Lidar-based bridge precast component external dimension detection apparatus and method

Abstract

The application relates to a bridge prefabricated component appearance size detection equipment and method based on a laser radar, wherein a mounting box, in which a laser radar and a rotating motor are arranged, is arranged on a base, the main appearance size of a bridge prefabricated component is detected by the laser radar, and an auxiliary unit in the mounting box supplements the detection work of the laser radar; data acquisition is carried out through the laser radar, three-dimensional reconstruction of the bridge prefabricated component is carried out, point cloud data of end faces and side faces are obtained after segmentation, filtering and noise reduction, point cloud data is wrapped by a convex hull algorithm, comparison between associated data and preset size design values of the prefabricated component is carried out, error data is calculated, and a size detection result is obtained. Pulse laser emitted by the laser radar scans the bridge prefabricated component to realize information acquisition, three-dimensional reconstruction and calculation are completed in real time, the whole detection process is not affected by external environmental conditions and complex backgrounds, application scenarios are widened, detection precision is ensured, detection speed is improved, and the automation and intelligent degree are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of measurement; test, in particular to a bridge prefabricated component appearance size detection device and method based on laser radar. BACKGROUND

[0002] The prefabricated component refers to the steel, wood or concrete component preformed in the factory or on site according to the design specification, and the bridge prefabricated component is one of them. According to the size of the general bridge, even if the prefabricated component form is used for assembling the bridge, the size and volume of the bridge prefabricated component are also considerable, such as the size of a conventional bent cap can reach 2600mm*2400mm*19050mm.

[0003] In the prior art, the bridge prefabricated component appearance size detection generally adopts the traditional full manual or semi-manual detection technology. The full manual detection obviously has problems of low detection result accuracy, low detection efficiency, poor detection data repeatability and the like, and cannot meet the requirements of modern assembly industry on the measurement process in terms of measurement efficiency, size accuracy and the like, in addition, the professional accomplishment and technical level of the measurement personnel directly affect the measurement result accuracy, and omission, misjudgment and even wrong judgment occur from time to time, and the size accuracy of the prefabricated component is the core of its quality control, and the components out of specification may seriously affect the on-site assembly process of the components; the semi-manual detection refers to the measurement by the computer vision of the 2D camera, and this measurement method depends on the external light conditions and is greatly affected by the environmental factors, and the application scene and application conditions are relatively harsh, although the method has certain improvement in efficiency compared with the traditional full manual measurement method, but due to the strict limitation on the external environmental conditions, the detection efficiency is still low and the measurement accuracy is unstable.

[0004] At present, in view of the problems of low detection accuracy and low detection efficiency in the bridge prefabricated component appearance size measurement, a detection device with high efficiency and accuracy and not affected by external environmental conditions is needed. SUMMARY

[0005] The present application solves the problems in the prior art and provides an optimized bridge prefabricated component appearance size detection device and method based on laser radar.

[0006] The technical scheme adopted by the present application is a bridge prefabricated component appearance size detection device based on laser radar, the device comprises a base, a mounting box is matched on the base, a laser radar is arranged in the mounting box, a rotating motor is matched with the laser radar, and an auxiliary unit is further arranged in the mounting box matched with the laser radar.

[0007] Preferably, the installation box comprises a first installation compartment arranged on the inner upper side, at least one side wall of the first installation compartment being detachable or transparent; the laser radar is arranged in the first installation compartment, and a working surface of the laser radar is arranged in cooperation with the detachable or transparent side wall; a fixing seat is arranged in the first installation compartment, a support is arranged on the fixing seat, a rotating motor is arranged on the upper portion of the support, and an output shaft of the rotating motor is connected with the laser radar.

[0008] Preferably, the auxiliary unit comprises an image acquisition device arranged in cooperation with the fixing seat, and the image acquisition device is arranged towards the detachable or transparent side wall.

[0009] Preferably, the installation box further comprises a second installation compartment arranged below the first installation compartment, the auxiliary unit comprises a plurality of auxiliary light sources arranged in cooperation with the second installation compartment, and a light emitting surface of the auxiliary light source is arranged below the detachable or transparent side wall; the auxiliary light sources are even and symmetrical about a vertical central axis of the installation box; the auxiliary unit further comprises a refrigeration device and a temperature sensor arranged in cooperation with the second installation compartment and / or the first installation compartment.

