CN118111345A - Tunnel foundation pit surrounding rock displacement, crack and ponding monitoring system - Google Patents
Tunnel foundation pit surrounding rock displacement, crack and ponding monitoring system Download PDFInfo
- Publication number
- CN118111345A CN118111345A CN202410109532.2A CN202410109532A CN118111345A CN 118111345 A CN118111345 A CN 118111345A CN 202410109532 A CN202410109532 A CN 202410109532A CN 118111345 A CN118111345 A CN 118111345A
- Authority
- CN
- China
- Prior art keywords
- foundation pit
- side wall
- camera
- model
- calibration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 39
- 238000006073 displacement reaction Methods 0.000 title claims abstract description 29
- 239000011435 rock Substances 0.000 title claims abstract description 17
- 230000008859 change Effects 0.000 claims abstract description 23
- 238000012806 monitoring device Methods 0.000 claims abstract description 14
- 238000005516 engineering process Methods 0.000 claims abstract description 13
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 9
- 238000007689 inspection Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 23
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000010276 construction Methods 0.000 claims description 15
- 238000001514 detection method Methods 0.000 claims description 15
- 238000005259 measurement Methods 0.000 claims description 15
- 239000013598 vector Substances 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 238000013519 translation Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000005520 cutting process Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 4
- 238000007476 Maximum Likelihood Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000011157 data evaluation Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000006467 substitution reaction Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims description 3
- 238000005336 cracking Methods 0.000 claims 3
- 238000009825 accumulation Methods 0.000 claims 1
- 230000000007 visual effect Effects 0.000 description 6
- 238000000605 extraction Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009435 building construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000011150 reinforced concrete Substances 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention belongs to the technical field of engineering monitoring, and particularly relates to a system for monitoring displacement, crack and ponding change of surrounding rock of a tunnel foundation pit based on laser radar and machine vision technology, which comprises the following components: the track is paved at the top of the foundation pit; the mobile trolley is arranged on the track and can carry out mobile inspection along the track; the monitoring device is carried on the mobile trolley and used for obtaining information of the side wall of the foundation pit inspected by the mobile trolley; the monitoring device scans the side wall of the foundation pit by using a laser radar and a structured light camera to acquire three-dimensional point cloud data, establishes a three-dimensional point cloud model of the wall of the foundation pit under a laser radar coordinate system, and accurately measures the whole and partial displacement deformation condition, side wall cracks and water seepage condition of the side wall of the foundation pit through regional division and data fitting.
Description
Technical Field
The invention belongs to the technical field of engineering monitoring, and particularly relates to a system for monitoring displacement, crack and ponding change of surrounding rock of a tunnel foundation pit based on laser radar and machine vision technology.
Background
With the development of society, the problems of environmental protection, safety production and the like are increasingly concerned, a large amount of electric power energy sources, reinforced concrete and other resources are consumed in the building construction process, the environment is greatly influenced, and meanwhile, the complexity and the danger of the construction site determine the importance of safety construction. Therefore, the concept of green and safe construction is increasingly paid attention to government, construction units and construction units, and relevant policies and standards of corresponding countries, places and enterprises are continuously released and updated. Under the background, technical research and application aiming at safety real-time monitoring and alarming of constructional engineering are developed, and real-time monitoring and management of resources, environment and safety information in the construction process are very necessary.
Along with the increasing difficulty of tunnel construction, the stability detection of the road tunnel structure is particularly important. In the traditional road tunnel stability detection, visual, broken or manual nondestructive detection methods are often adopted to detect the tunnel, and the detection of the methods is unstable and the detection efficiency is low, so how to quickly and accurately detect the tunnel stability is a research hot spot.
Disclosure of Invention
The invention aims to provide a tunnel foundation pit surrounding rock displacement, crack and ponding change monitoring system based on laser radar and machine vision technology so as to solve the technical problems.
