Pier rigidity detection method and system

Abstract

The invention provides a bridge pier rigidity detection method and system, and relates to the technical field of bridge pier detection. The method comprises the following steps: setting a to-be-detected sample point, and obtaining the central position of a to-be-detected area; setting a preset horizontal distance position of the central position as a hovering position of the unmanned aerial vehicle, enabling laser emitted by a laser range finder arranged on the unmanned aerial vehicle to irradiate the to-be-measured sample points, and determining a measurement distance measured by the laser range finder and a pitch angle and a yaw angle of the laser range finder when a plurality of to-be-measured sample points are measured; determining a rigidity reduction coefficient; obtaining a to-be-detected image of the to-be-detected area; determining a rigid risk coefficient; and determining a rigidity detection result. According to the invention, the unmanned aerial vehicle carries the laser range finder for non-contact detection, equipment does not need to be installed on the pier, the efficiency and safety of rigidity detection are improved, the rigidity of the pier can be comprehensively evaluated from two dimensions of geometric deformation and surface cracks, and the comprehensiveness and accuracy of rigidity detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge pier detection, and in particular to a bridge pier rigidity detection method and system. Background Art

[0002] Although the current relevant technologies can detect quality problems on bridge piers, installing equipment on the piers may cause potential interference to the pier structure, and the risk of pier rigidity degradation is not considered. In other words, it is impossible to comprehensively evaluate the rigidity of the piers from the two dimensions of geometric deformation and surface cracks.

[0003] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0004] The present invention provides a bridge pier rigidity detection method and system, which can solve the technical problems that related technologies require equipment to be installed on the bridge piers and cannot comprehensively evaluate the rigidity of the bridge piers from two dimensions of geometric deformation and surface cracks.

[0005] According to a first aspect of the present invention, a method for detecting rigidity of a pier is provided, comprising: setting the side of the pier to be detected as four areas to be detected, setting a plurality of sample points to be detected in the areas to be detected, and obtaining the center position of the areas to be detected; setting a preset horizontal distance of the center position as a hovering position of an unmanned aerial vehicle, irradiating the laser emitted by a laser rangefinder provided on the unmanned aerial vehicle to the sample points to be detected at multiple times in a current detection cycle, and determining the measured distances of the plurality of sample points to be detected measured by the laser rangefinder, as well as the pitch angle and yaw angle of the laser rangefinder when measuring the plurality of sample points to be detected; determining a rigidity reduction coefficient according to the measured distance, the pitch angle, the yaw angle and the preset horizontal distance; obtaining images of the areas to be detected at the start and end of the current detection cycle, respectively; determining a rigidity risk coefficient according to the images to be detected; and determining a rigidity detection result of the pier to be detected in the current detection cycle according to the rigidity reduction coefficient and the rigidity risk coefficient.

[0006] Furthermore, the number of the sample points to be tested in each area to be tested is the same, but the positions are randomly distributed.

[0007] Further, determining a rigid descent coefficient according to the measured distance, the pitch angle, the yaw angle, and the preset horizontal distance includes: taking the hovering position of the drone as the origin, taking the orientation when the pitch angle of the laser rangefinder is 0° and the yaw angle is 90° as the X-axis, taking the orientation when both the pitch angle and the yaw angle of the laser rangefinder are 0° as the Y-axis, and taking the vertical direction as the Z-axis to establish a coordinate system for the sample points to be measured; obtaining the central position coordinates of the central position in the coordinate system of the sample points to be measured according to the preset horizontal distance; determining the to-be-measured position vectors of multiple to-be-measured sample points in multiple to-be-measured regions at multiple moments in the current detection period according to the measured distance, the pitch angle, the yaw angle, and the central position coordinates; and determining the rigid descent coefficient according to the to-be-measured position vectors.

[0008] Further, determining the to-be-measured position vectors of multiple to-be-measured sample points in multiple to-be-measured regions at multiple moments in the current detection period according to the measured distance, the pitch angle, the yaw angle, and the central position coordinates includes: according to the formula , , determining the to-be-measured position vector of the j-th to-be-measured sample point in the i-th to-be-measured region at the k-th moment in the current detection period , where is the measured distance of the j-th to-be-measured sample point in the i-th to-be-measured region at the k-th moment in the current detection period, is the pitch angle of the j-th to-be-measured sample point in the i-th to-be-measured region at the k-th moment in the current detection period, and the unit of the pitch angle is degree, is the yaw angle of the j-th to-be-measured sample point in the i-th to-be-measured region at the k-th moment in the current detection period, and the unit of the yaw angle is degree, is the to-be-measured position coordinate of the j-th to-be-measured sample point in the i-th to-be-measured region at the k-th moment in the current detection period, is the preset horizontal distance, is the central position coordinate, and i, j, and k are all positive integers.

[0009] Further, determining the rigid descent coefficient according to the to-be-measured position vectors includes: obtaining the width and height of the to-be-measured bridge pier; and determining the rigid descent coefficient according to the to-be-measured position vectors, the width, and the height.

[0010] Further, determining the rigid descent coefficient according to the to-be-measured position vectors, the width, and the height includes: according to the formula , , , determining the rigid descent coefficient G, where is the cosine similarity between the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period and the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k - 1)-th moment in the current detection period, is the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k - 1)-th moment in the current detection period, is the cosine similarity between the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k + 1)-th moment in the current detection period and the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period, is the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k + 1)-th moment in the current detection period, is the abscissa of the position coordinate of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period, H is the height of the bridge pier to be measured, W is the width of the bridge pier to be measured, 4 is the number of areas to be measured, N is the number of sample points to be measured in each area to be measured, M is the number of moments in the detection period, i ≤ 4, j ≤ N, k ≤ M, and i, j, k, N, and M are all positive integers, if is a conditional function.

