Bridge pier surface and hole site measuring method based on unmanned aerial vehicle approaching photogrammetry

Abstract

The invention relates to a pier surface and hole site measurement method based on unmanned aerial vehicle approaching photogrammetry, and belongs to the technical field of engineering surveying, the method comprises the following steps: carrying out three-dimensional reconstruction based on an aerial image and an aerial triangulation measurement result, and obtaining a target model of a to-be-measured pier; the aerial image is obtained by performing image acquisition on the to-be-measured pier through the unmanned aerial vehicle; the aerial triangulation result is determined based on the aerial image; identifying a hole site in the target model based on a preset local window, and determining a reference point; and determining hole site information and surface information of the to-be-measured pier based on the relative position of the reference point in the preset local window. According to the pier surface and hole site measurement method based on unmanned aerial vehicle approaching photogrammetry, the problems of repeated climbing of personnel, reworking of the prefabricated member and the like are avoided, higher measurement precision, data density and operation efficiency are provided, and the method can serve as a mode for obtaining information of the prefabricated member of the pier and is popularized in bridge construction.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering measurement, and in particular to a method for measuring bridge pier surfaces and hole positions based on unmanned aerial vehicle (UAV) close-up photography measurement. Background Art

[0002] With the development of information technology, some digital and automated means have gradually emerged in the field of engineering construction. However, the current identification schemes for the surface shape of piers and the deviation of prefabricated hole positions are still relatively backward. Prefabricated parts such as handrails can only be produced uniformly according to design specifications. When mismatches are found during manual installation on site, feedback is required to redo the work, resulting in rework and low efficiency.

[0003] To achieve prior identification of pier surfaces and hole positions, methods such as total station measurement, lidar scanning, and photogrammetry can be used. Among them, the total station measurement method can only obtain a small number of discrete points. At the same time, due to the depression of the hole position, it is impossible to correctly obtain depth information and the efficiency is low; the scanning height of the station-type lidar is limited, and it is impossible to measure the higher parts of the pier; the airborne radar is also restricted by the scanning angle and alignment accuracy, and the hole position identification based on the point cloud is more difficult than that based on the image; the traditional oblique photogrammetry method has a long shooting distance and is mainly used for medium and low resolution large survey area three-dimensional model reconstruction or small and medium scale topographic map acquisition. Its typical resolution is 3-5cm, which is difficult to meet the needs of precise measurement of pier surfaces and hole positions.

[0004] The close-range photogrammetry method can effectively avoid the above constraints such as data density and accuracy, and obtain a high-resolution, high-relative-accuracy three-dimensional model. However, due to its texture characteristics, the model is not smooth near the hole position, so it is difficult to directly extract the precise hole position coordinates from the model. In addition, this method is also easily affected by factors such as uneven lighting. Summary of the invention

[0005] In view of this, it is necessary to provide a method for measuring the pier surface and hole position based on UAV close-up photogrammetry to solve the problems of low accuracy and efficiency of existing measurement methods.

[0006] In order to solve the above problems, the present invention provides a method for measuring the surface and hole position of a pier based on close-up photogrammetry using an unmanned aerial vehicle, comprising: Perform three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured; the aerial image is obtained by collecting images of the bridge pier to be measured by a drone; the aerial triangulation result is determined based on the aerial image; Identify the hole positions in the target model based on a preset local window and determine the reference points; Based on the relative position of the reference point within the preset local window, the hole position information and surface information of the pier to be measured are determined.

[0007] In a possible implementation, identifying the hole position in the target model based on a preset local window and determining the reference point includes: Identify the hole position in the preset local window based on the target detection method; The reference point is determined based on the coordinates of the center of the hole position in the preset local window.

[0008] In a possible implementation, determining the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window includes: The relative position is calibrated on the target model to obtain the calibrated position of the reference point: Based on the calibrated position, extracting the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model; Determine the hole position information based on the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model; the hole position information includes the three-dimensional coordinates of the hole position center in the coordinate system corresponding to the target model; The three-dimensional coordinates of the reference point in the corresponding coordinate system of the target model are fitted to obtain the surface information.

[0009] In a possible implementation, determining the hole position information based on the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model includes: Based on the three-dimensional coordinates of the reference point in the corresponding coordinate system of the target model, a distance weighted interpolation method based on least squares or a surface fitting method is used to optimally estimate the coordinates of the hole center to determine the hole information.

[0010] In a possible implementation, after determining the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window, the method further includes: Align the three-dimensional coordinates of the hole center in the coordinate system corresponding to the target model with the coordinates of the prefabricated hole center in the preset solution, and determine the relative relationship between the coordinate system corresponding to the target model and the coordinate system corresponding to the preset solution; Based on the relative relationship, a difference between the prefabricated hole position and the hole position is determined.

[0011] In a possible implementation, before performing three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured, the method further includes: Based on the shape and position distribution of the bridge pier to be measured, the preset route of the drone is determined; the preset route includes: a full-area controlled tilt photography route, a complete column-surrounding photography route, and a high-density photography route close to key parts; The aerial image is acquired by collecting images of the bridge pier to be measured according to the preset route using a drone.

[0012] In a possible implementation, before performing three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured, the method further includes: Preprocessing the aerial image to obtain POS information; Perform aerial triangulation based on the aerial image and the POS information to obtain the aerial triangulation result.

[0013] The present invention also provides a bridge pier surface and hole position measurement device based on drone close-up photogrammetry, comprising: The reconstruction module is used to perform three-dimensional reconstruction based on aerial images and aerial triangulation results to obtain the target model of the bridge pier to be measured; A recognition module, used to recognize the hole positions in the target model based on a preset local window and determine the reference points; The determination module is used to determine the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window.

[0014] On the other hand, the present invention also provides an electronic device, including a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the bridge pier surface and hole position measurement method based on drone close-up photogrammetry described in any of the above implementations.

[0015] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for measuring the pier surface and hole position based on drone close-up photogrammetry as described in any of the above-mentioned implementations.

[0016] The beneficial effects of the present invention are as follows: the bridge pier surface and hole position measurement method based on unmanned aerial vehicle close-up photogrammetry provided by the present invention can obtain all-round and multi-angle aerial images of the bridge pier by using the unmanned aerial vehicle to collect images of the bridge pier to be measured, and obtain aerial triangulation results by performing aerial triangulation based on the aerial images, so as to obtain the target model of the bridge pier to be measured, and identify the hole position according to the preset local window to determine multiple reference points near the hole position, so as to obtain the hole position information and surface information of the bridge pier to be measured according to the relative position of the reference point in the preset local window, and combine the flexibility of the unmanned aerial vehicle with the precision of photogrammetry, so as to realize fast and accurate measurement of the bridge pier surface and hole position, which not only improves the measurement efficiency, but also reduces the risk and cost of manual measurement, and compared with the traditional manual measurement method or other measurement methods, it avoids the problems of repeated climbing of personnel and rework of prefabricated parts, provides higher measurement accuracy, data density and operation efficiency, and can be used as a method for obtaining information of bridge pier prefabricated parts and promoted in bridge construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 One of the method flow charts of an embodiment of the method for measuring the bridge pier surface and hole position based on UAV close-up photogrammetry provided by the present invention; Figure 2 A second method flow chart of an embodiment of a method for measuring bridge pier surface and hole position based on UAV close-up photogrammetry provided by the present invention; Figure 3 A schematic diagram of a preset route provided by the present invention; Figure 4 The overall texture map of the top of the bridge pier provided by the present invention; Figure 5 The overall structural diagram of the top of the bridge pier provided by the present invention; Figure 6 A local texture map of the top of the bridge pier provided by the present invention; Figure 7 A partial structural diagram of the top of the bridge pier provided by the present invention; Figure 8 A schematic diagram of the observation window and reference points provided by the present invention; Fig. 9 A schematic structural diagram of an embodiment of a bridge pier surface and hole position measurement device based on drone close-up photogrammetry provided by the present invention; Fig.10 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] In the description of the embodiments of the present invention, unless otherwise specified, "multiple" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" may mean: A exists alone, A and B exist at the same time, and B exists alone.

[0021] The descriptions of "first", "second", etc. involved in the embodiments of the present invention are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the technical features defined as "first" or "second" may explicitly or implicitly include at least one of the features.

[0022] Figure 1 One of the method flow charts of an embodiment of the method for measuring the bridge pier surface and hole position based on UAV close-up photogrammetry provided by the present invention is as follows: Figure 1 As shown in the figure, the bridge pier surface and hole position measurement method based on UAV close-up photogrammetry includes: S101, performing three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain a target model of the bridge pier to be measured; the aerial image is obtained by collecting images of the bridge pier to be measured by a drone; and the aerial triangulation result is determined based on the aerial image; S102, identifying the hole positions in the target model based on a preset local window and determining a reference point; S103: Determine the hole position information and surface information of the pier to be measured based on the relative position of the reference point in the preset local window.

[0023] In S101, three-dimensional reconstruction is performed based on the aerial image and the aerial triangulation result to obtain a target model of the bridge pier to be measured.

[0024] Drones can be used to collect images of the bridge piers to be measured. The flight path and altitude of the drone can be planned in advance to ensure that all-round and multi-angle images of the bridge piers can be captured.

[0025] Aerial triangulation based on the collected aerial images involves matching the feature points in the images and calculating the three-dimensional coordinates of these feature points to construct an accurate three-dimensional model of the bridge pier to be measured, namely the target model.

[0026] In S102, the hole positions in the target model are identified based on a preset local window to determine reference points.

[0027] In order to accurately identify the hole position in the target model, a preset local window can be set. The size and position of the preset local window can be determined according to the actual situation of the bridge pier and the distribution characteristics of the hole position. The preset local window can be used to identify the holes in the target model. For example, an image detection method based on deep learning can be used for hole position identification. After the hole position is identified, several reference points can be automatically searched near the hole position.

[0028] In S103, based on the relative position of the reference point in the preset local window, the hole position information and the surface information of the pier to be measured are determined.

[0029] The relative positions of all reference points in the preset local window are recorded, and the calibrated positions of the reference points are calibrated in the three-dimensional model, so as to extract the three-dimensional coordinates of the reference points in the corresponding coordinate system of the three-dimensional model, thereby determining the three-dimensional coordinates of the hole center in the corresponding coordinate system of the target model, that is, obtaining the hole information, and fitting the three-dimensional coordinates of the reference points in the corresponding coordinate system of the target model to obtain the shape information, that is, obtaining the surface information.

[0030] Compared with the prior art, the method for measuring the surface and hole position of a pier based on close-up photogrammetry of an unmanned aerial vehicle provided in an embodiment of the present invention can obtain all-round and multi-angle aerial images of the pier by using an unmanned aerial vehicle to collect images of the pier to be measured, and obtain aerial triangulation results by performing aerial triangulation based on the aerial images, thereby obtaining a target model of the pier to be measured, and identifying the hole position according to a preset local window to determine multiple reference points near the hole position, thereby obtaining the hole position information and surface information of the pier to be measured according to the relative positions of the reference points in the preset local window. Combining the flexibility of the unmanned aerial vehicle with the precision of photogrammetry, rapid and accurate measurement of the surface and hole position of the pier can be achieved, which not only improves the efficiency of measurement, but also reduces the risk and cost of manual measurement. Compared with traditional manual measurement methods or other measurement methods, it avoids problems such as repeated climbing by personnel and rework of prefabricated parts, provides higher measurement accuracy, data density and operation efficiency, and can be used as a method for obtaining information on prefabricated parts of piers and promoted in bridge construction.

[0031] In some embodiments of the present invention, before performing three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured, the method further includes: Based on the shape and position distribution of the bridge pier to be measured, the preset route of the drone is determined; the preset route includes: a full-area controlled tilt photography route, a complete column-surrounding photography route, and a high-density photography route close to key parts; The aerial image is acquired by collecting images of the bridge pier to be measured according to the preset route using a drone.

[0032] According to the shape and location distribution of the bridge piers to be measured, preset drone routes with multiple levels and shooting densities are planned.

[0033] For example, according to the shape and location distribution of the bridge piers, the following three levels of preset routes can be designed: (1) Control the oblique photography route in the entire area.

[0034] According to the camera parameters and the height of the bridge piers, the oblique photogrammetry route is arranged with the requirements of no less than 80% overlap in the heading direction and no less than 70% overlap in the lateral direction, so as to achieve full coverage of the surrounding features of the survey area in the shooting area, so as to ensure the integrity and continuity of the model of the entire survey area. Due to the different heights of the bridge piers, the resolution requirements of the oblique route are relatively low, and the ground resolution is usually within 3-5cm.

[0035] (2) Complete cylindrical photography route.

[0036] Design a circling route based on the specific shape of each column, so that the shooting angle covers the plane of the column, and the shooting range covers the surface of the column and the bottom of the ground, to ensure the integrity of each column model. The circling route has higher requirements for resolution, and the ground resolution should usually be less than 1cm. It can also be adjusted when the column is higher.

[0037] (3) Key areas are close to high-density photography routes.

[0038] For the areas of interest on the surface of the column, such as near the prefabricated holes and the column number marks, close-up photography is used for high-density shooting to ensure the fineness of the key areas of the model. Close-up photography is usually within 10m from the column, and only focuses on whether the shooting of the area of ​​interest is fine and whether the coverage is sufficient, without considering issues such as ground connection. This step has the highest resolution requirements, but the shooting content may not include the ground, so the focus is on the resolution relative to the surface of the column, which should usually be within 2mm and no more than 3mm at most.

[0039] Control the drone to reach the designated location according to the preset route, perform designated actions, and obtain multi-level and multi-resolution aerial images.

[0040] The method for measuring the pier surface and hole position based on the close-up photogrammetry of the drone provided in the embodiment of the present invention determines the preset route of the drone according to the shape and position distribution of the pier to be measured. Through the above route design, it is possible to achieve data coverage of the entire measurement area and the entire column, and obtain sufficient high-resolution images, thereby ensuring the balance between the integrity and precision of the model. In addition, when laying out the route, the height, shape of the column and the resolution of the camera used should be fully considered to ensure that the overlap of the captured images is sufficient and that there is no missing shooting angle.

[0041] In some embodiments of the present invention, before performing three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured, the method further includes: Preprocessing the aerial image to obtain POS information; Perform aerial triangulation based on the aerial image and the POS information to obtain the aerial triangulation result.

[0042] Preprocessing can include airborne positioning and attitude data processing and image optical consistency processing. After acquiring the aerial image, airborne positioning and attitude data processing is performed to obtain POS information, which refers to the shooting position and attitude information of each image.

[0043] The aerial image is processed for optical consistency, including the following steps: (1) Adjust the exposure value of the image to reduce the brightness difference between the backlight area and the bright light area; (2) Unify the color style between images and reduce the differences in color temperature, white balance, and channel components; (3) Adjust the contrast and geometric distortion of the image to correctly highlight the features of the photographed object.

[0044] Through the above adjustment methods, the image content can be correct and the style can be unified to avoid structural misrecognition or texture dislocation during modeling, and to ensure the measurable properties and visual effects of the model. The photos can be pre-classified and then processed in batches.

[0045] Aerial triangulation is performed based on the adjusted original images and POS data, and the image control network is adjusted to obtain the best posture state of each image and the optimal estimation of camera parameters.

[0046] Therefore, three-dimensional reconstruction is performed based on aerial triangulation and the original aerial images to obtain a high-precision model of the bridge pier to be measured.

[0047] In some embodiments of the present invention, the identifying the hole position in the target model based on a preset local window and determining the reference point includes: Identify the hole position in the preset local window based on the target detection method; The reference point is determined based on the coordinates of the center of the hole position in the preset local window.

[0048] An appropriate observation window, i.e., a preset local window, can be selected near each group of hole positions in the target model, and the hole position information can be identified within the window and a number of reference points can be automatically generated.

[0049] Exemplarily, the method for selecting the preset local window and reference point is as follows: (1) The observation line of sight should be located in the vertical plane and approximately perpendicular to the local surface of the column to reduce the perspective effect of objects being larger near and smaller far away; (2) The holes in the model should be symmetrically distributed about the center of the window, and the line connecting the upper and lower boundaries should not exceed 2 / 3 of the height of the window to ensure the stability of point recognition; (3) Select a deep learning-based target detection method such as the YOLO series or the RCNN series to identify the hole positions in the window and extract the coordinates of the hole center in the two-dimensional window; (4) Based on the recognition results, several reference points are automatically searched near the hole positions. The reference points should be located in the relatively flat area around the hole positions on the column facade, evenly distributed and with high density. There should be at least 8 points around each hole position and at least 5 points on the line connecting every two holes. The relative positions of all points in the 2D window are recorded and calibrated in the 3D model.

[0050] In some embodiments of the present invention, determining the hole position information and surface information of the pier to be measured based on the relative position of the reference point in the preset local window includes: The relative position is calibrated on the target model to obtain the calibrated position of the reference point: Based on the calibrated position, extracting the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model; Determine the hole position information based on the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model; the hole position information includes the three-dimensional coordinates of the hole position center in the coordinate system corresponding to the target model; The three-dimensional coordinates of the reference point in the corresponding coordinate system of the target model are fitted to obtain the surface information.

[0051] In some embodiments of the present invention, determining the hole position information based on the three-dimensional coordinates of the reference point in the corresponding coordinate system of the target model includes: Based on the three-dimensional coordinates of the reference point in the corresponding coordinate system of the target model, a distance weighted interpolation method based on least squares or a surface fitting method is used to optimally estimate the coordinates of the hole center to determine the hole information.

[0052] Based on the model data corresponding to the two-dimensional window image, the three-dimensional coordinates of the reference point are extracted to calculate the coordinates of the hole center position and extract the surface information of the cylinder.

[0053] Based on the calibrated position of the reference point obtained in the previous step, the three-dimensional coordinates of all points in the model in the model independent coordinate system are extracted. Combined with the position relationship in the two-dimensional window, the hole center coordinates are optimally estimated using least squares-based distance weighted interpolation or surface fitting methods to obtain the three-dimensional coordinates of the hole center in the corresponding coordinate system of the target model.

[0054] The reason for doing this is that the holes in the model often have protrusions or depressions, making it difficult to directly collect the correct coordinates of the hole center. The surrounding reference points are relatively flat and can be used to express the cylindrical shape as well as to fit and calculate the hole center coordinates.

[0055] In some embodiments of the present invention, after determining the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window, the method further includes: Align the three-dimensional coordinates of the hole center in the coordinate system corresponding to the target model with the coordinates of the prefabricated hole center in the preset solution, and determine the relative relationship between the coordinate system corresponding to the target model and the coordinate system corresponding to the preset solution; Based on the relative relationship, a difference between the prefabricated hole position and the hole position is determined.

[0056] The points measured in the model coordinate system are aligned with the center coordinates of the prefabricated holes in the preset scheme to obtain the relative relationship between the two coordinate frames, that is, the relative relationship between the corresponding coordinate system of the target model and the corresponding coordinate system of the preset scheme.

[0057] According to the coordinates of the hole center position obtained in the previous step, the corresponding hole position coordinates are found in the pier construction design drawing. The scale parameters of the two coordinate systems are the same, but there is relative rotation and translation. Therefore, a rotation + translation mathematical model is established for the two coordinate systems. According to the two sets of coordinate sequences with the same name obtained, the parameters are estimated based on the least squares principle, so as to realize the alignment of the two coordinate frames and facilitate the calculation of the hole position deviation. Certain constraints or anti-error models can be added during the estimation to avoid the hole position with a large deviation from the design position affecting the registration result.

[0058] Based on the registered coordinate frame, the difference between the actual position and the designed position of each prefabricated hole is counted, and it is determined whether the deviation value exceeds the limit.

[0059] Therefore, prefabrication suggestions are provided based on the deviation value, column shape information is provided based on the coordinates of key points on the column surface, hole distribution information is provided based on the hole deviation statistics, and design suggestions for prefabricated parts are provided based on the above information.

[0060] Exemplarily, prefabrication suggestions may be provided as follows: (1) Based on the point measurement and coordinate registration in the previous steps, the three-dimensional coordinates of each prefabricated hole on the column in the preset scheme coordinate system and its deviation value relative to the designed position can be obtained; (2) By fitting the coordinates of the reference points in the region of interest, the surface shape information of the region can be obtained; (3) Based on the hole position deviation and cylindrical shape, provide reference manufacturing dimensions and shapes for prefabricated parts such as hanging rails.

[0061] The embodiment of the present invention provides a method for measuring the pier surface and hole position based on close-up photogrammetry using an unmanned aerial vehicle, and proposes a method for measuring the cylindrical shape and hole position deviation based on an independent cylindrical coordinate system. Since the focus is on relative position and relative shape, there is no need to rely on external control points or target points for absolute coordinate correction, and millimeter-level relative information measurement can be achieved on a high-precision model.

[0062] The pier surface and hole position measurement method based on UAV close-up photogrammetry provided in the embodiment of the present invention avoids the problems of repeated climbing by personnel, rework of prefabricated parts, etc. compared with traditional manual measurement methods or other measurement methods, and provides higher measurement accuracy, data density and operation efficiency. It can be used as a method for obtaining information on pier prefabricated parts and promoted in bridge construction.

[0063] The embodiment of the present invention is based on drone close-up photogrammetry technology and image recognition technology. The bridge pier surface and hole position measurement method based on drone close-up photogrammetry provided by the present invention is described in detail below in conjunction with specific scenarios. Figure 2 The second method flow chart of an embodiment of the method for measuring the bridge pier surface and hole position based on UAV close-up photogrammetry provided by the present invention is as follows: Figure 2 The following figure shows a typical implementation process, which includes the following steps: Step 1: Route planning: According to the shape and location distribution of the bridge piers to be measured, three levels of drone preset routes are planned. Figure 3 A schematic diagram of a preset route provided by the present invention, such as Figure 3 As shown, from top to bottom are the inclined route, the circling route and the close route, and their design principles have been introduced in the invention content.