In order to achieve the above purpose, the following technical scheme is provided: a tunnel foundation pit surrounding rock displacement, crack and ponding change monitoring system includes:
the track is paved at the top of the foundation pit;
the mobile trolley is arranged on the track and can carry out mobile inspection along the track;
The monitoring device is carried on the mobile trolley and used for obtaining information of the side wall of the foundation pit inspected by the mobile trolley;
The monitoring device scans the side wall of the foundation pit by using a laser radar and a structured light camera to acquire three-dimensional point cloud data, establishes a three-dimensional point cloud model of the wall of the foundation pit under a laser radar coordinate system, and accurately measures the whole and partial displacement deformation condition, side wall cracks and water seepage condition of the side wall of the foundation pit through regional division and data fitting.
In the above technical solution, further, the monitoring device includes:
A three-dimensional calibration module;
the image point cloud splicing and fusing module converts the image into three-dimensional point cloud data, splices the multi-frame point cloud into a three-dimensional model, performs smoothing and fusion on the spliced point cloud, and finally performs data processing to obtain a final three-dimensional model;
The model fitting module is used for selecting a corresponding linear or nonlinear model according to the data characteristics and a target to be fitted; secondly, calculating parameter values of the model by using a least square method, a maximum likelihood estimation method or other algorithms; evaluating the predictive performance of the model using the validation dataset and calculating a prediction error; if the prediction performance of the model does not meet the requirement, trying to fine-tune parameters of the model to improve the accuracy of the model; predicting values of unknown data using the resulting fitted model;
The detection module is used for preprocessing data, including noise removal and standardization; secondly, training an artificial intelligent model by using a machine learning algorithm, and evaluating the effect of the artificial intelligent model; then, adjusting model parameters to improve accuracy; the model was tested on a separate dataset and its accuracy was assessed.
In any of the above technical solutions, further, the three-dimensional calibration module includes:
The image acquisition unit is used for acquiring images of the calibration surface by using the structured light binocular camera;
A camera correction unit that corrects the acquired image;
the point cloud construction unit is used for carrying out three-dimensional reconstruction on the characteristic points on the calibration surface by utilizing the parallax image and the depth image to construct point cloud;
the parameter calibration unit is used for analyzing the point cloud data and calculating the internal parameters and the external parameters of the camera;
And the camera calibration unit is used for calibrating the internal and external parameters of the camera through the internal parameters and the external parameters calculated by the parameter calibration unit, wherein the internal parameters are a camera matrix and distortion coefficients, and the external parameters are a rotation matrix and translation vectors.
In any of the above technical solutions, further, the data evaluation unit evaluates the calibration result, so as to ensure the calibration accuracy.
In any one of the above technical solutions, further, in the process of monitoring the displacement change of the side wall of the foundation pit, acquiring the side wall data of the foundation pit under multiple measurements by using a laser radar, filtering noise data according to initial parameters measured by the foundation pit, and performing corresponding interpolation processing on the side wall data; and monitoring the position change condition of the side wall according to the change condition of the side wall data under the condition of comparing the shooting height with the unified height and measuring for a plurality of times, and calculating the displacement of the side wall.
In any one of the above technical solutions, further, in the process of monitoring the crack of the side wall of the foundation pit, a camera is used to collect the crack data of the side wall of the foundation pit; firstly cutting an original image into k multiplied by k small blocks by adopting an image cutting mode, and then performing model training on the image size;
The crack height positioning algorithm adopts an SVR regression algorithm, and the heights corresponding to four corner points of different crack detection frames in the test picture are obtained by constructing fitting relation functions of characteristics such as foundation pit side wall parameters, the heights shot by a camera, coordinates of pixel points in the picture and the like and the actual heights corresponding to the pixel points in the picture, so that the height range of the crack is obtained.
In any of the above technical solutions, further, the monitoring device extracts multiple groups of three-dimensional feature points in the image by using the RGB image and the depth image of the reference target of the fixed point at the top of the foundation pit shot by the laser radar and the structured light camera, calculates the pose transformation matrix of the structured light camera relative to the first shot under multiple measurements, and combines the combined calibration algorithm of the structured light camera and the laser radar to unify the laser radar coordinate system under multiple measurements to the laser radar coordinate system under the first scan, so as to realize the unification of the multiple measurement coordinate systems.