[0011] Further, according to the image to be measured, determining the rigidity risk coefficient includes: in the image to be measured, identifying whether there are cracks in the area to be measured through an image detection model; if there are cracks in the area to be measured, determining the target area to be measured with cracks; through a trained image recognition neural network model, performing crack feature extraction processing on the image to be measured at the start moment of the current detection period of the target area to be measured, to obtain a start target area to be measured feature vector; through a trained image recognition neural network model, performing crack feature extraction processing on the image to be measured at the end moment of the current detection period of the target area to be measured, to obtain an end target area to be measured feature vector; determining the rigidity risk coefficient according to the start target area to be measured feature vector and the end target area to be measured feature vector; if there are no cracks in the area to be measured, determining that the rigidity risk coefficient is 0.

[0012] Further, determining the rigidity risk coefficient according to the start target area to be measured feature vector and the end target area to be measured feature vector includes: according to the formula , determining the rigidity risk coefficient F, where is the start target area to be measured feature vector of the s-th target area to be measured, is the end target area to be measured feature vector of the s-th target area to be measured, n is the number of target areas to be measured, s ≤ n, and s and n are both positive integers.

[0013] Further, according to the rigid decline coefficient and the rigid risk coefficient, determine the rigid detection result of the pier to be measured in the current detection period, including: if the rigid risk coefficient is greater than or equal to the set rigid risk coefficient threshold, determine that the rigid detection result of the pier to be measured in the current detection period is severe rigid degradation and the pier structure is dangerous; if the rigid decline coefficient is greater than or equal to the set rigid decline coefficient threshold and the rigid risk coefficient is less than the set rigid risk coefficient threshold, determine that the rigid detection result of the pier to be measured in the current detection period is severe rigid degradation and the pier structure is dangerous; if the rigid decline coefficient is less than the set rigid decline coefficient threshold and the rigid risk coefficient is less than the set rigid risk coefficient threshold, determine that the rigid detection result of the pier to be measured in the current detection period is normal rigidity and the pier structure is safe.

[0014] According to a second aspect of the present invention, there is provided a pier rigid detection system, including: a sample point to be measured and a central position module, configured to set the side of the pier to be measured into four regions to be measured, set a plurality of sample points to be measured in the regions to be measured, and obtain the central position of the regions to be measured; a measurement distance, pitch angle and yaw angle module, configured to set the preset horizontal distance at the central position as the UAV hovering position, at multiple moments in the current detection period, make the laser emitted by the laser rangefinder provided on the UAV irradiate the sample points to be measured, and determine the measurement distances of the plurality of sample points to be measured measured by the laser rangefinder, as well as the pitch angle and yaw angle of the laser rangefinder when measuring the plurality of sample points to be measured; a rigid decline coefficient module, configured to determine the rigid decline coefficient according to the measurement distance, the pitch angle, the yaw angle and the preset horizontal distance; a sample image to be measured module, configured to respectively obtain the sample images to be measured of the regions to be measured at the start moment and the end moment of the current detection period; a rigid risk coefficient module, configured to determine the rigid risk coefficient according to the sample images to be measured; a rigid detection result module, configured to determine the rigid detection result of the pier to be measured in the current detection period according to the rigid decline coefficient and the rigid risk coefficient.

[0015] Technical effect: According to the present invention, non-contact detection is performed by carrying a laser rangefinder on a drone, and there is no need to install equipment on the pier, which reduces the potential interference with the pier structure and improves the efficiency and safety of rigidity detection. By measuring the distance, pitch angle, yaw angle and preset horizontal distance, the three-dimensional coordinates of the sample points to be tested of the pier to be tested can be accurately calculated to reflect the local deformation of the pier to be tested. In addition, the rigidity reduction coefficient and the rigidity risk coefficient are combined to comprehensively evaluate the rigidity of the pier from the two dimensions of geometric deformation and surface damage, which improves the comprehensiveness and accuracy of rigidity detection, and can be used for large-scale pier screening and long-term rigidity detection. When determining the position vectors to be tested of multiple sample points to be tested in multiple areas to be tested at multiple times in the current detection cycle, the position vectors to be tested of multiple sample points to be tested in multiple areas to be tested at multiple times in the current detection cycle can be determined based on the position coordinates to be tested and the center position coordinates. The position vectors to be tested realize the unification of the measurement benchmark, reduce the error caused by the fluctuation of the drone's hovering position, and for multiple measurements of the same sample point to be tested at different times, by comparing the changes in the position vectors to be tested, the time course of rigidity degradation can be quantified. When determining the rigidity reduction coefficient, the weight can be set based on the relative difference of the cosine similarity of the position vector to be tested at adjacent moments, and based on the characteristics that the closer the position vector to be tested is to the horizontal direction, the more dangerous it is when deformed, and the characteristics that the farther the sample point to be tested is from the center position, the stronger the deformation representation ability is. The weight is set, so as to weight and average the relative difference of the cosine similarity of the position vector to be tested at multiple adjacent moments in the current detection cycle of multiple sample points to be tested in multiple test areas to obtain the rigidity reduction coefficient, and improve the accuracy of identifying the degree of rigidity reduction. When determining the rigidity risk coefficient, the rigidity risk coefficient can be determined by the similarity of the feature vector of the starting target test area and the feature vector of the ending target test area, so as to quantify the degree of change of the crack characteristics within the detection cycle, judge the degree of danger of the bridge pier structure to be tested, and improve the reliability of the rigidity risk coefficient.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work. Figure 1 A schematic diagram of a flow chart of a bridge pier rigidity detection method according to an embodiment of the present invention is exemplarily shown; Figure 2Exemplarily shown is a flowchart for calculating the rigidity decline coefficient according to an embodiment of the present invention; Figure 3 Exemplarily shown is a flowchart for calculating the rigidity risk coefficient according to an embodiment of the present invention; Figure 4 Exemplarily shown is a flowchart for determining the rigidity detection result according to an embodiment of the present invention; Figure 5 Exemplarily shown is a block diagram of a pier rigidity detection system according to an embodiment of the present invention. Detailed implementation manners