[0010] Preferably, the installation box further comprises a third installation compartment arranged below the second installation compartment, the third installation compartment is provided with a data processing device and a battery module, the data processing device is arranged in cooperation with the laser radar and the auxiliary unit, and the battery module is electrically connected with the laser radar, the rotating motor and the auxiliary unit; a touch screen and an indicator lamp are arranged outside the installation box in cooperation with the data processing device and the battery module.

[0011] Preferably, a moving mechanism is arranged below the bottom plate.

[0012] A bridge precast member appearance size detection method based on the laser radar device, the method comprising the following steps:

[0013] S1 performing data acquisition at a plurality of angles on any bridge precast member by the laser radar; and performing three-dimensional reconstruction on the bridge precast member based on the acquired data;

[0014] S2 segmenting the three-dimensionally reconstructed bridge precast member and invalid information around the bridge precast member;

[0015] S3 filtering and denoising the cut point cloud data by using a voxel down-sampling algorithm, and segmenting the end surface and the side surface of the bridge precast member by using a random sample consensus algorithm to obtain point cloud data of the end surface and the side surface of the bridge precast member;

[0016] S4 performing wrapping processing on the point cloud data of the end surface and the side surface of the bridge precast member respectively by using a convex hull algorithm to obtain associated data;

[0017] S5 comparing the associated data with a preset size design value of the bridge precast member, calculating error data, and obtaining a size detection result.

[0018] Preferably, in S2, the data of the bridge prefabricated component after three-dimensional reconstruction is acquired, the preset size design value of any end face and / or side face of the bridge prefabricated component is extracted, corresponding equal proportion is enlarged to obtain a reference segmentation surface; the center of any end face and / or side face of the bridge prefabricated component after three-dimensional reconstruction is found and aligned with the center of the corresponding reference segmentation surface, the data of the end face and / or side face corresponding to the reference segmentation surface is extracted, and the surrounding invalid information is removed.

[0019] Preferably, in S3, the iteration number k of the random sample consensus algorithm is

[0020]

[0021] wherein p is the probability that all the points randomly selected from the data set in the iteration process are local points, ω is the probability of selecting a local point from the data set each time, n is the number of selected points required by the assumed estimation model; k is the standard deviation SD(k),

[0022] Preferably, in S4, the convex hull algorithm adopts Graham scanning method or Jarvis stepping method; and in S4, the associated data includes column end face size, steel bar spacing, steel bar exposed size, column surface flatness and column height data.

[0023] The application provides an optimized bridge prefabricated component appearance size detection device and method based on laser radar, which comprises a base and a mounting box matched with the base, a laser radar and a rotating motor are arranged in the mounting box, the main appearance size of the bridge prefabricated component is detected by the laser radar, and an auxiliary unit is arranged in the mounting box to supplement the detection work of the laser radar; the bridge prefabricated component is three-dimensionally reconstructed by collecting data by the laser radar, and the point cloud data of the end face and the side face of the bridge prefabricated component is obtained after segmentation, filtering and noise reduction; the point cloud data is wrapped by a convex hull algorithm, the associated data is obtained, compared with the preset size design value of the bridge prefabricated component, error data is calculated, and a size detection result is obtained.

[0024] Compared with the traditional visual method which measures and calculates by means of a 2D camera, the application realizes information collection by scanning the bridge prefabricated component by the pulsed laser emitted by the laser radar, realizes real-time three-dimensional reconstruction and calculation, and is not affected by external environmental conditions, especially light, and is not affected by a complex background, which widens the application scene, greatly improves the detection speed while ensuring the detection precision, and greatly improves the automation and intelligence. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 It is a perspective view of the application;

[0026] Figure 2 It is a schematic diagram of the front view structure of the present application;

[0027] Figure 3 It is a schematic diagram of the side view structure of the present application;

[0028] Figure 4 It is a flow chart of the detection method of the present application;

[0029] Figure 5 It is two types of point cloud data of the bridge prefabricated part collected by the laser radar in the embodiment of the present application, wherein 5a is the end surface point cloud data of the column, and 5b is the side surface point cloud data of the column;

[0030] Figure 6 It is a gray scale image of the preliminary three-dimensional reconstruction of the two types of point cloud data in the embodiment of the present application, wherein 6a corresponds to the image of a, and 6b corresponds to the image of b; Figure 5 Figure 5