In any of the above technical solutions, further, the joint calibration algorithm performs joint calibration on the laser radar and the structured light camera by adopting a calibration method based on plane constraint.
In any of the above technical solutions, further, the camera calculates coordinates of the calibration plate plane under the camera coordinate system by identifying the tag on the calibration plate; the laser radar emits light beam to strike the calibration board, the plane constraint is constructed by utilizing the coordinates of the laser point under the laser coordinate system and the coordinates under the camera coordinate system to solve the external parameters,
The plane constraint calculation formula is as follows:
ncT(RPl+t)+dc＝0
Wherein, the vector n is a three-dimensional normal vector of the label plane, d is a distance from the origin of the camera coordinate system O c to the label plane, the coordinate of the point P under the laser coordinate system O l is P l, and the coordinate under the camera coordinate system O c is P c; the transformation of point P from laser coordinate system O l to camera coordinate system O c is:
Pc＝RT(Pl-t)
Substitution P l＝(x,y,0)T:
The matrix H is solved for nonlinear least squares as a new unknown to recover and obtain a rotation matrix and translation vector r_t'.
The beneficial effects of the invention are as follows: according to the method, three-dimensional modeling calculation is carried out on the side wall of the foundation pit in different time periods by adopting a three-dimensional calibration technology, an image splicing technology, a point cloud splicing technology, an image and point cloud fusion technology, a linear and nonlinear fitting technology, an artificial intelligent detection technology and the like, the displacement deformation condition of the side wall is calculated according to the position change of characteristic points in different time periods, and the crack and water seepage condition of the side wall are identified and detected by adopting the artificial intelligent detection technology so as to realize real-time monitoring of the side wall of the foundation pit.
Drawings
FIG. 1 is a flow chart of the target area point cloud construction of the present invention;
FIG. 2 is a schematic diagram of the joint calibration of the camera and lidar of the present invention;
Detailed Description
The technical solutions of the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which are obtained by a person skilled in the art based on the embodiments of the present application, fall within the scope of protection of the present application.
In the description of the present application, it is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments in accordance with the present application. For ease of description, the dimensions of the various features shown in the drawings are not drawn to actual scale. Techniques, methods, and apparatus known to one of ordinary skill in the relevant art may not be discussed in detail, but should be considered part of the specification where appropriate. In all examples shown and discussed herein, any specific values should be construed as merely illustrative, and not a limitation. Thus, other examples of the exemplary embodiments may have different values. It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further discussion thereof is necessary in subsequent figures.
The intelligent visual monitoring, analyzing and early warning system for displacement belongs to the field of engineering monitoring and mainly comprises an adjustable target, an intelligent visual identification module, an intelligent acquisition control module, a cloud platform, a client and the like.
The adjustable targets are fixed at the top of the foundation pit and are arranged opposite to the intelligent visual recognition module; the intelligent visual recognition module is connected with the intelligent acquisition control module through a wire; the intelligent acquisition control module is connected with the cloud platform through a network; the cloud platform analyzes and calculates the received data, compares the received data with a set allowable deformation threshold, marks different threshold intervals with corresponding colors, characterizes the security state of the measuring point with different colors of the measuring point, and sends alarm information to a designated receiver; and finally, visually visualizing deformation and early warning information of each measuring point by using the client.
The existing method is based on single visual monitoring analysis, has high requirements on site environment, is easily influenced by dust and the like caused by site construction and personnel activities, and the scheme needs to use a target as a measured point, has high requirements on point selection and installation, is greatly influenced by construction, for example, the condition that a construction instrument vehicle shields the target and the like, can lead to the reduction of the monitoring effect, and has single monitoring item. Based on the displacement, crack and ponding change monitoring system of tunnel foundation pit surrounding rock based on laser radar and machine vision technology is provided by the applicant.
