]

[0068] Where S is the crack area, is the set of crack pixels in the converted binary image, Φ is the set of all pixels in the converted target binary image, and W is the width of the converted image.

[0069] The area of ​​the bridge crack is calculated according to the above formula. Based on the above target binary image, Fig. 9 As shown in the figure, the crack edge is obtained by finding the edge, and then the edge is divided into Q edge and P edge by normal vector estimation. And search for the point closest to Q0 on the P side Calculate the Euclidean distance I(Q0,P0) and obtain the midpoint coordinates I0 between the two points. The calculation formula is as follows:

[0070]

[0071]

[0072] Let ΔI be the step size, Q0 searches for two points on the edge of P whose distance is equal to the step size. The calculation formula of ΔI is:

[0073] ΔI i =δI(Q i-1 ,P i-1 )

[0074] Among them, δ is the step size factor;

[0075] Then search for distance P on the Q side 1+ , P 1- The two nearest points Calculate its Euclidean distance I(Q 1± ,P 1± ), and obtain the midpoint coordinates between the two points I 1± , the calculation formula is as follows:

[0076]

[0077]

[0078] Using the same formula as in the previous step, Q 1+ , Q 1- By searching for two points on the edge of P whose distance is equal to the step size Search for distance P on Q edge 2+ , P 2- The two nearest points Calculate its Euclidean distance I(Q 2± ,P 2± ), and obtain the midpoint coordinates between the two points I 2±;

[0079] According to the above calculation method, all I(Q i± ,P i± ), I i± , and connect all midpoints I i± A connecting line I is formed, and the connecting line I is the skeleton line of the bridge crack, and the length of the skeleton line is the target length of the bridge crack;

[0080] Further calculation of the average width of bridge cracks The specific formula is as follows:

[0081]

[0082] Among them, I(Q i± ,P i± ) is the Euclidean distance between the i-th P and Q, and the average width This is the target width of the bridge crack.

[0083] Finally, the area, target length and target width of the bridge crack are obtained, and the bridge crack is measured. The key to the Z-type step skeleton method is to set the step factor. Adjusting the step factor can change the calculation accuracy of the crack length. By setting a suitable step factor, an effective balance between operation speed and accuracy can be achieved.

[0084] In the above embodiment, damage data of the bridge surface is quickly collected by a laser array, the spatial coordinates of the laser projection points are established, and the initial image is obtained by converting the mapping relationship between the curved surface and the straight surface in combination with the microelement method. The target binary image of the initial image is obtained by using a deep learning segmentation algorithm. Finally, based on the target binary image, the ratio of crack pixels to all pixels is calculated and the Z-type step skeleton line method is used to accurately obtain the area, length and width of the actual crack. This method solves the problem of rapid and accurate measurement of curved surface cracks when image data is collected at an inclination angle, and can realize crack measurement of bridge planes and curved surfaces. At the same time, non-contact measurement by drones is adopted, which can be widely used in measurement projects in other fields. Moreover, the present invention can be realized by only using drones equipped with cameras and laser sensors, and the overall cost is relatively low.

[0085] In one embodiment, the drone in step S1 maintains a safety distance of at least 0.3 meters from the bridge when flying. Specifically, when the drone flies around the bridge surface, it needs to maintain a safety distance of at least 0.3 meters, so as to avoid the drone getting too close to the bridge surface and causing a collision.

[0086] In one embodiment, step S3 specifically includes: obtaining the distance data between the five projection points and point O and the angle data of the five laser sensors according to the N-shaped laser array; setting a spatial coordinate system with point O as the origin, and establishing the coordinates of the five projection points according to the distance data and the angle data. Specifically, the distance data here is the distance L between the five laser sensors and the damaged part of the bridge. OA , L OB , L OC , L OD , L OE ; The angle data here is the horizontal angle between each sensor θ A ,θ B ,θ D ,θ E (like Figure 7 ), and the angle α between each sensor and the plumb line A , α B , α C , α D , α E (like Figure 6 As shown in ). Based on the distance data and angle data, the coordinates of the projection points with point O as the origin can be calculated, and the coordinate matrix of the five projection points can be further obtained. The projection points of the five laser sensors on the damaged surface of the bridge form an N-shaped uniform distribution to reduce calculation errors. The coordinate matrix of the projection points is established using angle data and distance data, which solves the data conversion problem of camera tilt shooting and improves the accuracy of the size measurement of bridge cracks.

[0087] In one embodiment, after step S9, the method further includes: measuring the cracks of the bridge regularly and drawing a crack trend graph; and analyzing and predicting the change trend of the cracks of the bridge according to the crack trend graph. Specifically, the cracks of the bridge are measured regularly, the data are numbered in chronological order, and a crack damage trend graph is drawn. The graph should include but not be limited to the numerical values ​​and change rates of the crack quantity, length, height and area in each period, and combined with professional evaluation standards, the change trend of the cracks of the bridge is analyzed and predicted, so as to ensure traffic safety and the safety of people's lives and property.

[0088] In one embodiment, Fig.10 As shown, a bridge crack measurement device 200 based on laser dot matrix is ​​provided, which includes a drone building module 210, a bridge shooting module 220, a coordinate establishment module 230, a function building module 240, an image building module 250, a height calculation module 260, an image processing module 270, a crack area calculation module 280 and a crack length and width calculation module 290, wherein:

[0089] The drone building module 210 is used to set up a drone equipped with a camera to fly around the bridge at a close distance, and the drone is equipped with five laser sensors;

[0090] The bridge shooting module 220 is used to shoot images of the damaged part of the bridge using a camera, and at the same time, five laser sensors all emit lasers from point O to the damaged part of the bridge to form five projection points, and the five projection points constitute an N-shaped laser array;

[0091] The coordinate establishing module 230 is used to establish the coordinates of five projection points based on the N-shaped laser array with point O as the origin;

[0092] The function construction module 240 is used to substitute the coordinates into a preset curve function to obtain a target curve function;

[0093] The image construction module 250 is used to convert the coordinates of the target curve function using the differential method to obtain the pixel coordinates, and to perform linear interpolation to supplement the pixel coordinates to obtain the initial image;

[0094] The height calculation module 260 is used to calculate the left and right heights of the initial image according to the N-shaped laser array;

[0095] The image processing module 270 is used to process the initial image using a segmentation algorithm based on deep learning to obtain a target binary image;

[0096] The crack area calculation module 280 is used to obtain the bridge crack pixel set in the target binary image and the entire pixel set of the target binary image based on the target binary image, and calculate the area of ​​the bridge crack according to the pixel point coordinates and the left and right heights;

[0097] The crack length and width calculation module 290 is used to calculate the target length of the bridge crack according to the Z-type step skeleton line method, and obtain the target width of the bridge crack.

[0098] In one embodiment, the device also includes a crack analysis module, which includes a trend drawing unit and a trend analysis unit, wherein: the trend drawing unit is used to regularly measure bridge cracks and draw a crack trend graph; the trend analysis unit is used to analyze and predict the changing trend of bridge cracks based on the crack trend graph.

[0099] In one embodiment, a device is provided, which may be a server, and its internal structure diagram may be as follows: Fig.11As shown. The device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the device is used to provide computing and control capabilities. The memory of the device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the device is used to store configuration templates and can also be used to store target web page data. The network interface of the device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a bridge crack measurement method based on a laser dot matrix is ​​implemented.

[0100] Those skilled in the art will understand that Fig.11 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the device to which the scheme of the present application is applied. The specific device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0101] Obviously, those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a computer storage medium (ROM / RAM, magnetic disk, optical disk) and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than that here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Therefore, the present invention is not limited to any specific combination of hardware and software.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A bridge crack measurement method based on laser array, characterized in that: include: S1 sets a drone equipped with a camera to fly around the bridge at close range. The drone is equipped with five laser sensors. S2 uses the camera to take an image of the damaged part of the bridge, and at the same time, the five laser sensors all emit lasers from point O to the damaged part of the bridge to form five projection points, and the five projection points constitute an N-shaped laser array; S3, according to the N-shaped laser array, taking the point O as the origin, establishing the coordinates of the five projection points; S4: Substituting the coordinates into a preset curve function to obtain a target curve function; S5 uses a differential method to perform coordinate conversion on the target curve function to obtain pixel point coordinates, and performs linear interpolation to supplement the pixel point coordinates to obtain an initial image; S6, calculating the left and right heights of the initial image according to the N-shaped laser array; S7 processes the initial image using a segmentation algorithm based on deep learning to obtain a target binary image; S8, based on the target binary image, obtaining a bridge crack pixel set in the target binary image and a whole pixel set of the target binary image, and calculating an area of ​​the bridge crack according to the pixel point coordinates and the left and right heights; S9: calculating the target length of the bridge crack according to the Z-type step skeleton method, and obtaining the target width of the bridge crack; The formula used by the differential element method in step S5 is specifically: Where n is the pixel length of the arc FM projected on the x-axis, Δx i is the length of the i-th unit pixel difference of the curve function on the x-axis, Δy i is the curve function on the y-axis with Δx i The corresponding pixel difference length of the i-th unit, l n is the pixel arc length of arc FM, and the required pixel point coordinates are obtained from the pixel arc length.

2. The method according to claim 1, characterized in that The drone in step S1 maintains a safety distance of at least 0.3 meters from the bridge when flying.

3. The method according to claim 1, characterized in that The step S3 is specifically as follows: According to the N-shaped laser array, obtaining distance data between the five projection points and the O point, and angle data of the five laser sensors; A spatial coordinate system with the point O as the origin is set, and coordinates of the five projection points are established according to the distance data and the angle data.