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only a part rather than all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] The technical solutions of the present invention will be described in detail below with specific embodiments. These specific embodiments may be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0020] Figure 1 Exemplarily shown is a schematic flowchart of a pier rigidity detection method according to an embodiment of the present invention. The method includes: Step S1, setting the side surface of the pier to be measured as four measurement regions to be measured, setting a plurality of measurement points to be measured in the measurement regions to be measured, and obtaining the central positions of the measurement regions to be measured; Step S2, setting the preset horizontal distance at the central position as the hovering position of the unmanned aerial vehicle (UAV). At multiple moments in the current detection period, making the laser emitted by the laser rangefinder arranged on the UAV irradiate the measurement points to be measured, and determining the measurement distances of the multiple measurement points to be measured measured by the laser rangefinder, as well as the pitch angle and yaw angle of the laser rangefinder when measuring the multiple measurement points to be measured; Step S3, determining the rigidity decline coefficient according to the measurement distance, the pitch angle, the yaw angle, and the preset horizontal distance; Step S4, respectively obtaining the images to be measured of the measurement regions to be measured at the start moment and the end moment of the current detection period; Step S5, determining the rigidity risk coefficient according to the images to be measured; Step S6, determining the rigidity detection result of the pier to be measured in the current detection period according to the rigidity decline coefficient and the rigidity risk coefficient.

[0021] According to the pier rigidity detection method of an embodiment of the present invention, non-contact detection is carried out by a drone carrying a laser rangefinder, without installing equipment on the pier, reducing potential interference to the pier structure and improving the efficiency and safety of rigidity detection. By measuring the distance, pitch angle, yaw angle and a preset horizontal distance, the three-dimensional coordinates of the sample points to be measured of the pier to be measured can be accurately calculated, reflecting the local deformation of the pier to be measured. And by comprehensively considering the rigidity decline coefficient and the rigidity risk coefficient, the pier rigidity is comprehensively evaluated from two dimensions of geometric deformation and surface damage, improving the comprehensiveness and accuracy of rigidity detection, and large-scale pier screening and long-term rigidity detection can be carried out.

[0022] According to an embodiment of the present invention, in step S1, the pier to be measured can be selected according to the service years of the pier, and the side of the pier to be measured is set as four measurement areas. The setting method can be determined according to the cross-sectional geometric shape of the pier to be measured. If the cross-section of the pier to be measured is rectangular, the four rectangular sides of the pier to be measured are directly set as four measurement areas. If the cross-section of the pier to be measured is circular, the side of the pier is evenly divided into four equal measurement areas along the circumferential direction (for example, divided by 0°-90°, 90°-180°, 180°-270°, 270°-360°). For each measurement area, its central position is determined by geometric calculation or measurement tools (such as total station, laser rangefinder), and the central position is defined as the centroid of the regional geometric figure (for example, the intersection of the diagonals of a rectangular area).

[0023] According to an embodiment of the present invention, step S1 includes: the number of the sample points to be measured in each measurement area is the same, but the positions are randomly distributed.

[0024] According to an embodiment of the present invention, in step S2, a drone hovering position is set at a preset horizontal distance (for example, 2.5 m) from the central position (at the same height as the central position and directly facing the central position), that is, the centroid of the cross-section of the pier to be measured at the height of the centroid of the measurement area (for example, the above-mentioned center of the circle) is connected to the centroid of the measurement area, and the drone hovering position is located on the extension line of this connection. Therefore, there are four drone hovering positions, so that the laser emitted by the laser rangefinder installed on the drone can irradiate all the sample points to be measured, reducing manual operation and lowering the labor cost and safety risk. The interval between adjacent moments can be set to 3 days, 7 days, etc., and each detection cycle can be set to half a month, 1 month, etc. The present invention does not limit this. When the drone hovers for measurement, the laser rangefinder rotates up, down, left and right to make the emitted laser accurately irradiate the sample points to be measured, and determine the measurement distance, pitch angle and yaw angle of the laser rangefinder at this time. The pitch angle is used to describe the angle at which the laser rangefinder rotates up or down, and the yaw angle is used to describe the angle at which the laser rangefinder rotates left or right.

[0025] According to an embodiment of the present invention, in step S3, a rigidity reduction coefficient is determined according to the measured distance, the pitch angle, the yaw angle and the preset horizontal distance.

[0026] Figure 2 The flowchart for calculating the rigidity reduction coefficient according to an embodiment of the present invention is exemplarily shown.