The monitoring system of the present application is described in detail by the following examples.
Example 1:
The embodiment provides a tunnel foundation ditch country rock displacement, crack and ponding change monitoring system, include: the track is paved at the top of the foundation pit; the mobile trolley is arranged on the track and can carry out mobile inspection along the track; the monitoring device is mounted on the mobile trolley and used for obtaining information of the side wall of the foundation pit inspected by the mobile trolley;
The monitoring device scans the side wall of the foundation pit by utilizing a laser radar and a structured light camera to acquire three-dimensional point cloud data, establishes a three-dimensional point cloud model of the wall of the foundation pit under a laser radar coordinate system, and accurately measures the whole and partial displacement deformation condition, side wall cracks and water seepage condition of the side wall of the foundation pit through region division and data fitting.
Specifically, the monitoring device includes:
A three-dimensional calibration module; the three-dimensional calibration module comprises: the image acquisition unit is used for acquiring images of the calibration surface by using the structured light binocular camera; a camera correction unit that corrects the acquired image; the point cloud construction unit is used for carrying out three-dimensional reconstruction on the characteristic points on the calibration surface by utilizing the parallax image and the depth image to construct point cloud; the parameter calibration unit is used for analyzing the point cloud data and calculating the internal parameters and the external parameters of the camera; the camera calibration unit is used for calibrating the internal and external parameters of the camera through the internal parameters and the external parameters calculated by the parameter calibration unit, wherein the internal parameters are a camera matrix and distortion coefficients, and the external parameters are a rotation matrix and a translation vector; the data evaluation unit evaluates the calibration result to ensure the calibration accuracy; the point cloud curved surface reconstruction aims to enable the point cloud curved surface reconstruction to reflect geometrical characteristics of the object surface vividly, and is convenient for extraction and analysis of model physical information and geometrical information;
the image point cloud splicing and fusing module converts the image into three-dimensional point cloud data, splices the multi-frame point cloud into a three-dimensional model, performs smoothing and fusion on the spliced point cloud, and finally performs data processing to obtain a final three-dimensional model;
The model fitting module is used for selecting a corresponding linear or nonlinear model according to the data characteristics and a target to be fitted; secondly, calculating parameter values of the model by using a least square method, a maximum likelihood estimation method or other algorithms; evaluating the predictive performance of the model using the validation dataset and calculating a prediction error; if the prediction performance of the model does not meet the requirement, trying to fine-tune parameters of the model to improve the accuracy of the model; predicting values of unknown data using the resulting fitted model;
The detection module is used for preprocessing data, including noise removal and standardization; secondly, training an artificial intelligent model by using a machine learning algorithm, and evaluating the effect of the artificial intelligent model; then, adjusting model parameters to improve accuracy; the model was tested on a separate dataset and its accuracy was assessed.
In the technical scheme, the working principle of the laser radar for collecting point cloud information is as follows: firstly, a laser transmitter transmits a high-energy laser beam to an observed target entity through an observation lens, the laser beam is reflected after reaching the surface of an object, returns along the same path and is received by a scanner receiver, so that the distance between the scanner and the measured object is obtained; then, acquiring horizontal angle and vertical angle information between a target object and a scanner by using an angle measurement control module, and completing acquisition of target point position information by using a driving motor and a steerable lens; and finally, converting the acquired distance information and angle information into coordinate information by utilizing a microprocessor, and storing the information such as the coordinate information, the reflection intensity, the texture characteristics and the like into a memory.
Due to the influence of factors such as an instrument, an external environment, personnel operation and the like, noise points which interfere with geometric characteristic information of an object can exist in the collected foundation pit side wall point cloud data. The side wall point cloud data which is not subjected to noise elimination processing can influence the accuracy of subsequent point cloud data registration and side wall three-dimensional solid model establishment, and can influence the later-stage side wall dynamic deformation monitoring. For this reason, denoising is required.