[0064] In the embodiment of the present invention: the height of the column is about 10m; The relative altitude of the inclined route is 100m, and the ground resolution is about 1.3cm; The relative altitude of the circumnavigation route is 30m, and the ground resolution is about 3.8mm; The distance between the close flight line and the pier surface varies from about 5 to 10 meters, corresponding to a resolution of 0.6 to 1.2 mm on the pier surface. All of the above meet the requirements for image resolution.

[0065] Step 2: Implementation of flight mission.

[0066] According to the above preset route, control the drone to the designated location for shooting. To ensure image quality, the best choice is cloudy weather with sufficient light to avoid weak texture in the shadow area. The second best choice is around noon on a sunny day to reduce the shadow area caused by the sun's altitude angle.

[0067] Step 3: Data preprocessing.

[0068] Step 3.1: Based on the airborne positioning and attitude data, obtain the POS information corresponding to each image based on RTK or PPK technology.

[0069] Step 3.2: Check and adjust the brightness, contrast, and hue of the drone image to make the target clear, the gloss even, reduce the color difference between photos, and highlight the target features.

[0070] Step 4: Aerial triangulation.

[0071] A unified block is established for the multiple cylindrical continuous areas in the embodiment, and aerial triangulation is performed on all original images in the block in combination with POS information and camera parameters. The bundle method is used for regional network adjustment to obtain the best posture state of each image and the optimal estimation of camera parameters.

[0072] Step 5: Modeling.

[0073] Perform high-precision 3D reconstruction on the designated area within the block. Due to the high image density, the memory usage of unified modeling is too large. Therefore, the modeling can be divided into partitions or cylinders, and the model quality can be checked. Figures 4 to 7 A high-resolution 3D model of the top of a bridge pier in the embodiment is shown. Figure 4 The overall texture map of the top of the bridge pier provided by the present invention, Figure 5 The overall structural diagram of the top of the bridge pier provided by the present invention is as follows: Figure 4 and 5 Corresponding to the same area. Figure 6 The local texture map of the top of the bridge pier provided by the present invention, Figure 7 A partial structural diagram of the top of the bridge pier provided by the present invention, Figure 6 and 7 Corresponding to the same area.

[0074] There are many aerial triangulation and modeling software on the market, such as DJI Zhitu, ContextCapture, Reconstruction Master, Metashape, etc. As long as the model effect is normal, they can be used. In the embodiment, Reconstruction Master is selected, which can automatically complete the image optical consistency adjustment in step 3.2 and simplify the process.

[0075] Step 6: Select the local window and reference point.

[0076] An appropriate observation window is selected near each group of hole positions in the model, and a deep learning algorithm is called within the window to identify the hole position information, and several reference points are automatically generated around it.

[0077] In the embodiment, the YOLOv8 model is selected, and two parts and three loss functions are mainly used for the target detection task: 1. Regression loss.

[0078] (1) Distribution Focal Loss (DFL): calculates the offset from the center of the anchor point to the upper left corner and the lower right corner.

[0079]

[0080] in, , is the predicted value and the near-prediction value output by the network, , , are the actual value, integral value and adjacent integral value of the label. The specific conversion process is: y = the distance from the center to a certain edge / the current downsampling multiple.

[0081] (2) IoU Loss positioning loss, using CIoU loss, only calculates the positioning loss of positive samples, which is a metric to measure the degree of overlap between the predicted box and the true box.

[0082]

[0083]

[0084]

[0085] In the formula, and Represent the forecast and actual situation respectively. , and denote the center, width, and height of the corresponding bounding box, respectively. is the diagonal distance of the enclosing bounding box, is the Euclidean distance between the coordinates of the two box center points, is the original Intersection over Union.

[0086] 2. Classification loss.

[0087] (1) The classification loss uses binary cross entropy BCE loss (Binary CrossEntropy Loss), which only calculates the classification loss of positive samples;

[0088] In the formula, is the sample size, It is The true labels of samples, The model predicts The probability that a sample is a positive class, BCE Loss can measure the gap between the probability distribution predicted by the model and the actual observed label.

[0089] Figure 8 A schematic diagram of the observation window and reference points provided by the present invention, such as Figure 8 As shown, the model structure near all reference points was checked and confirmed to be flat.

[0090] Step 7: Measure the target point.

[0091] Based on the model data corresponding to the two-dimensional window image, the three-dimensional coordinates of the reference point are extracted, and then the coordinates of the center position of the hole are estimated by using distance weighted interpolation or surface fitting methods. This process is repeated until every group of hole positions on each cylinder is measured.

[0092] A distance interpolation method used in the embodiment is:

[0093] In the formula, etc. represent a set of hole center coordinates obtained by interpolation of a pair of reference points. To assign weights to them, they can be weighted according to distance, and the coordinates of a hole position are obtained by at least 4 pairs of reference points in different directions.

[0094] The estimation criteria for the registration parameters are:

[0095] The parameter calculation steps include: (1) Calculate the centroid of each of the two sets of point clouds and , and translate to the centroid coordinate system; (2) Calculate the covariance matrix:

[0096] (3) Yes The matrix is ​​decomposed by SVD to obtain and :

[0097] (4) Estimation of rotation matrix:

[0098] (5) Estimated translation:

[0099] If there are large deviations in hole positions during implementation, an outlier elimination strategy can be used to eliminate gross errors.

[0100] Step 9, point deviation statistics: Based on the alignment result of the coordinate frame, the collected coordinates of each prefabricated hole are converted into the design coordinate system, the difference between it and the design coordinate is obtained, and it is determined whether the deviation value exceeds the limit. The judgment method needs to be formulated in combination with actual needs. The judgment standard in the embodiment is 1 cm.

[0101] Step 10: Provide prefabrication suggestions: Based on the statistical results of hole position deviation, we can provide production suggestions for prefabricated parts such as hanging rails. For holes with deviation values ​​exceeding the limit, we can provide the actual shape between the two holes in combination with the reference points on the column surface.

[0102] The present invention combines UAV close-range photogrammetry technology, computer vision 3D reconstruction technology and deep learning-based image recognition technology to propose a method for measuring the surface shape of bridge piers and prefabricated hole positions. Its direct purpose is to optimize the production and installation process of bridge pier prefabricated parts, avoid repeated climbing caused by manual trial and error, and improve construction efficiency. At the same time, it provides high-precision and high-density data on the surface shape of bridge piers, which can be used for bridge pier construction quality inspection, etc.

[0103] In order to better implement the bridge pier surface and hole position measurement method based on drone close-up photogrammetry in the embodiment of the present invention, on the basis of the bridge pier surface and hole position measurement method based on drone close-up photogrammetry, the embodiment of the present invention also provides a bridge pier surface and hole position measurement device based on drone close-up photogrammetry. Fig. 9 The schematic diagram of the structure of an embodiment of the bridge pier surface and hole position measurement device based on drone close-up photogrammetry provided by the present invention is as follows: Fig. 9 As shown, the bridge pier surface and hole position measurement device 900 based on drone close-up photography measurement includes: A reconstruction module 910 is used to perform three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain a target model of the bridge pier to be measured; An identification module 920 is used to identify the hole positions in the target model based on a preset local window and determine the reference points; The determination module 930 is used to determine the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window.

[0104] The pier surface and hole position measurement device 900 based on drone close-up photogrammetry provided in the above embodiment can realize the technical solution described in the above embodiment of the pier surface and hole position measurement method based on drone close-up photogrammetry. The specific implementation principles of the above modules or units can refer to the corresponding contents in the embodiment of the pier surface and hole position measurement method based on drone close-up photogrammetry, which will not be repeated here.

[0105] like Fig.10 As shown, the present invention also provides an electronic device 1000. The electronic device 1000 includes a processor 1001, a memory 1002 and a display 1003. Fig.10 Only some components of the electronic device 1000 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0106] In some embodiments, the processor 1001 can be a central processing unit (CPU), a microprocessor or other data processing chip, which is used to run the program code or process data stored in the memory 1002, such as the bridge pier surface and hole position measurement method based on drone close-up photogrammetry in the present invention.

[0107] In some embodiments, processor 1001 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, processor 1001 may be local or remote. In some embodiments, processor 1001 may be implemented in a cloud platform. In some embodiments, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0108] In some embodiments, the memory 1002 may be an internal storage unit of the electronic device 1000, such as a hard disk or memory of the electronic device 1000. In other embodiments, the memory 1002 may also be an external storage device of the electronic device 1000, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 1000.

[0109] Furthermore, the memory 1002 may include both an internal storage unit of the electronic device 1000 and an external storage device. The memory 1002 is used to store application software installed in the electronic device 1000 and various data.

[0110] In some embodiments, the display 1003 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an organic light-emitting diode (OLED) touch device, etc. The display 1003 is used to display information on the electronic device 1000 and to display a visual user interface. The components 1001-1003 of the electronic device 1000 communicate with each other via a system bus.

[0111] In one embodiment, when the processor 1001 executes the bridge pier surface and hole position measurement program based on drone close-up photogrammetry in the memory 1002, the following steps may be implemented: Perform three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured; the aerial image is obtained by collecting images of the bridge pier to be measured by a drone; the aerial triangulation result is determined based on the aerial image; Identify the hole positions in the target model based on a preset local window and determine the reference points; Based on the relative position of the reference point within the preset local window, the hole position information and surface information of the pier to be measured are determined.

[0112] It should be understood that when the processor 1001 executes the bridge pier surface and hole position measurement program based on drone close-up photogrammetry in the memory 1002, in addition to the above functions, other functions can also be realized. For details, please refer to the description of the corresponding method embodiment above.

[0113] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 1000 mentioned, and the electronic device 1000 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic devices may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 1000 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0114] Correspondingly, an embodiment of the present invention also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, it can implement the steps or functions of the bridge pier surface and hole position measurement method based on drone close-up photogrammetry provided in the above-mentioned method embodiments.

[0115] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0116] The above is a detailed introduction to the bridge pier surface and hole position measurement method based on UAV close-up photogrammetry provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for measuring bridge pier surface and hole position based on drone close-up photogrammetry, characterized in that: include: Perform three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured; the aerial image is obtained by collecting images of the bridge pier to be measured by a drone; the aerial triangulation result is determined based on the aerial image; Identify the hole positions in the target model based on a preset local window and determine the reference points; Based on the relative position of the reference point within the preset local window, the hole position information and surface information of the pier to be measured are determined.

2. The method for measuring bridge pier surface and hole position based on UAV close-up photogrammetry according to claim 1 is characterized in that: The identifying the hole position in the target model based on the preset local window and determining the reference point includes: Identify the hole position in the preset local window based on the target detection method; The reference point is determined based on the coordinates of the center of the hole position in the preset local window.

3. The method for measuring bridge pier surface and hole position based on UAV close-up photogrammetry according to claim 1 is characterized in that: The determining the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window includes: The relative position is calibrated on the target model to obtain the calibrated position of the reference point: Based on the calibrated position, extracting the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model; Determine the hole position information based on the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model; the hole position information includes the three-dimensional coordinates of the hole position center in the coordinate system corresponding to the target model; The three-dimensional coordinates of the reference point in the corresponding coordinate system of the target model are fitted to obtain the surface information.

4. The method for measuring bridge pier surface and hole position based on UAV close-up photogrammetry according to claim 3 is characterized in that: The determining the hole position information based on the three-dimensional coordinates of the reference point in the coordinate system corresponding to the target model includes: Based on the three-dimensional coordinates of the reference point in the corresponding coordinate system of the target model, a distance weighted interpolation method based on least squares or a surface fitting method is used to optimally estimate the coordinates of the hole center to determine the hole information.

5. The method for measuring bridge pier surface and hole position based on UAV close-up photogrammetry according to claim 3 is characterized in that: After determining the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window, the method further includes: Align the three-dimensional coordinates of the hole center in the coordinate system corresponding to the target model with the coordinates of the prefabricated hole center in the preset solution, and determine the relative relationship between the coordinate system corresponding to the target model and the coordinate system corresponding to the preset solution; Based on the relative relationship, a difference between the prefabricated hole position and the hole position is determined.

6. The method for measuring bridge pier surface and hole position based on UAV close-up photogrammetry according to claim 1 is characterized in that: Before performing three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured, the method further includes: Based on the shape and position distribution of the bridge pier to be measured, the preset route of the drone is determined; the preset route includes: a full-area controlled tilt photography route, a complete column-surrounding photography route, and a high-density photography route close to key parts; The aerial image is acquired by collecting images of the bridge pier to be measured according to the preset route using a drone.

7. The method for measuring bridge pier surface and hole position based on UAV close-up photogrammetry according to claim 1 is characterized in that: Before performing three-dimensional reconstruction based on the aerial image and the aerial triangulation result to obtain the target model of the bridge pier to be measured, the method further includes: Preprocessing the aerial image to obtain POS information; Perform aerial triangulation based on the aerial image and the POS information to obtain the aerial triangulation result.

8. A device for measuring bridge pier surface and hole position based on drone close-up photogrammetry, characterized in that: include: The reconstruction module is used to perform three-dimensional reconstruction based on aerial images and aerial triangulation results to obtain the target model of the bridge pier to be measured; A recognition module, used to recognize the hole positions in the target model based on a preset local window and determine the reference points; The determination module is used to determine the hole position information and surface information of the bridge pier to be measured based on the relative position of the reference point in the preset local window.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the method for measuring the pier surface and hole position based on drone close-up photogrammetry as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the method for measuring the bridge pier surface and hole position based on drone close-up photogrammetry as described in any one of claims 1 to 7 is implemented.

Images