[0027] According to one embodiment of the present invention, step S3 includes: step S31, taking the hovering position of the drone as the origin, the direction of the laser rangefinder when the pitch angle is 0° and the yaw angle is 90° as the X-axis, the direction of the laser rangefinder when the pitch angle and the yaw angle are both 0° as the Y-axis, and the vertical direction as the Z-axis, to establish a coordinate system of the sample point to be measured; step S32, according to the preset horizontal distance, obtaining the center position coordinates of the center position in the coordinate system of the sample point to be measured; step S33, determining the position vectors of multiple sample points to be measured in multiple test areas at multiple times in the current detection cycle according to the measured distance, the pitch angle, the yaw angle and the center position coordinates; step S34, determining the rigidity reduction coefficient according to the position vector to be measured.

[0028] According to one embodiment of the present invention, four coordinate systems of sample points to be measured are established with four drone hovering positions (i.e., the preset horizontal distances of the four center positions) as the origins, respectively, without interfering with each other. The direction of the X-axis of the coordinate system of the sample points to be measured is the direction in which the pitch angle of the laser rangefinder is 0° and the yaw angle is 90° (90° to the right), the direction of the Y-axis is the direction in which the pitch angle of the laser rangefinder is 0° (no deviation up and down) and the yaw angle is 0° (no deviation left and right), and the Z-axis is vertically upward. When the pitch angle of the laser rangefinder is 0° and the yaw angle is 0°, the emitted laser is directly facing the center position, and in each coordinate system of the sample points to be measured, the coordinates of the center position are the same.

[0029] According to an embodiment of the present invention, based on the measured distance, the pitch angle, the yaw angle and the center position coordinates, determining the position vectors of the multiple test points in the multiple test areas at multiple moments in the current detection cycle includes: determining the position vector of the jth test point in the ith test area at the kth moment in the current detection cycle according to formulas (1) and (2): , (1), (2), in, is the measured distance of the jth sample point to be tested in the i-th test area at the kth moment of the current detection cycle, is the pitch angle of the jth sample point to be tested in the i-th test area at the kth moment of the current detection cycle. The unit of the pitch angle is degree. is the yaw angle of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period. The unit of the yaw angle is degree. is the coordinate of the position to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period. is the preset horizontal distance. is the central position coordinate, and i, j, and k are all positive integers.

[0030] According to an embodiment of the present invention, in formula (1), the laser rangefinder is at the origin of the coordinate system of the sample point to be measured. The distance from the origin to the sample point to be measured is the measured distance of the laser rangefinder. is the projection length of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period in the X-axis direction in the coordinate system of the sample point to be measured, that is, the coordinate value of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period on the X-axis. is the projection length of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period in the Y-axis direction in the coordinate system of the sample point to be measured, that is, the coordinate value of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period on the Y-axis. is the projection length of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period in the Z-axis direction in the coordinate system of the sample point to be measured, that is, the coordinate value of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period on the Z-axis. From the above three coordinate values, the coordinate of the position to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment can be obtained. In formula (2), taking the origin at the preset horizontal distance from the central position, and when the pitch angle of the laser rangefinder is 0° and the yaw angle is 0°, the emitted laser is directly facing the central position. Therefore, is the central position coordinate. When the coordinate of the position to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment is subtracted from the central position coordinate, that is, it represents the vector from the central position coordinate to the coordinate of the position to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment, and the vector of the position to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment can be obtained.

[0031] In this way, based on the coordinate of the position to be measured and the central position coordinate, the vectors of the positions to be measured of multiple sample points to be measured in multiple areas to be measured at multiple moments in the current detection period can be determined. The vectors of the positions to be measured unify the measurement reference, reduce the error caused by the fluctuation of the hovering position of the UAV. For multiple measurements of the same sample point to be measured at different moments, by comparing the changes in the vectors of the positions to be measured, the time-course law of rigid degradation can be quantified.

[0032] According to an embodiment of the present invention, step S34 includes: step S341, obtaining the width and height of the pier to be measured; step S342, determining the rigid descent coefficient according to the position vector to be measured, the width and the height.

[0033] According to an embodiment of the present invention, the width and height of the pier to be measured are obtained through the construction drawings of the pier to be measured, and the rigid descent coefficient is determined through the position vector to be measured, the width and the height.

[0034] According to an embodiment of the present invention, determining the rigid descent coefficient according to the position vector to be measured, the width and the height includes: determining the rigid descent coefficient G according to formulas (3), (4) and (5), (3), (4), (5), where, is the cosine similarity between the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period and the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k - 1)-th moment in the current detection period, is the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k - 1)-th moment in the current detection period, is the cosine similarity between the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k + 1)-th moment in the current detection period and the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period, is the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k + 1)-th moment in the current detection period, is the abscissa of the position coordinate of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period, H is the height of the pier to be measured, W is the width of the pier to be measured, 4 is the number of areas to be measured, N is the number of sample points to be measured in each area to be measured, M is the number of moments in the detection period, i ≤ 4, j ≤ N, k ≤ M, and i, j, k, N and M are all positive integers, and if is a conditional function.

[0035] According to an embodiment of the present invention, in formula (3), is the cosine similarity between the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period and the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k - 1)-th moment in the current detection period. The closer this cosine similarity is to 1, the more similar the position vectors to be measured of the sample point to be measured at the (k - 1)-th moment and the k-th moment in the current detection period are. That is, between the (k - 1)-th moment and the k-th moment in the current detection period, the displacement of the sample point to be measured is smaller. Similarly, in formula (4), is the cosine similarity between the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the (k + 1)-th moment in the current detection period and the position vector to be measured of the j-th sample point to be measured in the i-th area to be measured at the k-th moment in the current detection period. The closer this cosine similarity is to 1, the more similar the position vectors to be measured of the sample point to be measured at the k-th moment and the (k + 1)-th moment in the current detection period are. That is, between the k-th moment and the (k + 1)-th moment in the current detection period, the displacement of the sample point to be measured is smaller. In formula (5), is the relative difference between the cosine similarity between the position vectors to be measured of the sample point to be measured at the (k - 1)-th moment and the k-th moment in the current detection period and the cosine similarity between the position vectors to be measured of the sample point to be measured at the k-th moment and the (k + 1)-th moment in the current detection period. The larger this relative difference is, the greater the change in the displacement of the sample point to be measured within the adjacent time intervals in the current detection period, that is, the greater the degree of rigidity decline of the pier to be measured. is the minimum angle between the position vector to be measured and the vertical direction (i.e., the Z-axis). For example, if the position vector to be measured points to the positive direction of the Z-axis (such as the sample point at the top of the pier), the minimum angle is the angle between the position vector to be measured and the positive direction of the Z-axis. If the position vector to be measured points to the negative direction of the Z-axis (such as the sample point at the bottom of the pier), the minimum angle is the angle between the position vector to be measured and the negative direction of the Z-axis. is the ratio between the minimum angle and Since bridge design can usually tolerate minor vertical settlement, but horizontal displacement will directly change the structural force path, leading to overturning or shear failure. Therefore, the larger this ratio is, the larger the minimum angle between the position vector to be measured and the Z-axis (the closer the position vector to be measured is to the horizontal direction), indicating that the pier to be measured is more vulnerable to damage when the position vector to be measured changes. That is, the more dangerous the pier to be measured is when deforming in this direction, and the greater the degree of rigidity decline of the pier to be measured. Therefore, its weight value is higher. indicates that when the height of the pier to be measured is greater than or equal to the width of the pier to be measured, the value of the conditional function is the height of the pier to be measured. Otherwise, the value of the conditional function is the width of the pier to be measured. is the ratio between the modulus of the position vector to be measured and the above corresponding conditional function. The larger this ratio is, the larger the modulus of the position vector to be measured is, indicating that the position of the sample point to be measured is farther from the central position, and the more accurate the measurement result of the position vector to be measured is. When the position vector to be measured changes, the stronger the ability of this measurement result to characterize the overall deformation of the bridge pier is, that is, the position change of the sample point to be measured farther from the center of the bridge pier can better reflect the overall deformation trend of the bridge pier. Therefore, a higher weight is assigned. Using and , the relative difference in the cosine similarity of the position vectors to be measured of multiple sample points to be measured in multiple adjacent moments in the current detection period in multiple regions to be measured is multiplied by the corresponding weights and then averaged to obtain the rigidity degradation coefficient. The larger this rigidity degradation coefficient is, the greater the degree of rigidity degradation of the bridge pier to be measured is, that is, the more serious the rigidity degradation is.

[0036] In this way, based on the relative difference in the cosine similarity of the position vectors to be measured at adjacent moments, and based on the characteristics that the closer the position vector to be measured is to the horizontal direction, the more dangerous the deformation is, weights are set, and the farther the sample point to be measured is from the central position, the stronger the ability to characterize the deformation is, weights are set. Thus, the relative difference in the cosine similarity of the position vectors to be measured of multiple sample points to be measured in multiple adjacent moments in the current detection period in multiple regions to be measured is weighted and averaged to obtain the rigidity degradation coefficient, improving the accuracy of identifying the degree of rigidity degradation.

[0037] According to an embodiment of the present invention, in step S4, through an image acquisition device, for example, a camera carried by a drone can be used to respectively acquire the images to be measured of the regions to be measured at the start time and the end time of the current detection period. By comparing the images to be measured of the corresponding regions to be measured at the start time and the end time of the current detection period, the change of the surface cracks of the bridge pier to be measured can be analyzed.

[0038] According to an embodiment of the present invention, in step S5, the rigidity risk coefficient is determined according to the image to be measured.

[0039] Figure 3 Exemplarily, a flowchart for calculating the rigidity risk coefficient according to an embodiment of the present invention is shown.

[0040] According to an embodiment of the present invention, step S5 includes: step S51, in the to-be-detected image, identify whether there is a crack in the to-be-detected area through an image detection model; step S52, if there is a crack in the to-be-detected area, determine the target to-be-detected area where the crack exists; step S53, through the trained image recognition neural network model, perform crack feature extraction processing on the to-be-detected image of the target to-be-detected area at the start moment of the current detection cycle, and obtain the start target to-be-detected area feature vector; step S54, through the trained image recognition neural network model, perform crack feature extraction processing on the to-be-detected image of the target to-be-detected area at the end moment of the current detection cycle, and obtain the end target to-be-detected area feature vector; step S55, determine the rigidity risk coefficient according to the start target to-be-detected area feature vector and the end target to-be-detected area feature vector; step S56, if there is no crack in the to-be-detected area, determine that the rigidity risk coefficient is 0.

[0041] According to an embodiment of the present invention, the image detection model belongs to a type of deep learning model (for example, a convolutional neural network model). The image detection model is trained through historical data so that the image detection model can identify whether there is a crack in the to-be-detected image. If the image detection model detects crack features (such as linear structures, edge discontinuities, etc.) in any to-be-detected image, then mark this to-be-detected area as the target to-be-detected area. If no crack is detected, determine that the rigidity risk coefficient is 0, indicating that the to-be-detected pier structure is safe. The image recognition neural network model can be a convolutional neural network model. This image recognition neural network model has the ability to extract features such as crack morphology, length, and width. Perform crack feature extraction processing on the to-be-detected image of the target to-be-detected area at the start moment of the current detection cycle, and the start target to-be-detected area feature vector can be obtained. Perform crack feature extraction processing on the to-be-detected image of the target to-be-detected area at the end moment of the current detection cycle, and the end target to-be-detected area feature vector can be obtained. Compare the start target to-be-detected area feature vector and the end target to-be-detected area feature vector to determine the rigidity risk coefficient.

[0042] According to an embodiment of the present invention, determining the rigidity risk coefficient according to the start target to-be-detected area feature vector and the end target to-be-detected area feature vector includes: determining the rigidity risk coefficient F according to formula (6), (6), where, is the start target to-be-detected area feature vector of the s-th target to-be-detected area, is the end target to-be-detected area feature vector of the s-th target to-be-detected area, n is the number of target to-be-detected areas, s ≤ n, and both s and n are positive integers.

[0043] According to an embodiment of the present invention, in formula (6), is the similarity between the start target area to be measured feature vector and the end target area to be measured feature vector of the s-th target area to be measured. The greater this similarity, the more similar the cracks in the target area to be measured are at the start and end moments of the current detection period, that is, the structure of the target area to be measured is stable. is the average of the similarities between the start target area to be measured feature vectors and the end target area to be measured feature vectors of multiple target areas to be measured, and then subtracting 1 from to obtain the rigidity risk coefficient. The greater this rigidity risk coefficient, the more unstable the structure of the target area to be measured, that is, the more serious the rigidity degradation of the pier to be measured, and thus the more dangerous the structure of the pier to be measured.

[0044] In this way, the rigidity risk coefficient can be determined through the similarity between the start target area to be measured feature vector and the end target area to be measured feature vector, so as to quantify the change degree of the crack feature during the detection period, judge the danger degree of the structure of the pier to be measured, and improve the reliability of the rigidity risk coefficient.

[0045] According to an embodiment of the present invention, in step S6, the rigidity detection result of the pier to be measured in the current detection period is determined according to the rigidity decline coefficient and the rigidity risk coefficient.

[0046] Figure 4 Exemplarily, a flowchart of determining the rigidity detection result according to an embodiment of the present invention is shown.

[0047] According to an embodiment of the present invention, step S6 includes: step S61, if the rigidity risk coefficient is greater than or equal to the set rigidity risk coefficient threshold, it is determined that the rigidity detection result of the pier to be measured in the current detection period is severe rigidity degradation and the pier structure is dangerous; step S62, if the rigidity decline coefficient is greater than or equal to the set rigidity decline coefficient threshold and the rigidity risk coefficient is less than the set rigidity risk coefficient threshold, it is determined that the rigidity detection result of the pier to be measured in the current detection period is severe rigidity degradation and the pier structure is dangerous; step S63, if the rigidity decline coefficient is less than the set rigidity decline coefficient threshold and the rigidity risk coefficient is less than the set rigidity risk coefficient threshold, it is determined that the rigidity detection result of the pier to be measured in the current detection period is normal rigidity and the pier structure is safe.

[0048] According to an embodiment of the present invention, if the rigid risk coefficient is greater than or equal to the set rigid risk coefficient threshold (for example, 0.6), then regardless of the rigid risk coefficient, it is determined that the rigid detection result of the pier to be measured in the current detection period is severe rigid degradation, and the pier structure is dangerous, and maintenance measures need to be taken immediately. For example, traffic should be closed immediately and the emergency reinforcement program should be started. If the rigid decline coefficient is greater than or equal to the set rigid decline coefficient threshold (for example, 0.7), and the rigid risk coefficient is less than the set rigid risk coefficient threshold, it is determined that the rigid detection result of the pier to be measured in the current detection period is severe rigid degradation, and the pier structure is dangerous, and maintenance measures should be taken immediately. If the rigid decline coefficient is less than the set rigid decline coefficient threshold, and the rigid risk coefficient is less than the set rigid risk coefficient threshold, it is determined that the rigid detection result of the pier to be measured in the current detection period is normal rigidity, and the pier structure is safe, and routine monitoring can be maintained without taking maintenance measures.

[0049] The pier rigid detection method according to the embodiment of the present invention performs non-contact detection by a drone carrying a laser rangefinder, without installing equipment on the pier, reducing potential interference to the pier structure and improving the efficiency and safety of rigid detection. By measuring the distance, pitch angle, yaw angle and preset horizontal distance, the three-dimensional coordinates of the sample points to be measured of the pier to be measured can be accurately calculated, reflecting the local deformation of the pier to be measured. And by comprehensively considering the rigid decline coefficient and the rigid risk coefficient, the rigidity of the pier is comprehensively evaluated from two dimensions of geometric deformation and surface damage, improving the comprehensiveness and accuracy of rigid detection, and large-scale pier screening and long-term rigid detection can be carried out. When determining the position vectors to be measured of multiple sample points in multiple measurement areas at multiple moments in the current detection period, based on the position coordinates to be measured and the central position coordinates, the position vectors to be measured of multiple sample points in multiple measurement areas at multiple moments in the current detection period can be determined. The position vectors to be measured achieve the unification of the measurement reference, reducing the error caused by the fluctuation of the drone hovering position. For multiple measurements of the same sample point to be measured at different moments, by comparing the changes in the position vectors to be measured, the time-course law of rigid degradation can be quantified. When determining the rigid decline coefficient, based on the relative difference in the cosine similarity of the position vectors to be measured at adjacent moments, and setting weights based on the characteristics that the closer the position vector to be measured is to the horizontal direction, the more dangerous the deformation is, and the stronger the deformation characterization ability is when the sample point to be measured is farther from the central position, the relative differences in the cosine similarity of the position vectors to be measured of multiple sample points in multiple measurement areas at multiple adjacent moments in the current detection period are weighted and averaged to obtain the rigid decline coefficient, improving the accuracy of identifying the degree of rigid decline. When determining the rigid risk coefficient, the rigid risk coefficient can be determined by the similarity between the starting feature vector of the target measurement area and the ending feature vector of the target measurement area, so as to quantify the change degree of the crack feature during the detection period, judge the danger degree of the pier structure to be measured, and improve the reliability of the rigid risk coefficient.

[0050] Figure 5 A block diagram of a pier rigidity detection system according to an embodiment of the present invention is exemplarily shown. The system includes: a sample point to be measured and a central position module, which are used to set the side of the pier to be measured into four regions to be measured, set a plurality of sample points to be measured in the regions to be measured, and obtain the central positions of the regions to be measured; a module for measuring distance, pitch angle, and yaw angle, which is used to set a hovering position of a drone at a preset horizontal distance from the central position, and at multiple moments in the current detection cycle, make the laser emitted by the laser rangefinder installed on the drone irradiate on the sample points to be measured, and determine the measured distances of the multiple sample points to be measured measured by the laser rangefinder, as well as the pitch angle and yaw angle of the laser rangefinder when measuring the multiple sample points to be measured; a rigidity decrease coefficient module, which is used to determine a rigidity decrease coefficient according to the measured distance, the pitch angle, the yaw angle, and the preset horizontal distance; a to-be-measured image module, which is used to respectively obtain to-be-measured images of the regions to be measured at the start moment and the end moment of the current detection cycle; a rigidity risk coefficient module, which is used to determine a rigidity risk coefficient according to the to-be-measured images; a rigidity detection result module, which is used to determine the rigidity detection result of the pier to be measured in the current detection cycle according to the rigidity decrease coefficient and the rigidity risk coefficient.

[0051] The present invention can be a method, an apparatus, a system, and / or a computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions for performing various aspects of the present invention loaded thereon.

[0052] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the drawings are only examples and do not limit the present invention. The object of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and described in the embodiments. Without departing from the principle, the embodiments of the present invention can have any deformation or modification.

Claims

1. A method for detecting the rigidity of a bridge pier, characterized in that, Including: Set the side of the pier to be measured as four measurement areas, set multiple measurement points in the measurement areas, and obtain the central positions of the measurement areas; Set the preset horizontal distance from the central position as the UAV hovering position. At multiple moments in the current detection cycle, make the laser emitted by the laser rangefinder installed on the UAV irradiate on the measurement points to be measured, and determine the measurement distances of multiple measurement points measured by the laser rangefinder, as well as the pitch angle and yaw angle of the laser rangefinder when measuring multiple measurement points; Determine the rigid descent coefficient according to the measurement distance, the pitch angle, the yaw angle and the preset horizontal distance; At the start time and end time of the current detection cycle, respectively obtain the measurement images of the measurement areas; Determine the rigid risk coefficient according to the measurement images; Determine the rigid detection result of the pier to be measured in the current detection cycle according to the rigid descent coefficient and the rigid risk coefficient.

2. The pier rigidity detection method according to claim 1, wherein The number of measurement points to be measured in each measurement area is the same, but the positions are randomly distributed.

3. The pier rigidity detection method according to claim 1, characterized in that Determine the rigid descent coefficient according to the measurement distance, the pitch angle, the yaw angle and the preset horizontal distance, including: Taking the UAV hovering position as the origin, taking the orientation when the pitch angle of the laser rangefinder is 0° and the yaw angle is 90° as the X axis, taking the orientation when both the pitch angle and yaw angle of the laser rangefinder are 0° as the Y axis, and taking the vertical direction as the Z axis, establish a measurement point coordinate system to be measured; Obtain the central position coordinates of the central position in the measurement point coordinate system to be measured according to the preset horizontal distance; Determine the position vectors to be measured of multiple measurement points in multiple measurement areas at multiple moments in the current detection cycle according to the measurement distance, the pitch angle, the yaw angle and the central position coordinates; Determine the rigid descent coefficient according to the position vectors to be measured.

4. The pier rigidity detection method according to claim 3, characterized in that, Determine the position vectors to be measured of multiple sample points to be measured in multiple regions to be measured at multiple moments in the current detection period according to the measured distance, the pitch angle, the yaw angle, and the central position coordinates, including: According to the formula , , determine the position vector to be measured of the j-th sample point to be measured in the i-th region to be measured at the k-th moment in the current detection period , where is the measured distance of the j-th sample point to be measured in the i-th region to be measured at the k-th moment in the current detection period, is the pitch angle of the j-th sample point to be measured in the i-th region to be measured at the k-th moment in the current detection period, and the unit of the pitch angle is degree, is the yaw angle of the j-th sample point to be measured in the i-th region to be measured at the k-th moment in the current detection period, and the unit of the yaw angle is degree, is the position coordinate to be measured of the j-th sample point to be measured in the i-th region to be measured at the k-th moment in the current detection period, is the preset horizontal distance, is the central position coordinate, and i, j, and k are all positive integers.

5. The pier rigidity detection method according to claim 4, characterized in that, Determine the rigid descent coefficient according to the position vectors to be measured, including: Obtain the width and height of the pier to be measured; Determine the rigid descent coefficient according to the position vectors to be measured, the width and the height.

6. The pier rigidity detection method according to claim 5, wherein Determine the rigid descent coefficient according to the to-be-detected position vector, the width, and the height, including: According to the formula , , , determine the rigid descent coefficient G, where is the cosine similarity between the to-be-detected position vector of the j-th to-be-detected sample point in the i-th to-be-detected area at the k-th moment in the current detection period and the to-be-detected position vector of the j-th to-be-detected sample point in the i-th to-be-detected area at the (k - 1)-th moment in the current detection period, is the to-be-detected position vector of the j-th to-be-detected sample point in the i-th to-be-detected area at the (k - 1)-th moment in the current detection period, is the cosine similarity between the to-be-detected position vector of the j-th to-be-detected sample point in the i-th to-be-detected area at the (k + 1)-th moment in the current detection period and the to-be-detected position vector of the j-th to-be-detected sample point in the i-th to-be-detected area at the k-th moment in the current detection period, is the to-be-detected position vector of the j-th to-be-detected sample point in the i-th to-be-detected area at the (k + 1)-th moment in the current detection period, is the abscissa of the to-be-detected position coordinate of the j-th to-be-detected sample point in the i-th to-be-detected area at the k-th moment in the current detection period, H is the height of the to-be-detected bridge pier, W is the width of the to-be-detected bridge pier, 4 is the number of to-be-detected areas, N is the number of to-be-detected sample points in each to-be-detected area, M is the number of detection period moments, i ≤ 4, j ≤ N, k ≤ M, and i, j, k, N, and M are all positive integers, and if is a conditional function.

7. The pier rigidity detection method according to claim 1, characterized in that Determine the rigid risk coefficient according to the measurement images, including: In the measurement images, identify whether there are cracks in the measurement areas through an image detection model; If there are cracks in the measurement areas, determine the target measurement areas with cracks; Through the trained image recognition neural network model, perform crack feature extraction processing on the measurement image of the target measurement area at the start time of the current detection cycle to obtain the start target measurement area feature vector; Through the trained image recognition neural network model, perform crack feature extraction processing on the measurement image of the target measurement area at the end time of the current detection cycle to obtain the end target measurement area feature vector; Determine the rigid risk coefficient according to the start target measurement area feature vector and the end target measurement area feature vector; If there are no cracks in the measurement areas, determine that the rigid risk coefficient is 0.

8. The pier rigidity detection method according to claim 7, characterized in that Determine the rigidity risk coefficient according to the starting target area to be measured feature vector and the ending target area to be measured feature vector, including: according to the formula , determine the rigidity risk coefficient F, where is the starting target area to be measured feature vector of the s-th target area to be measured, is the ending target area to be measured feature vector of the s-th target area to be measured, n is the number of target areas to be measured, s ≤ n, and both s and n are positive integers.

9. The pier rigidity detection method according to claim 1, characterized in that Determine the rigidity detection result of the pier to be measured in the current detection period according to the rigidity decline coefficient and the rigidity risk coefficient, including: if the rigidity risk coefficient is greater than or equal to the set rigidity risk coefficient threshold, determine that the rigidity detection result of the pier to be measured in the current detection period is serious rigidity degradation and the pier structure is dangerous; if the rigidity decline coefficient is greater than or equal to the set rigidity decline coefficient threshold and the rigidity risk coefficient is less than the set rigidity risk coefficient threshold, determine that the rigidity detection result of the pier to be measured in the current detection period is serious rigidity degradation and the pier structure is dangerous; if the rigidity decline coefficient is less than the set rigidity decline coefficient threshold and the rigidity risk coefficient is less than the set rigidity risk coefficient threshold, determine that the rigidity detection result of the pier to be measured in the current detection period is normal rigidity and the pier structure is safe.

10. A pier rigidity detection system for performing the pier rigidity detection method according to any one of claims 1-9, characterized in that, Including: The sample point to be measured and the central position module are used to set the side of the pier to be measured as four measurement areas, set multiple sample points to be measured in the measurement areas, and obtain the central positions of the measurement areas; the measurement distance, pitch angle and yaw angle module is used to set the preset horizontal distance at the central position as the UAV hovering position, at multiple moments in the current detection period, make the laser emitted by the laser rangefinder set on the UAV irradiate the sample points to be measured, and determine the measurement distances of multiple sample points to be measured measured by the laser rangefinder, as well as the pitch angle and yaw angle of the laser rangefinder when measuring multiple sample points to be measured; The rigidity decline coefficient module is used to determine the rigidity decline coefficient according to the measurement distance, the pitch angle, the yaw angle and the preset horizontal distance; The image to be measured module is used to obtain the images to be measured of the measurement areas at the start and end moments of the current detection period respectively; The rigidity risk coefficient module is used to determine the rigidity risk coefficient according to the image to be measured; The rigidity detection result module is used to determine the rigidity detection result of the pier to be measured in the current detection period according to the rigidity decline coefficient and the rigidity risk coefficient.

Images