The large-scale noise in the point cloud data mainly refers to useless information points which are far away from the point cloud of the main body and are easy to distinguish, and include isolated points distributed near the main body of the point cloud and drift points which are far away from the point cloud data of the main body and have small quantity.
Small-scale noise points which are mixed with effective data points or are close to main point cloud data and difficult to distinguish are also present in the point cloud data acquired by the laser radar due to the external environment. The model surface built by the point cloud data containing small-scale noise is not smooth, coarse and poor in effect, so that the points mixed in the point cloud data need to be removed or filtered.
Therefore, optimally, in the process of monitoring the displacement change of the side wall of the foundation pit, acquiring side wall data of the foundation pit under multiple measurement by a laser radar, filtering noise data according to initial parameters (such as the initial distance between an instrument and the side wall, the height of the foundation pit and the like) of the foundation pit measurement, and carrying out corresponding interpolation processing on the side wall data to enable the side wall data to be denser; and monitoring the position change condition of the side wall according to the change condition of the side wall data under the condition of comparing the shooting height with the unified height and measuring for a plurality of times, and calculating the displacement of the side wall.
In the process of monitoring cracks of the side wall of the foundation pit, the acquired original image is oversized and is not suitable for being directly used as input of a network model, so that the camera is used for acquiring the crack data of the side wall of the foundation pit; the original image is cut into k multiplied by k small blocks by adopting an image cutting mode, and then model training is carried out on the image size.
Because the foundation pit crack detection model only obtains the position of the crack in the plane image, the specific position of the crack on the side wall of the foundation pit cannot be obtained, the subsequent repair work is not facilitated, the height of the identified crack needs to be clarified after the crack is identified in the plane image, the SVR regression algorithm is adopted for the crack height positioning algorithm, and the height corresponding to the four corner points of different crack detection frames in the test picture is obtained by constructing fitting relation functions of characteristics such as the parameters (height) of the side wall of the foundation pit, the height shot by a camera, the coordinates of pixel points in the picture and the like and the actual height corresponding to the pixel points in the picture, so that the height range of the crack is obtained.
In the condition of monitoring crack water seepage of the side wall of the foundation pit, the same image recognition is adopted, and the water seepage position and the height are positioned through a water seepage recognition algorithm.
In this embodiment, the monitoring device extracts multiple groups of three-dimensional feature points in the image by using the RGB image and the depth image of the reference target of the fixed point at the top of the foundation pit shot by the laser radar and the structured light camera, calculates the pose transformation matrix of the structured light camera relative to the first shot under multiple measurements, and combines the laser radar coordinate system under multiple measurements to the laser radar coordinate system under the first scan by combining the combined calibration algorithm of the structured light camera and the laser radar, thereby realizing the integration of multiple measurement coordinate systems.
Since the RGB image is a projection of a three-dimensional scene on a two-dimensional plane, no depth information is recorded, in order to determine the position and the posture of a target in a three-dimensional space, the depth image needs to be aligned to the RGB image, a point cloud image is constructed, and finally the spatial pose of the target is determined based on a point cloud processing related algorithm. As shown in fig. 1, a point cloud construction flowchart of the target area is constructed.
The method for computing the image feature point extraction and rotation matrix is as follows,
SIFT features are an image local feature extraction method, and have scale invariance, rotation invariance and certain illumination invariance. The specific steps of the algorithm are as follows.
1) Scale space extremum detection
A Gaussian pyramid of the image is created and a Gaussian difference image (DoG) at different scales is calculated. Local extremal points are detected in the DoG image, which may be potential key points. The gaussian kernel function is shown as follows:
L(x，y，σ)＝G(x，y，σ)*I(x，y)
Wherein G (x, y, σ) is a Gaussian kernel function:
2) Key point positioning
The key points that are suboptimal in the DoG algorithm have a larger principal curvature in the direction of the parallel edges and a smaller curvature in the direction perpendicular to the edges, and if the ratio of the two is above a certain threshold, the key points are considered as boundaries and are ignored.