[0031] Figure 7 It is a gray scale image of the three-dimensional reconstruction result after the mode recognition and cutting processing in the embodiment of the present application, wherein 7a corresponds to the image of a, and 7b corresponds to the image of b; Figure 6 Figure 6

[0032] Figure 8 It is a detection table of the actual size data and error of the bridge prefabricated part calculated in the embodiment of the present application, including the column end surface size, the steel bar spacing, the steel bar exposed size, the column surface flatness and the column height, and the unit is mm. DETAILED DESCRIPTION

[0033] The present application will be further described in detail below in combination with the embodiments, but the protection scope of the present application is not limited thereto.

[0034] As shown in Figures 1-3 The present application relates to a bridge prefabricated part appearance size detection equipment based on laser radar, which comprises a base 1, an installation box is matched on the base 1, a laser radar 2 is arranged in the installation box, a rotating motor 3 is matched with the laser radar 2, and an auxiliary unit is further arranged in the installation box matched with the laser radar 2.

[0035] In the present application, the base 1 and the installation box thereon are taken as the main body of the equipment, and the laser radar 2 and the associated devices and components are integrated for bridge prefabricated part appearance size detection.

[0036] ​​​​In the application, the laser radar 2 uses laser as a signal source, the pulsed laser emitted by the laser hits the target object, causes scattering, and part of the light wave is reflected to the receiver of the laser radar 2, and the distance from the laser radar 2 to the target point can be obtained according to the laser ranging principle calculation; by continuously scanning the target object with pulsed laser, the data of all target points on the target object can be obtained, and after the data are transmitted to the main control system, mode recognition, cutting and imaging processing are performed on the data by the point cloud data processing module in the main control system, accurate three-dimensional images can be obtained, the size information of the target bridge prefabricated component can be calculated, and reading can be realized; the three-dimensional images and the related size information obtained after the calculation of the point cloud processing module can be compared with the design size database preset in the main control system, and the error can be calculated, and a detection report is generated subsequently, and the subsequent adjustment and processing work of the bridge prefabricated component can be directly performed by the staff according to the detection report, so that the detection accuracy is ensured, the detection speed is greatly improved, and the automation and intelligent degree are greatly improved.

[0037] In the application, in the process of realizing the above functions, the rotating motor 3 is arranged in cooperation with the laser radar 2, so that multi-angle data acquisition of the laser radar 2 is realized, and more accurate detection results are obtained with the aid of various auxiliary units.

[0038] The mounting box comprises a first mounting bin 4 arranged on the upper part of the inner side, at least one side wall of the first mounting bin 4 is detachable or transparent; the laser radar 2 is arranged in the first mounting bin 4, and the working surface of the laser radar 2 is arranged in cooperation with the detachable or transparent side wall; a fixing seat 5 is arranged in the first mounting bin 4, a support 6 is arranged on the fixing seat 5, a rotating motor 3 is arranged on the upper part of the support 6, and the output shaft of the rotating motor 3 is connected with the laser radar 2.

[0039] The auxiliary unit comprises an image acquisition device 7 arranged in cooperation with the fixing seat 5, and the image acquisition device 7 is arranged towards the detachable or transparent side wall.

[0040] In the application, the mounting box is arranged in layers, so that the space in the mounting box can be fully utilized.

[0041] In the application, one side wall of the first installation bin 4 is provided with a detachable or transparent image acquisition window, generally a transparent image acquisition window, a fixing seat 5 is arranged on the bottom plate of the first installation bin 4, a support 6 is vertically arranged on the fixing seat 5, a servo rotating platform 8 is arranged on the support 6, the servo rotating platform 8 is connected with the support 6 through a nut, a through hole (not shown in the figure) is formed in the support 6, the servo rotating platform 8 is installed on the support 6 through the through hole, a rotating motor 3 is arranged on the servo rotating platform 8, a laser radar 2 is connected with the rotating motor 3, and the working surface of the laser radar 2 is arranged towards the bridge prefabricated member, that is, matched with the detachable or transparent side wall; the cooperation of the servo rotating platform 8, the support 6 and the rotating motor 3 is easy to be understood by those skilled in the art, and those skilled in the art can arrange them according to the requirements.

[0042] In the application, the laser radar 2 is driven to rotate by the rotating motor 3 to acquire the point cloud data of the bridge prefabricated member, and the pitch motion range of the laser radar 2 driven by the rotating motor 3 is 0-40°.

[0043] In the application, the horizontal scanning field angle of the laser radar 2 is 190° at most.

[0044] In the application, the first installation bin 4 is further provided with an image acquisition device 7, including but not limited to an industrial camera and a video camera, the working surface of the image acquisition device 7 is arranged towards the bridge prefabricated member, that is, matched with the detachable or transparent side wall, and the image acquisition device 7 is used for acquiring the image data of the bridge prefabricated member.

[0045] In the application, the image acquisition device 7 further comprises a camera support 9 arranged below the fixing seat 5, the industrial camera is arranged on the camera support 9, and the industrial camera is located directly below the laser radar 2.

[0046] The installation box further comprises a second installation bin 10 arranged below the first installation bin 4, and the auxiliary unit comprises a plurality of auxiliary light sources 11 arranged in the second installation bin 10, the light emitting surface of the auxiliary light source 11 is arranged below the detachable or transparent side wall; the auxiliary light sources 11 are even and symmetrical about the vertical central axis of the installation box; the auxiliary unit further comprises a refrigeration device 12 and a temperature sensor (not shown in the figure) arranged in the second installation bin 10 and / or the first installation bin 4.

[0047] In the application, the second installation bin 10 is arranged below the first installation bin 4, and the main purpose is to arrange the auxiliary light source 11, the illumination end of the auxiliary light source 11 is arranged towards the bridge prefabricated member, that is, arranged below the detachable or transparent side wall. Generally, the auxiliary light source 11 is arranged as two, symmetrically arranged at the second installation bin 10 and can emit parallel light beams through the parallel light source emission structure, so as to ensure that the industrial camera can still shoot clear photos of the bridge prefabricated member in the case that the external light source condition is not good.

[0048] In the present application, the second installation compartment 10 and / or the first installation compartment 4 can be provided with a refrigeration device 12, such as a built-in air conditioner, which generally covers two compartments, according to the needs. When the ambient temperature is too high or the laser radar 2 works for a long time, the temperature in the installation box can be adjusted by the refrigeration device 12 to provide a working temperature of not more than 55℃ in the installation box. When the refrigeration device 12 is configured, an air outlet 15 is generally provided on the side wall of the installation box to discharge heat.

[0049] The installation box further comprises a third installation compartment 13 arranged below the second installation compartment 10, wherein the third installation compartment 13 is provided with a data processing device 14 and a battery module, the data processing device 14 is arranged in cooperation with the laser radar 2 and the auxiliary unit, and the battery module is electrically connected with the laser radar 2, the rotating motor 3 and the auxiliary unit; the installation box cooperating with the data processing device 14 and the battery module is provided with a touch screen 16 and an indicator light 17.

[0050] In the present application, the third installation compartment 13 is arranged below the second installation compartment 10, and the data processing device 14 and the battery module (not shown in the figure) are arranged in the third installation compartment 13; the data processing device 14 is used for receiving and processing image data collected by the industrial camera and point cloud data collected by the laser radar 2, and is provided with a programmable logic controller, a power converter, a cooling fan, a servo driver, a wiring terminal and the like, the data processing device 14 is connected with the laser radar 2, the image acquisition device 7 and the indicator light 17, and the data processing device 14 performs real-time processing of the point cloud data and the image data, and finally obtains the size data of the bridge prefabricated component; the battery module provides stable working voltage for the equipment, and generally selects a rechargeable lithium battery pack. The lithium battery pack stores energy and supplies power to the device, and a charging interface can be arranged on the installation box to charge the lithium battery pack, or the lithium battery pack can be taken out for independent charging.

[0051] In the present application, the touch screen 16 is arranged outside the installation box, and the touch screen 16 is provided with equipment operation buttons, including but not limited to a power switch, a reset button, an end surface scanning button and a side surface scanning button.

[0052] In the present application, the indicator light 17 is arranged outside the installation box, and the indicator light 17 displays different colors when the equipment is in different working states, such as yellow constant light after the device reset self-checking work is completed, green constant light after the scanning work of the laser radar 2 is completed, and the indicator light 17 is in an extinguished state during the rest of the time.

[0053] The bottom plate is provided with a moving mechanism 18.

[0054] In the present application, the moving mechanism under the base 1 can realize the rapid movement of the equipment, and the moving mechanism includes but is not limited to wheels, fork openings cooperating with forklifts and the like.

[0055] In the present application, the skilled person can set up the communication antenna, power switch and the like of the device as needed to complete different functions and function expansion, which is easily understood by the skilled person; the operation flow embodiment of the device is proposed, including the following steps:

[0056] Step 1: start the device, start the corresponding detection program on the touch screen 16, and move the device to the front of the end face of the prefabricated part;

[0057] Step 2: observe whether the measured prefabricated part is completely in the shooting range of the industrial camera in the touch screen 16, make corresponding adjustment, press the reset key, and the indicator light 17 is always yellow after self-checking;

[0058] Step 3: start end face scanning, the indicator light 17 is always green after completing the instruction task, and the collected data is transmitted to the point cloud information processing module of the data processing device 14;

[0059] Step 4: move the device to the side of the prefabricated part, repeat steps 1 to 3, until all point cloud data collection is completed;

[0060] Step 5: the data processing device 14 processes the point cloud information, calculates the three-dimensional model information of the prefabricated part, and displays it on the touch screen 16; after the interception work is completed, click the touch screen 16 to start automatic measurement;

[0061] Step 6: the data processing device calculates the size of the intercepted three-dimensional model, compares the calculation result with the prefabricated part design size database in the system, and displays the comparison result on the touch screen 16, and generates a detection report.

[0062] As shown in Figure 4 The present application also relates to a bridge prefabricated part appearance size detection method based on laser radar using the device, and the method comprises the following steps:

[0063] S1: data collection of several angles is carried out on any bridge prefabricated part by laser radar 2; three-dimensional reconstruction is carried out on the bridge prefabricated part based on the collected data;

[0064] In the present application, the laser radar 2 acquires data of the bridge prefabricated part at different angles, or the laser radar 2 acquires data of the bridge prefabricated part at different distances and angles by going around the bridge prefabricated part, including point cloud data of the column end face and the column side face of the bridge prefabricated part; the greater the deflection angle of the laser radar 2, the higher the sampling frequency.

[0065] S2: segmentation is carried out on the three-dimensionally reconstructed bridge prefabricated part and the invalid information around the bridge prefabricated part;

[0066] In S2, the three-dimensional reconstructed bridge prefabricated component data is acquired, preset size design values of any end face and / or side face of the bridge prefabricated component are extracted, corresponding equal proportion magnification is performed to obtain a reference segmentation face; the center of any end face and / or side face of the three-dimensional reconstructed bridge prefabricated component is found and aligned with the center of the corresponding reference segmentation face, data of the end face and / or side face corresponding to the reference segmentation face is extracted, and invalid information around is removed.

[0067] In the application, the obtained point cloud data containing the end face and the side face of the bridge prefabricated component can be directly subjected to real-time three-dimensional reconstruction processing by means of the point cloud visualization software Open3D in python, and after the reconstruction is completed, cutting processing is directly performed at the operation end to preliminarily remove irrelevant data such as environmental background. Compared with the existing binocular vision technology using parallax to calculate depth, this method is more intuitive and rapid, and the cutting processing can greatly reduce the subsequent computer workload.

[0068] In S3, the filtered and denoised point cloud data after cutting is subjected to filtering and denoising by means of a voxel downsampling algorithm, and the point cloud data of the end face and the side face of the bridge prefabricated component is segmented by means of a random sample consensus algorithm.

[0069] In S3, the iteration number k of the random sample consensus algorithm is k = 3.

[0070]

[0071] In the formula, p is the probability that all the points randomly selected from the data set in the iteration process are local points, ω is the probability that a local point is selected from the data set each time, that is, ω = the number of local points / the number of data sets, and n is the number of selected points required by the hypothesis estimation model; k takes the standard deviation SD(k), that is, k = 3. The local point refers to a point in the target plane, and the hypothesis estimation model refers to a mathematical model that we want to iterate out by means of the random sample consensus algorithm, that is, a target model.

[0072] In the application, when the point cloud is preprocessed, a large amount of three-dimensional point cloud data is obtained in actual application, which often contains some noise information. In order to improve the processing effect and calculation efficiency of the point cloud, some preprocessing is usually performed. Downsampling is a very key step in the point cloud preprocessing process, which not only affects the processing speed of the point cloud, but also affects the final processing effect. Here, the voxel downsampling algorithm is used to filter and denoise the cut point cloud data, to improve the subsequent point cloud processing speed and processing effect. The voxel downsampling algorithm is a conventional preprocessing means technology. Compared with other downsampling algorithms, the voxel downsampling algorithm has high calculation efficiency and uniform sampling point distribution, and has better effect in the downsampling processing of dense three-dimensional point cloud data. Here, the voxel downsampling algorithm in the open three-dimensional library is used to simplify the point cloud, and the performance in maintaining the data integrity of the high-density component is better.

[0073] In the application, in practical application, the ability of finding arbitrary shape clusters in noisy spatial data can be further utilized by a spatial clustering algorithm to find column end surface point clouds, embedded bar end point clouds and column bottom edge point clouds that meet the conditions, and to correct the points that are incorrectly marked. The clustering algorithm is used in the process of finding the target plane. After segmentation and downsampling, there are still many error description points in the point cloud data. In order to correct the points that are incorrectly marked and find the target plane, point clustering is performed using DBSCAN (density-based spatial clustering of applications with noise) and the like, and then the small clusters with noise and dispersion are filtered. The DBSCAN spatial clustering algorithm is a density-based spatial clustering algorithm that divides regions with sufficient density into clusters and finds clusters of arbitrary shape in noisy spatial databases. The cluster is defined as the largest set of densely connected points.

[0074] In the application, the random sample consensus algorithm (RANSAC) is then used to finely segment the column end surface, column side surface and column bottom edge point cloud data found, to provide more accurate point cloud data for subsequent distance calculation. Specifically, the goal of this algorithm is to find the plane with the most support in the point cloud. The parameters of a mathematical model are iteratively estimated from a set of observation data containing "outliers". In order to increase the probability, the number of iterations must be increased. The data set here is the point cloud data that has been segmented, preprocessed and downsampled. Of course, there is still some noise in this data set.

[0075] S4 wraps the point cloud data of the end surface and side surface of the bridge precast component respectively by a convex hull algorithm to obtain associated data;

[0076] In S4, the convex hull algorithm uses the Graham scanning method or the Jarvis stepping method. In S4, the associated data includes column end surface size, reinforcement spacing, reinforcement exposed size, column surface flatness and column height data.

[0077] In the application, the Graham scanning method is implemented through two steps of point set sorting and stack scanning. The Jarvis stepping method selects the point with the smallest horizontal coordinate as the starting point, and then finds the outermost point by using the cross product comparison method to wrap the point, and iterates all points until the starting point is found again.

[0078] In the application, the convex hull algorithm is not equal to the clustering algorithm. The clustering algorithm and the random sample consensus algorithm are both means of optimizing and removing noise in the process of finding the target plane. The convex hull algorithm is used to finally perform wrapping calculation on the point cloud data that has been processed and optimized.

[0079] S5 compares the association data with the preset size design value of the bridge prefabricated component, calculates error data, and obtains a size detection result.

[0080] In the present application, the error includes direct error, average error and maximum error, which are used to generate a corresponding factory detection report.

[0081] As shown in Figures 5-8 , an embodiment of the device and method of the present application is shown; the present application shows that the size precision reaches the millimeter level through measurement, the maximum error of the exposed size of the steel bar is 4.1 mm, the average error is only 2.03 mm; the maximum error of the steel bar spacing is 3.9 mm, the average error is only 1.88 mm; the end face size error is less than 5 mm; the maximum difference of the flatness is 7.8 mm, which is much smaller than the design requirement of 15 mm;

[0082] The traditional manual detection method needs about 1 hour to complete the appearance size detection of a single bridge prefabricated component, and the detection precision cannot reach the millimeter level, while the present application obtains the point cloud data of the bridge prefabricated component by means of the non-contact laser radar 7, performs real-time three-dimensional reconstruction and size calculation, and only needs five minutes to complete the appearance size detection of a single bridge prefabricated component, which guarantees the precision, greatly improves the detection efficiency, and reduces the production cost.

Claims

1. A method for detecting the appearance and dimensions of prefabricated bridge components based on laser radar, characterized by: The method adopts a detection device, including a base, a mounting box is provided on the base, a laser radar is provided in the mounting box, a rotating motor is provided in conjunction with the laser radar, and an auxiliary unit is also provided in the mounting box in conjunction with the laser radar; The method comprises the following steps: S1: collecting data from several angles of any prefabricated bridge component using a laser radar; and performing three-dimensional reconstruction of the prefabricated bridge component based on the collected data; S2 segments the invalid information of the prefabricated bridge components and their surroundings after 3D reconstruction; S3 uses the voxel downsampling algorithm to filter and reduce noise on the cropped point cloud data, and then uses the random sampling consistency algorithm to segment the end and side point cloud data of the bridge prefabricated components; the number of iterations k of the random sampling consistency algorithm is, ， Among them, p is the probability that all randomly selected points from the data set during the iteration are in-game points, is the probability of selecting an internal point from the data set each time, n is the number of points required to be selected for the assumed estimation model; k is the standard deviation SD ( ), ; S4: using a convex hull algorithm to perform wrapping processing on the point cloud data of the end face and the side face of the prefabricated bridge component to obtain associated data; S5 compares the associated data with the preset size design values ​​of the bridge prefabricated components, calculates the error data, and obtains the size detection results.

2. The method for detecting the appearance dimensions of prefabricated bridge components based on laser radar according to claim 1, characterized in that: In S2, the three-dimensionally reconstructed prefabricated bridge component data is obtained, and the preset size design value of any end face and / or side face of the prefabricated bridge component is extracted, and the corresponding geometric enlargement is performed to obtain the reference segmentation surface; Find the center of any end face and / or side face of the 3D reconstructed prefabricated bridge component and align it with the center of the corresponding reference segmentation surface, extract the data of the end face and / or side face corresponding to the reference segmentation surface, and eliminate the surrounding invalid information.

3. The method for detecting the appearance and dimensions of prefabricated bridge components based on laser radar according to claim 1, characterized in that: In S4, the convex hull algorithm adopts the Graham scanning method or the Jarvis stepping method; in S4, the associated data includes the column end face size, steel bar spacing, steel bar exposed size, column surface flatness and column height data.

4. The method for detecting the appearance and dimensions of prefabricated bridge components based on laser radar according to claim 1, characterized in that: The installation box includes a first installation compartment located on the upper inner side, and at least one side wall of the first installation compartment is detachable or transparent; the laser radar is arranged in the first installation compartment, and the working surface of the laser radar is cooperated with the detachable or transparent side wall; a fixing seat is provided in the first installation compartment, and a bracket is provided on the fixing seat, and a rotating motor is provided on the upper part of the bracket, and the output shaft of the rotating motor is connected to the laser radar.

5. The method for detecting the appearance and dimensions of prefabricated bridge components based on laser radar according to claim 4, characterized in that: The auxiliary unit includes an image acquisition device arranged in conjunction with the fixing seat, and the image acquisition device is arranged toward the detachable or transparent side wall.

6. The method for detecting the appearance and dimensions of prefabricated bridge components based on laser radar according to claim 4, characterized in that: The installation box also includes a second installation bin arranged under the first installation bin, and the auxiliary unit includes a plurality of auxiliary light sources arranged in conjunction with the second installation bin, and the light-emitting surface of the auxiliary light source is arranged under the removable or transparent side wall; the auxiliary light sources are an even number and are symmetrical about the vertical central axis of the installation box; the auxiliary unit also includes a refrigeration device and a temperature sensor arranged in conjunction with the second installation bin and / or the first installation bin.

7. The method for detecting the appearance and dimensions of prefabricated bridge components based on laser radar according to claim 6, characterized in that: The installation box also includes a third installation compartment located under the second installation compartment, and a data processing device and a battery module are installed in the third installation compartment. The data processing device is arranged in conjunction with the laser radar and the auxiliary unit, and the battery module is electrically connected to the laser radar, the rotating motor and the auxiliary unit; a touch screen and an indicator light are provided outside the installation box that cooperates with the data processing device and the battery module.

8. The method for detecting the appearance and dimensions of prefabricated bridge components based on laser radar according to claim 1, characterized in that: A moving mechanism is provided under the base.

Images