4. The method according to claim 1, characterized in that: The preset curve function in step S4 is specifically: x 2 +k1xy+k2y 2 +k3x+k4y+k5=0 Among them: k1, k2, k3, k4, k5 are the coefficients of the curve equation.

5. The method according to claim 1, characterized in that The step S9 is specifically as follows: S901 determines the edge of the bridge crack, divides the edge into a Q edge and a P edge using normal vector estimation, and takes a point Q0(x Q0 ,y Q0 ), and search for the point P0(x P0 ,y P0 ), calculate its Euclidean distance I(Q0,P0), and obtain the midpoint coordinates I0 between the two points. The calculation formula is as follows: S902 sets ΔI as the step size, and Q0 searches for two points P on the edge of P whose distance is equal to the step size. 1+ (x P1+ ,y P1+ ), P 1- (x P1- ,y P1- ), the calculation formula of ΔI is: ΔI i =δI(Q i-1 ,P i-1 ) Among them, δ is the step size factor; S903 Search distance P on Q edge 1+ , P 1- The two nearest points Q 1+ (x Q1+ ,y Q1+ ), Q 1- (x Q1- ,y Q1- ), calculate its Euclidean distance I(Q 1± ,P 1± ), and obtain the midpoint coordinates between the two points I 1± , the calculation formula is as follows: S904 uses the same formula as in step S902 and step S903, Q 1+ , Q 1- By searching for two points P on the edge of P whose distance is equal to the step size 2+ (x P2+ ,y P2+ ), P 2- (x P2- ,y P2- ), search for distance P on the Q edge 2+ , P 2- The two nearest points Q 2+ (x Q2+ ,y Q2+ ), Q 2- (x Q2- ,y Q2- ), calculate its Euclidean distance I(Q 2± ,P 2± ), and obtain the midpoint coordinates between the two points I 2± ; S905 repeats step S904, and calculates all I(Q i± ,P i± ), I i± , and connect all midpoints I i± A connecting line I is formed, wherein the connecting line I is the skeleton line of the bridge crack, and the length of the skeleton line is the target length of the bridge crack; S906 Calculate the average width of the bridge crack The specific formula is as follows: Among them, I(Q i± ,P i± ) is the Euclidean distance between the i-th P and Q, and the average width That is the target width of the bridge crack.

6. The method according to claim 1, characterized in that After step S9, the method further includes: Regularly measuring the cracks of the bridge and drawing a crack trend diagram; According to the crack trend diagram, the changing trend of bridge cracks is analyzed and predicted.

7. A bridge crack measurement device based on a laser array, characterized in that: It includes a drone building module, a bridge shooting module, a coordinate building module, a function building module, an image building module, a height calculation module, an image processing module, a crack area calculation module and a crack length and width calculation module, among which: The drone building module is used to set up a drone equipped with a camera to fly around the bridge at a close distance, and the drone is equipped with five laser sensors; The bridge shooting module is used to use the camera to shoot images of the damaged part of the bridge, and at the same time, the five laser sensors all emit lasers from point O to the damaged part of the bridge to form five projection points, and the five projection points constitute an N-shaped laser array; The coordinate establishment module is used to establish the coordinates of the five projection points based on the N-shaped laser array and taking the O point as the origin; The function construction module is used to substitute the coordinates into a preset curve function to obtain a target curve function; The image construction module is used to convert the coordinates of the target curve function by using a differential element method to obtain pixel coordinates, and to perform linear interpolation on the pixel coordinates to obtain an initial image; The height calculation module is used to calculate the left and right heights of the initial image according to the N-shaped laser array; The image processing module is used to process the initial image using a segmentation algorithm based on deep learning to obtain a target binary image; The crack area calculation module is used to obtain, based on the target binary image, a bridge crack pixel set in the target binary image and a whole pixel set of the target binary image, and calculate the area of ​​the bridge crack according to the pixel point coordinates and the left and right heights; The crack length and width calculation module is used to calculate the target length of the bridge crack according to the Z-type step skeleton line method, and obtain the target width of the bridge crack; The image construction module is used for: the formula used by the differential element method is specifically: Where n is the pixel length of the arc FM projected on the x-axis, Δx i is the length of the i-th unit pixel difference of the curve function on the x-axis, Δy i is the curve function on the y-axis with Δx i The corresponding pixel difference length of the i-th unit, l n is the pixel arc length of arc FM, and the required pixel point coordinates are obtained from the pixel arc length.

8. The device according to claim 7, characterized in that It also includes a crack analysis module, which includes a trend drawing unit and a trend analysis unit, wherein: The trend drawing unit is used to regularly measure the cracks of the bridge and draw a crack trend diagram; The trend analysis unit is used to analyze and predict the change trend of bridge cracks according to the crack trend graph.

9. A device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 6 when executing the computer program.

Images