3) Keypoint direction determination
A direction is assigned to each keypoint to achieve rotational invariance. For any key point, gradient characteristics (amplitude and amplitude angle) of all pixels in the region of the Gaussian pyramid image with r as a radius are acquired.
4) Description of key points
The image information is abstracted by partitioning the image area around the key points, calculating gradient histograms in the blocks, generating feature vectors.
Extracting a plurality of three-dimensional characteristic point pairs of target areas at different time points according to image processing: p= { P 1,...,pn}P′＝{p′1,...,p′n }, P i＝Rpi +t is made by the euclidean transform r_t. The problem can be solved with Iterative Closest Points (ICP). Specifically, solution using linear algebra (SVD). By defining the error term, a least squares problem is constructed, solving such that the sum of squares of the errors reaches a minimum R_T. (ICP) can be solved in three steps:
The first step: calculating centroid positions p, p 'of the two groups of points, and then calculating a barycenter removing coordinate q i＝pi-p,q′i＝p′i -p' of each point;
And a second step of: defining an optimization problem calculation rotation matrix
And a third step of: calculating translation variables
t＝p-Rp′。
Regarding the combined calibration of the structured light camera and the laser radar, specifically, a combined calibration algorithm adopts a calibration method based on plane constraint to perform combined calibration on the laser radar and the structured light camera.
As shown in fig. 2, the camera calculates the coordinates of the calibration plate plane in the camera coordinate system by identifying the labels on the calibration plate; the laser radar emits light beam to strike the calibration board, the plane constraint is constructed by utilizing the coordinates of the laser point under the laser coordinate system and the coordinates under the camera coordinate system to solve the external parameters,
The plane constraint calculation formula is as follows:
ncT(RPl+t)+dc＝0
Wherein, the vector n is a three-dimensional normal vector of the label plane, d is a distance from the origin of the camera coordinate system O c to the label plane, the coordinate of the point P under the laser coordinate system O l is P l, and the coordinate under the camera coordinate system O c is P c; the transformation of point P from laser coordinate system O l to camera coordinate system O c is:
Pc＝RT(Pl-t)
Substitution P l＝(x,y,0)T:
The matrix H is solved for nonlinear least squares as a new unknown to recover and obtain a rotation matrix and translation vector r_t'.
The embodiments of the present application have been described above with reference to the accompanying drawings, in which the embodiments of the present application and features of the embodiments may be combined with each other without conflict, the present application is not limited to the above-described embodiments, which are merely illustrative, not restrictive, of the present application, and many forms may be made by those of ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are protected by the present application.
Claims (9)
1. Tunnel foundation ditch country rock displacement, crack and ponding change monitoring system, its characterized in that includes:
the track is paved at the top of the foundation pit;
the mobile trolley is arranged on the track and can carry out mobile inspection along the track;
The monitoring device is carried on the mobile trolley and used for obtaining information of the side wall of the foundation pit inspected by the mobile trolley;
The monitoring device scans the side wall of the foundation pit by using a laser radar and a structured light camera to acquire three-dimensional point cloud data, establishes a three-dimensional point cloud model of the wall of the foundation pit under a laser radar coordinate system, and accurately measures the whole and partial displacement deformation condition, side wall cracks and water seepage condition of the side wall of the foundation pit through regional division and data fitting.
2. The system for monitoring displacement, cracking and water accumulation of surrounding rock of tunnel foundation pit according to claim 1, wherein said monitoring device comprises:
A three-dimensional calibration module;
the image point cloud splicing and fusing module converts the image into three-dimensional point cloud data, splices the multi-frame point cloud into a three-dimensional model, performs smoothing and fusion on the spliced point cloud, and finally performs data processing to obtain a final three-dimensional model;
The model fitting module is used for selecting a corresponding linear or nonlinear model according to the data characteristics and a target to be fitted; secondly, calculating parameter values of the model by using a least square method, a maximum likelihood estimation method or other algorithms; evaluating the predictive performance of the model using the validation dataset and calculating a prediction error; if the prediction performance of the model does not meet the requirement, trying to fine-tune parameters of the model to improve the accuracy of the model; predicting values of unknown data using the resulting fitted model;
The detection module is used for preprocessing data, including noise removal and standardization; secondly, training an artificial intelligent model by using a machine learning algorithm, and evaluating the effect of the artificial intelligent model; then, adjusting model parameters to improve accuracy; the model was tested on a separate dataset and its accuracy was assessed.
3. The system for monitoring displacement, cracking and ponding change of foundation pit surrounding rock based on machine vision technology as claimed in claim 2, wherein the three-dimensional calibration module comprises:
The image acquisition unit is used for acquiring images of the calibration surface by using the structured light binocular camera;
A camera correction unit that corrects the acquired image;
the point cloud construction unit is used for carrying out three-dimensional reconstruction on the characteristic points on the calibration surface by utilizing the parallax image and the depth image to construct point cloud;
the parameter calibration unit is used for analyzing the point cloud data and calculating the internal parameters and the external parameters of the camera;
And the camera calibration unit is used for calibrating the internal and external parameters of the camera through the internal parameters and the external parameters calculated by the parameter calibration unit, wherein the internal parameters are a camera matrix and distortion coefficients, and the external parameters are a rotation matrix and translation vectors.
4. A machine vision technology based foundation pit surrounding rock displacement, cracking and ponding change monitoring system as claimed in claim 3, further comprising: and the data evaluation unit evaluates the calibration result and ensures the calibration accuracy.
5. The tunnel foundation pit surrounding rock displacement, crack and ponding change monitoring system according to claim 4, wherein,
Acquiring foundation pit side wall data under multiple measurements through a laser radar in the process of monitoring displacement changes of the foundation pit side wall, filtering noise data according to initial parameters of foundation pit measurement, and carrying out corresponding interpolation processing on the side wall data; and monitoring the position change condition of the side wall according to the change condition of the side wall data under the condition of comparing the shooting height with the unified height and measuring for a plurality of times, and calculating the displacement of the side wall.
6. The tunnel foundation pit surrounding rock displacement, crack and ponding change monitoring system according to claim 4, wherein,
In the process of monitoring cracks of the side wall of the foundation pit, acquiring the crack data of the side wall of the foundation pit by adopting a camera; firstly cutting an original image into k multiplied by k small blocks by adopting an image cutting mode, and then performing model training on the image size;
The crack height positioning algorithm adopts an SVR regression algorithm, and the heights corresponding to four corner points of different crack detection frames in the test picture are obtained by constructing fitting relation functions of characteristics such as foundation pit side wall parameters, the heights shot by a camera, coordinates of pixel points in the picture and the like and the actual heights corresponding to the pixel points in the picture, so that the height range of the crack is obtained.
7. The tunnel foundation pit surrounding rock displacement, crack and ponding change monitoring system according to claim 5, wherein,
The monitoring device extracts a plurality of groups of three-dimensional characteristic points in an image by utilizing RGB (red, green and blue) images and depth images of a reference target of a fixed point at the top of a foundation pit shot by a laser radar and a structured light camera, calculates a pose transformation matrix of the structured light camera relative to the first shot under the multi-measurement, and unifies a laser radar coordinate system under the multi-measurement to a laser radar coordinate system under the first scanning by combining a combined calibration algorithm of the structured light camera and the laser radar, so that unification of the multi-measurement coordinate system is realized.
8. The tunnel foundation pit surrounding rock displacement, crack and ponding change monitoring system according to claim 7, wherein,
The joint calibration algorithm adopts a calibration method based on plane constraint to perform joint calibration on the laser radar and the structured light camera.
9. The tunnel foundation pit surrounding rock displacement, crack and ponding change monitoring system according to claim 8, wherein,
The camera calculates the coordinates of the plane of the calibration plate under the camera coordinate system by identifying the labels on the calibration plate; the laser radar emits light beam to strike the calibration board, the plane constraint is constructed by utilizing the coordinates of the laser point under the laser coordinate system and the coordinates under the camera coordinate system to solve the external parameters,
The plane constraint calculation formula is as follows:
ncT(RPl+t)+dc＝0
Wherein, the vector n is a three-dimensional normal vector of the label plane, d is a distance from the origin of the camera coordinate system O c to the label plane, the coordinate of the point P under the laser coordinate system O l is P l, and the coordinate under the camera coordinate system O c is P c; the transformation of point P from laser coordinate system O l to camera coordinate system O c is:
Pc＝RT(Pl-t)
Substitution P l＝(x,y,0)T:
The matrix H is solved for nonlinear least squares as a new unknown to recover and obtain a rotation matrix and translation vector r_t'.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410109532.2A CN118111345A (en) | 2024-01-26 | 2024-01-26 | Tunnel foundation pit surrounding rock displacement, crack and ponding monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410109532.2A CN118111345A (en) | 2024-01-26 | 2024-01-26 | Tunnel foundation pit surrounding rock displacement, crack and ponding monitoring system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118111345A true CN118111345A (en) | 2024-05-31 |
Family
ID=91217262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410109532.2A Pending CN118111345A (en) | 2024-01-26 | 2024-01-26 | Tunnel foundation pit surrounding rock displacement, crack and ponding monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118111345A (en) |
-
2024
- 2024-01-26 CN CN202410109532.2A patent/CN118111345A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | Structural health monitoring and inspection of dams based on UAV photogrammetry with image 3D reconstruction | |
| Guan et al. | Automated pixel-level pavement distress detection based on stereo vision and deep learning | |
| Liu et al. | Image‐based crack assessment of bridge piers using unmanned aerial vehicles and three‐dimensional scene reconstruction | |
| Inzerillo et al. | Image-based 3D reconstruction using traditional and UAV datasets for analysis of road pavement distress | |
| Yang et al. | Structure monitoring and deformation analysis of tunnel structure | |
| Yu et al. | 3D reconstruction of road surfaces using an integrated multi-sensory approach | |
| RU2609434C2 (en) | Detection of objects arrangement and location | |
| Kwak et al. | Extrinsic calibration of a single line scanning lidar and a camera | |
| Rau et al. | Bridge crack detection using multi-rotary UAV and object-base image analysis | |
| CN104091168B (en) | Line of electric force based on unmanned plane image extracts localization method automatically | |
| Zhang et al. | 3D mapping of discontinuity traces using fusion of point cloud and image data | |
| CN104574393A (en) | Three-dimensional pavement crack image generation system and method | |
| Guldur | Laser-based structural sensing and surface damage detection | |
| Sarker et al. | Multi-scale 3D roughness quantification of concrete interfaces and pavement surfaces with a single-camera set-up | |
| Wang et al. | Automated pavement distress survey: a review and a new direction | |
| Ioli et al. | UAV photogrammetry for metric evaluation of concrete bridge cracks | |
| CN105023270A (en) | Proactive 3D stereoscopic panorama visual sensor for monitoring underground infrastructure structure | |
| Dhiman et al. | A multi-frame stereo vision-based road profiling technique for distress analysis | |
| CN103688289A (en) | Method and system for estimating a similarity between two binary images | |
| CN117036326A (en) | Defect detection method based on multi-mode fusion | |
| Guldur et al. | Damage detection on structures using texture mapped laser point clouds | |
| CN118111345A (en) | Tunnel foundation pit surrounding rock displacement, crack and ponding monitoring system | |
| CN115562284A (en) | Method for realizing automatic inspection by high-speed rail box girder inspection robot | |
| Kim et al. | An automatic robust point cloud registration on construction sites | |
| Johnston et al. | Estimating building floor plans from exterior using laser scanners |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |