Early warning method for concrete collapse in tunnel, controller and storage medium

Abstract

The invention discloses a method for early warning of collapse of concrete in a tunnel, a controller and a storage medium, and belongs to the technical field of engineering machinery. The method comprises the steps that in the operation process of the wet spraying machine, video data of a tunnel top construction face at the position where a constructor is located is collected, and the video data comprises multiple frames of construction face images; extracting angular points of each construction surface image; according to the angular points, screening matching point pairs among all the construction surface images to obtain a plurality of common feature points of the plurality of frames of construction surface images and coordinate data of each feature point in each construction surface image; according to the coordinate data of each feature point in each construction surface image, judging whether the concrete of the construction surface has a collapse risk or not; and outputting an early warning signal under the condition that the collapse risk exists. According to the invention, the early warning signal can be output in time by monitoring the concrete collapse condition of the top construction surface of the tunnel where the constructors are located in real time, so that the constructors can evacuate in time, and the safety of concrete construction in the tunnel is improved.

Description

Technical Field

[0001] This application relates to the technical field of construction machinery, and specifically relates to a method, a controller, and a storage medium for early warning of concrete collapse in a tunnel. Background Art

[0002] A wet shotcreting machine is a mobile construction machinery that sprays concrete doped with additives onto a construction surface such as a tunnel surface or a slope through high-pressure air. In recent years, due to the increasing requirements of the country for the degree of construction mechanization year by year, wet shotcreting machines have become essential equipment for tunnel construction. With the increase in the usage of wet shotcreting machines, there has been a frequent occurrence of safety incidents where equipment operators are hit by collapsed concrete in the tunnel, which can seriously lead to paralysis or even death. Currently, there are no targeted preventive measures for such accidents. Since construction workers' attention is usually focused on the spraying surface and they cannot notice the situation of the construction surface above their heads, if concrete collapse occurs, safety accidents are likely to occur. Therefore, there is a problem of relatively low safety in the current concrete construction in tunnels. Summary of the Invention

[0003] The purpose of the embodiments of this application is to provide a method, a controller, and a storage medium for early warning of concrete collapse in a tunnel, so as to solve the problem of relatively low safety in the current concrete construction in tunnels in the prior art.

[0004] To achieve the above purpose, the first aspect of this application provides a method for early warning of concrete collapse in a tunnel, and the method includes:

[0005] During the operation of the wet shotcreting machine, collect video data of the construction surface at the top of the tunnel where the construction worker is located, and the video data includes multiple frames of construction surface images;

[0006] Extract the corner points of each construction surface image;

[0007] According to the corner points, screen the matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image;

[0008] According to the coordinate data of each feature point in each construction surface image, determine whether there is a risk of collapse of the concrete on the construction surface;

[0009] In the case of a collapse risk, output a warning signal.

[0010] In the embodiments of the present application, collecting video data of the tunnel top construction surface where the construction workers are located includes: collecting an initial image containing the construction workers through an image acquisition device; based on a pre-trained object detection model, identifying the safety helmets worn by the construction workers in the initial image and determining the position information of the safety helmets; adjusting the shooting angle of the image acquisition device according to the position information of the safety helmets so that the image acquisition device collects video data of the tunnel top construction surface where the construction workers are located.

[0011] In the embodiments of the present application, extracting the corner points of each construction surface image includes: performing Gaussian blur processing on the construction surface image at different scales to generate multiple scale images; calculating the gradient of the image in a set direction on each scale image to obtain a gradient image; determining the gradient covariance matrix of each pixel point in the image according to the gradient image; determining the corner response value of each pixel point according to the gradient covariance matrix to obtain a corner response map; determining the local maximum points in the corner response map through a non-maximum suppression method to obtain the corner points of the construction surface image.

[0012] In the embodiments of the present application, according to the corner points, screening the matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image includes: determining the feature vector corresponding to each corner point through a preset feature description algorithm; matching the adjacent features between each pair of adjacent construction surface images through a preset feature matching algorithm to obtain an initial matching point pair; screening stable and correct target matching point pairs from the initial matching point pairs based on the random sample consensus algorithm; performing an intersection operation on the target matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images; determining the coordinate data of each feature point in each construction surface image.

[0013] In the embodiments of the present application, determining the coordinate data of each feature point in each construction surface image includes: determining the coordinate data of each feature point in each construction surface image according to the target matching point pair and a pre-determined transformation model.

[0014] In an embodiment of the present application, to determine whether there is a risk of collapse of the concrete on the construction surface based on the coordinate data of each feature point in each construction surface image, the method includes: determining the coordinate change situation of each feature point based on the coordinate data of each feature point in each construction surface image to screen out target feature points, where the target feature points are the feature points whose coordinates have changed; determining the average change amount, sample standard deviation, and sample size corresponding to the target feature points based on the coordinate data of the target feature points in each construction surface image; determining the test statistic of the target feature points based on the average change amount, sample standard deviation, and sample size; determining the preset threshold corresponding to the test statistic based on the preset confidence level and the degrees of freedom corresponding to the target feature points; determining that there is no risk of collapse of the concrete on the construction surface when the test statistic is less than the preset threshold; and determining that there is a risk of collapse of the concrete on the construction surface when the test statistic is greater than or equal to the preset threshold.

[0015] In an embodiment of the present application, to determine whether there is a risk of collapse of the concrete on the construction surface based on the coordinate data of each feature point in each construction surface image, the method includes: determining the transformation matrix between adjacent construction surface images based on the coordinate data of each feature point in each construction surface image through an image change estimation algorithm; mapping the feature points in the construction surface image to the same spatial coordinate system based on the transformation matrix; determining the displacement amount of each feature point in the spatial coordinate system; determining that there is a risk of collapse of the concrete on the construction surface when the displacement amount of any feature point is greater than or equal to the preset displacement threshold; and determining that there is no risk of collapse of the concrete on the construction surface when the displacement amounts of all feature points are less than the preset displacement threshold.

[0016] In an embodiment of the present application, before extracting the corner points of each construction surface image, the method further includes: preprocessing each construction surface image; performing image enhancement and dehazing processing on the preprocessed construction surface image through a preset dehazing algorithm; and adjusting the contrast of the construction surface image after the image enhancement and dehazing processing.

[0017] In an embodiment of the present application, the method further includes: continuously outputting a warning signal until there is no risk of collapse of the concrete on the construction surface at the top of the tunnel where the construction personnel are located.

[0018] A second aspect of the present application provides a controller, including:

[0019] a memory configured to store instructions; and

[0020] a processor configured to call instructions from the memory and capable of implementing the above method for early warning of concrete collapse in a tunnel when executing the instructions.

[0021] A third aspect of the present application provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to cause a machine to execute the method for tunnel concrete collapse warning described above.

[0022] Through the above technical solution, during the operation of the wetting machine, video data of the construction surface at the top of the tunnel where the construction personnel are located is collected, and then the corner points of each frame of the construction surface image in the video data are extracted. According to the corner points, matching point pairs between all construction surface images are screened to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image. Then, according to the coordinate data of each feature point in each construction surface image, it is determined whether there is a risk of collapse of the concrete on the construction surface. Finally, in the case of a collapse risk, a warning signal is output. The present application can monitor in real time the concrete collapse situation of the construction surface at the top of the tunnel where the construction personnel are located and output a warning signal in a timely manner, so that the construction personnel can evacuate in time, improving the safety of concrete construction on the top of the tunnel.

[0023] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation part. Description of the Drawings

[0024] The drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementation, they are used to explain the embodiments of the present application, but do not constitute a limitation to the embodiments of the present application. In the drawings:

[0025] Figure 1 It is a schematic flowchart of a method for tunnel concrete collapse warning provided by an embodiment of the present application

[0026] Figure 2 It is a structural block diagram of a controller provided by an embodiment of the present application. Specific Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiments of the present application and does not limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0028] It should be noted that if there are directional indications (such as up, down, left, right, front, back, etc.) involved in the embodiments of the present application, the directional indications are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the attached drawings). If the specific posture changes, the directional indications will also change accordingly.

[0029] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the fact that those skilled in the art can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present application.

[0030] Generally, a wet shotcreting machine is equipped with an operator and a laborer. The laborer is responsible for auxiliary work such as wiring, feeding, and shoveling the surface. When the wet shotcreting machine is in spraying operation, the laborer must operate the discharge rod of the mixer truck to control the amount of concrete in the hopper of the wet shotcreting machine. The operator stands not far from the spraying surface and constantly adjusts the distance and angle between the nozzle and the target spraying surface to prevent bulges, pits, etc. from appearing on the target spraying surface. They are both unable to take into account the situations in other parts of the tunnel.

[0031] The wet shotcreting operation in the tunnel is generally continuous construction for several sets of steel arch frames. The operator must operate the equipment boom to continuously spray towards the steel arch frames until the exposed steel arch frames are filled with concrete of a specified thickness. Although the addition of accelerator in the sprayed material can greatly accelerate the setting of the concrete, the concrete still takes several seconds to set. If the quality or dosage of the accelerator is insufficient, the setting time of the concrete can even reach several minutes. When the above situation occurs at the crown of the tunnel, large-area concrete collapse is likely to occur due to the influence of gravity. Since the operator's attention is concentrated on the spraying surface and cannot observe the situation above the head, if one happens to be in the collapse area at this time, safety accidents will inevitably occur, and in severe cases, even life will be endangered.

[0032] In the prior art, when a wet spraying machine operates in a tunnel, the following problems exist: First, the personnel configuration is simple. The operator and the laborer perform their respective duties, and it is impossible to take into account the possible concrete collapse. Second, it is difficult to ensure the consistency of the quality of concrete and accelerator. There is a possibility of large-area concrete collapse in the same construction section. Third, there is no obvious sign of concrete collapse. Due to the limited height in the tunnel, once a collapse occurs, the operator basically has no reaction time. If it is a large-area collapse, even if the operator is not located below, personal injury may be caused by the sputtering after the concrete hits the ground. In view of this, to improve the safety of the wet spraying machine during operation in the tunnel, this application proposes a method for warning of concrete collapse in the tunnel.

[0033] Figure 1 It is a schematic flowchart of a method for warning of concrete collapse in the tunnel provided by an embodiment of this application. As Figure 1 shown, an embodiment of this application provides a method for warning of concrete collapse in the tunnel. Taking the case where this method is applied to a processor as an example, this method may include the following steps:

[0034] Step 101, during the operation of the wet spraying machine, collect video data of the construction surface at the top of the tunnel where the construction personnel are located. The video data includes multiple frames of construction surface images.

[0035] Specifically, during the operation of the wet spraying machine, in order to monitor the safety condition of the construction surface at the top of the tunnel where the construction personnel are located, it is first necessary to collect the video data of this area. The video data contains multiple consecutive images, and each image reflects the state of the construction surface at a certain moment.

[0036] In one example, the camera can be installed on the wet spraying machine and mounted on the vehicle body of the wet spraying machine through an electric pan-tilt with excellent anti-shake function, and ensure that the activity area of the construction personnel is always within the shooting range of the camera. Since the position of the construction personnel is not fixed, in order to collect the video data of the construction surface at the top of the tunnel where the construction personnel are located in real time, an identifiable identifier can be set on the construction personnel, and the position of the identifier in the picture within the shooting range of the recognition camera can be recognized, and the shooting angle of the camera can be adjusted so that the camera can shoot the construction surface at the top of the tunnel where the construction personnel are located.

[0037] In another example, a camera can be installed on the safety equipment worn by construction workers, such as a safety helmet. The safety helmet is improved by adding an image acquisition device, which includes a camera, a wireless communication module, and a counterweight. Among them, the camera is used to collect video data; the wireless communication module is used to enable the camera to communicate with the processor to transmit the video data; the counterweight is arranged below the camera and is used to keep the shooting angle of the camera always upward by the action of gravity, regardless of the actions of the construction workers. In this way, no matter how the construction workers move, the video data of the construction surface above the heads of the construction workers can be collected in real time, ensuring the accuracy of the collected data.

[0038] Step 102: Extract the corner points of each construction surface image.

[0039] Specifically, after obtaining the video data, the next step is to extract the corner points from each frame of the construction surface image. Corner points refer to points with drastic changes in brightness in the image, usually the edges or inflection points of objects in the image. In one example, these corner points can be automatically identified from the image through corner detection algorithms such as Harris corner detection and Shi-Tomasi corner detection. Corner point extraction is an important step in image feature extraction and provides a basis for subsequent feature matching.

[0040] Step 103: According to the corner points, screen the matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image.

[0041] Specifically, after extracting the corner points of all construction surface images, it is necessary to screen out the matching point pairs between each image. In one example, this can be achieved through feature matching algorithms (such as SIFT, SURF, ORB, etc.), which can find similar feature points in different images and calculate the matching relationship between them. By screening the matching point pairs, multiple common feature points of multiple frames of construction surface images and the coordinate data of these feature points in each construction surface image can be obtained. These coordinate data reflect the positions of the feature points in the image and are important bases for subsequent judgment of the safety status of the construction surface.

[0042] Step 104: According to the coordinate data of each feature point in each construction surface image, judge whether there is a risk of collapse of the concrete on the construction surface.

[0043] Specifically, after obtaining the common feature points and their coordinate data of multiple frames of construction surface images, the risk of concrete collapse on the construction surface can be judged by analyzing the position changes of these feature points. In one example, the displacement vectors of the same feature points in two adjacent frames of images can be calculated, and the displacement conditions of all feature points can be counted. If there are abnormal changes in the magnitude or direction of the displacement vector, such as a sudden increase or a sudden change in direction, it may indicate that there is a risk of concrete collapse on the construction surface. In addition, other factors can be combined for comprehensive judgment, such as geological conditions, construction methods, etc.

[0044] Step 105, when there is a risk of collapse, output a warning signal.

[0045] Specifically, if it is judged through analysis that there is a risk of concrete collapse on the construction surface, a warning signal needs to be output immediately. The warning signal can be presented in the form of sound, light, text information, etc., to remind construction personnel to pay attention to safety and take corresponding measures. At the same time, the warning signal can also be transmitted to the remote monitoring center so that managers can timely understand the on-site situation and make decisions.

[0046] Through the above technical solution, during the operation of the wetting machine, video data of the construction surface at the top of the tunnel where the construction personnel are located is collected, and then the corner points of each frame of construction surface image in the video data are extracted. According to the corner points, matching point pairs between all construction surface images are screened to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image. Then, according to the coordinate data of each feature point in each construction surface image, it is judged whether there is a risk of concrete collapse on the construction surface. Finally, when there is a risk of collapse, a warning signal is output. This application can monitor the concrete collapse situation at the top of the tunnel where the construction personnel are located in real time and output a warning signal in time, so that the construction personnel can evacuate in time, improving the safety of concrete construction on the top of the tunnel.

[0047] In the embodiment of this application, collecting the video data of the construction surface at the top of the tunnel where the construction personnel are located includes: collecting an initial image containing the construction personnel through an image acquisition device; based on a pre-trained object detection model, identifying the safety helmet worn by the construction personnel in the initial image and determining the position information of the safety helmet; adjusting the shooting angle of the image acquisition device according to the position information of the safety helmet so that the image acquisition device can collect the video data of the construction surface at the top of the tunnel where the construction personnel are located.

[0048] It can be understood that the image acquisition device is installed on the body of the wetting machine through a pan-tilt with excellent anti-shake function, and it is ensured that the activity area of the construction workers is always within the shooting range of the camera. Since the positions of the construction workers are not fixed, in order to collect the video data of the tunnel top construction surface at the positions of the construction workers in real time, recognizable identifiers can be set on the construction workers. By identifying the positions of the recognizable identifiers in the shooting range of the recognition camera in the shooting picture, the shooting angle of the camera can be adjusted so that the camera can shoot the tunnel top construction surface at the positions of the construction workers. Since the construction workers must wear safety helmets that meet safety requirements during the operation process, and the shape of the safety helmet is highly recognizable in the tunnel, the positions of the construction workers can be determined by identifying the safety helmets through computer vision recognition technology, providing a reference orientation for the motion control of the pan-tilt and the photographing of the camera, so as to adjust the shooting angle of the image acquisition device.

[0049] Specifically, first, collect common safety helmet images on the market to form a sample data set, and then input the sample data into a deep learning model for object detection for training, so that the model has the function of identifying safety helmets to obtain a trained object detection model. Then load the trained target model into the system. During the actual construction process, the position information of the safety helmet can be obtained by inputting the real-time image collected by the camera into the object detection model. Based on the position information of the safety helmet, determine the relative position relationship between the construction worker and the tunnel top construction surface, and adjust the shooting angle of the image acquisition device, such as the pitch angle and yaw angle, etc., so that the image acquisition device can accurately collect the video data of the tunnel top construction surface at the position of the construction worker to ensure the accuracy of the subsequent tunnel top concrete collapse detection result.

[0050] In the embodiment of the present application, extracting the corner points of each construction surface image may include: performing Gaussian blur processing on the construction surface image at different scales to generate multiple scale images; calculating the gradient of the image in a set direction on each scale image to obtain a gradient image; determining the gradient covariance matrix of each pixel point in the image according to the gradient image; determining the corner point response value of each pixel point according to the gradient covariance matrix to obtain a corner point response map; determining the local maximum points in the corner point response map through a non-maximum suppression method to obtain the corner points of the construction surface image.

[0051] Specifically, perform Gaussian blur processing on the construction surface image at different scales to generate a series of images with different degrees of blur, that is, multiple scale images, to capture the features of the image at different scales, which helps improve the accuracy and robustness of subsequent corner detection. Then, calculate the gradient of the image in the set direction on each scale image through the gradient function to generate a gradient image. The gradient is the rate of change of a function in a certain direction at a certain point. In image processing, the gradient is usually used to describe the change of image brightness or color. The set direction refers to the x-direction and y-direction of the selected coordinate system. The gradient image can reflect the rate of change of image brightness and is an important basis for detecting image edges and corners. Further, according to the gradient image, calculate the gradient covariance matrix of each pixel point in the image. The gradient covariance matrix can describe the statistical characteristics of the gradient vectors around the pixel point and is an important basis for judging corners. Further, according to the gradient covariance matrix, calculate the corner response value of each pixel point to generate a corner response map. The corner response value is used to reflect the possibility of a pixel point being a corner and is the basis for subsequent corner screening. Finally, in the corner response map, determine the local maximum points through the non-maximum suppression method as the corners of the construction surface image. Non-maximum suppression can remove redundant points in the corner response map and retain the most prominent corners.

[0052] In this way, through the above steps of multi-scale Gaussian blur processing, gradient calculation, gradient covariance matrix determination, corner response value calculation, and non-maximum suppression, the corners in the construction surface image can be accurately extracted. These corners, as important feature points of the image, provide a basis for subsequent construction surface safety monitoring and analysis.

[0053] In the embodiment of the present application, according to the corners, screen the matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image, which may include: determining the feature vector corresponding to each corner through a preset feature description algorithm; matching the feature vectors between adjacent construction surface images through a preset feature matching algorithm to obtain initial matching point pairs; screening stable and correct target matching point pairs from the initial matching point pairs based on the random sample consensus algorithm; performing an intersection operation on the target matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images; and determining the coordinate data of each feature point in each construction surface image.

[0054] Specifically, the preset feature description algorithms can be SIFT, SURF, ORB, etc., which are used to determine the feature vectors corresponding to each corner point. The feature vector is an abstract representation of the image information around the corner point. It contains the uniqueness and stability information of the corner point and is the basis for subsequent feature matching. Preferably, the feature description algorithm in the embodiment of the present application can adopt SIFT. SIFT (Scale-Invariant Feature Transform) is a computer vision algorithm used to detect and describe local features in images. The preset feature matching algorithms can be brute-force matching, FLANN matching, etc., which are used to match feature vectors between each pair of adjacent construction surface images to obtain initial matching point pairs. Feature matching is the process of finding the same or similar feature points in different images and is the basis for tasks such as image stitching and 3D reconstruction. Further, since the initial matching point pairs may contain incorrect matches, to remove these incorrect matches, stable and correct target matching point pairs can be screened from the initial matching point pairs based on the Random Sample Consensus (RANSAC) algorithm. Finally, through intersection operation, the feature points that appear in multiple frames of images can be found, that is, multiple common feature points in multiple frames of construction surface images. These feature points are the key to subsequent tasks such as image stitching and 3D reconstruction.

[0055] Further, after determining multiple common feature points of all construction communications, it is necessary to determine the coordinate data of each feature point in each construction surface image, that is, pixel coordinates.

[0056] In one example, the spatial transformation matrix between each construction surface image can also be determined by the least squares method according to the target matching point pairs. According to the spatial transformation matrix, the feature points in each construction surface image are transformed from the original coordinates to the same spatial coordinate system, so as to obtain the pixel coordinates of these feature points in the new coordinate system, in order to obtain the pixel coordinates of the feature points in each construction surface image in the new coordinate system.

[0057] In the embodiment of the present application, determining the coordinate data of each feature point in each construction surface image may include: determining the coordinate data of each feature point in each construction surface image according to the target matching point pairs and a pre-determined transformation model.

[0058] Specifically, the pre-determined spatial transformation model can be the best transformation model obtained from RANSAC. Using the screened target matching point pairs and the transformation model, calculate the pixel coordinates of each common feature point in all images. Record the pixel coordinates of each common feature point in all images for subsequent analysis and processing.

[0059] In the embodiments of the present application, to determine whether there is a risk of collapse of the concrete on the construction surface based on the coordinate data of each feature point in the images of each construction surface may include: determining the coordinate change situation of each feature point based on the coordinate data of each feature point in the images of each construction surface to screen out target feature points, where the target feature points are the feature points with coordinate changes; determining the average change amount, sample standard deviation and sample size corresponding to the target feature points based on the coordinate data of the target feature points in the images of each construction surface; determining the test statistic of the target feature points based on the average change amount, sample standard deviation and sample size; determining the preset threshold corresponding to the test statistic according to the preset confidence level and the degrees of freedom corresponding to the target feature points; determining that there is no risk of collapse of the concrete on the construction surface when the test statistic is less than the preset threshold; and determining that there is a risk of collapse of the concrete on the construction surface when the test statistic is greater than or equal to the preset threshold.

[0060] It can be understood that after the concrete on the construction surface of the tunnel crown solidifies, theoretically, the coordinates of the feature points in the images of each construction surface are nearly unchanged, that is, as shown in formula (1):

[0061]

[0062] where n is the number of homologous feature points, and k is the k-th image.

[0063] Therefore, the target feature points with coordinate changes can be screened out through the coordinate data of the feature points in the images of each construction surface. Based on the coordinate data of the target feature points in the images of each construction surface, the average change amount, sample standard deviation and sample size corresponding to each target feature point can be calculated. Then, for the target feature points, the sample pivot quantity of the t-distribution and the confidence level α are used to judge the movement of the points, as shown in formula (2):

[0064]

[0065] where t is the test statistic; is the average change amount of the coordinate data of the feature points; μ is the population mean, that is, the assumed mean of no coordinate change, usually 0; S is the sample standard deviation; and n is the sample size, that is, the sample size.

[0066] Further, after determining the test statistic corresponding to the target feature points, set the preset confidence level according to actual needs, such as 95%. Then, according to the degrees of freedom corresponding to the target feature points, that is, the sample size minus 1, combined with the preset confidence level and the test statistic, look up the t-distribution table to determine the preset threshold tmax corresponding to the test statistic t, and compare the test statistic and the preset threshold; when t < tmax, it is determined that there is no risk of collapse of the concrete on the construction surface; when t ≥ tmax, it is determined that there is a risk of collapse of the concrete on the construction surface.

[0067] In the embodiments of the present application, to determine whether there is a risk of concrete collapse on the construction surface based on the coordinate data of each feature point in each construction surface image, it may include: determining the transformation matrix between adjacent construction surface images through an image change estimation algorithm according to the coordinate data of each feature point in each construction surface image; mapping the feature points in the construction surface image to the same spatial coordinate system based on the transformation matrix; determining the displacement amount of each feature point in the spatial coordinate system; determining that there is a risk of concrete collapse on the construction surface when the displacement amount of any feature point is greater than or equal to a preset displacement threshold; and determining that there is no risk of concrete collapse on the construction surface when the displacement amounts of all feature points are less than the preset displacement threshold.

[0068] Specifically, to determine whether there is a risk of collapse on the construction surface, based on the coordinate data of each feature point in each construction surface image, an image change estimation algorithm (such as affine transformation, perspective transformation, or more complex non-linear transformation algorithms) can be used to determine the transformation matrix between adjacent construction surface images. These matrices are used to describe the geometric relationship between the images. Then, based on the transformation matrix, the feature points in each construction surface image are mapped to a unified, global spatial coordinate system to ensure that the feature points in different images can be compared and analyzed in the same coordinate system. Further, in the spatial coordinate system, the displacement amount of each feature point relative to its initial position is calculated. The displacement amount reflects the change of the feature point in the time series or spatial position. Then, the displacement amount of each feature point in the spatial coordinate system is compared with a preset displacement threshold, which is set according to actual engineering experience, safety standards, and monitoring requirements. It reflects the allowable range of feature point displacement under normal circumstances. A displacement amount exceeding this range may indicate structural instability or potential safety risks. Specifically, if the displacement amount of any feature point is greater than or equal to the preset displacement threshold, which means that a significant unstable change has occurred in the structure, it is determined that there is a potential risk of concrete collapse on the construction surface. If the displacement amounts of all feature points are less than the preset displacement threshold, which means that the structure has remained relatively stable during the monitoring period, it is determined that there is no significant risk of concrete collapse on the construction surface.

[0069] In the embodiments of the present application, before extracting the corner points of each construction surface image, the method may further include: preprocessing each construction surface image; performing image enhancement and dehazing processing on the preprocessed construction surface image through a preset dehazing algorithm; and adjusting the contrast of the construction surface image after the image enhancement and dehazing processing.

[0070] Specifically, when the wet spraying machine is working, the mixture of concrete and accelerator will inevitably rebound when it is sprayed onto the tunnel surface by high-pressure air, resulting in a situation similar to smog. In a relatively closed tunnel, the smog generated on the working surface will exist throughout the construction process. In order to reduce the possible impact of smog on the system, the pictures collected by the camera are first processed by a defogging algorithm based on image enhancement, such as histogram equalization, wavelet transform, homomorphic filtering, etc., to reduce the photo noise as much as possible and improve the photo contrast to obtain a more ideal defogging photo, thereby improving the accuracy of subsequent detection results.

[0071] In an embodiment of the present application, the method may further include: continuously outputting a warning signal until there is no risk of collapse of concrete on the top construction surface of the tunnel where the construction personnel are located.

[0072] Specifically, to ensure the safety of construction workers, after the alarm signal is output, it is necessary to continuously detect whether there is a risk of collapse of the concrete on the top construction surface of the tunnel where the construction workers are located. If there is a risk of collapse, the early warning signal is continuously output until there is no risk of collapse of the concrete on the top construction surface of the tunnel where the construction workers are located.

[0073] In one example, the processor can communicate with an alarm device configured by a construction worker to send an execution warning instruction to the alarm device so that the alarm device outputs a warning signal. The alarm device can be set on a safety helmet worn by the construction worker, such as a helmet. The alarm device can also be a mobile phone of the construction worker.

[0074] In the embodiment of the present application, when the construction workers are working, the remote controller and the alarm device communicate via wireless radio frequency. In the embodiment of the present application, a rotor motor is arranged inside the transmitter of the remote controller. When the warning signal sent by the receiver is received via wireless radio frequency, the rotor motor is activated to remind the construction workers that there is a risk of concrete collapse at their current location. When the construction workers move, the system repositions, photographs, processes, judges or warns according to the new location of the construction workers until the alarm is eliminated when the movement of the tunnel arc top concrete is not detected at the location of the construction workers. In this way, the safety of the construction workers can be guaranteed.

[0075] Figure 2 This is a structural block diagram of a controller provided in an embodiment of the present application. Figure 2 As shown, an embodiment of the present application provides a controller, which may include:

[0076] Memory 210, configured to store instructions; and

[0077] The processor 220 is configured to call instructions from the memory 210 and implement the above-mentioned method for early warning of concrete collapse in a tunnel when executing the instructions.

[0078] Specifically, in the embodiments of the present application, the processor 220 may be configured to:

[0079] During the operation of the wetting machine, collect video data of the construction surface at the top of the tunnel where the construction worker is located, and the video data includes multiple frames of construction surface images;

[0080] Extract the corner points of each construction surface image;

[0081] According to the corner points, screen the matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image;

[0082] Based on the coordinate data of each feature point in each construction surface image, determine whether there is a risk of collapse of the concrete on the construction surface;

[0083] In the case of a collapse risk, output a warning signal.

[0084] Further, the processor 220 may also be configured to: collect an initial image including the construction worker through an image acquisition device; based on a pre-trained object detection model, identify the safety helmet worn by the construction worker in the initial image and determine the position information of the safety helmet; adjust the shooting angle of the image acquisition device according to the position information of the safety helmet so that the image acquisition device collects video data of the construction surface at the top of the tunnel where the construction worker is located.

[0085] Further, the processor 220 may also be configured to: perform Gaussian blur processing on the construction surface image at different scales to generate multiple scale images; calculate the gradient of the image in a set direction on each scale image to obtain a gradient image; determine the gradient covariance matrix of each pixel point in the image according to the gradient image; determine the corner response value of each pixel point according to the gradient covariance matrix to obtain a corner response map; determine the local maximum points in the corner response map through a non-maximum suppression method to obtain the corner points of the construction surface image.

[0086] Further, the processor 220 may also be configured to: determine the feature vector corresponding to each corner point through a preset feature description algorithm; match the features between adjacent construction surface images through a preset feature matching algorithm to obtain an initial matching point pair; screen stable and correct target matching point pairs from the initial matching point pairs based on the random sample consensus algorithm; perform an intersection operation on the target matching point pairs between all construction surface images to obtain multiple common feature points of multiple frames of construction surface images; determine the coordinate data of each feature point in each construction surface image.

[0087] Further, the processor 220 may also be configured to: determine the coordinate data of each feature point in each construction surface image according to the target matching point pair and a pre-determined transformation model.

[0088] Further, the processor 220 can also be configured to: determine the coordinate change situation of each feature point according to the coordinate data of each feature point in each construction surface image, so as to screen out the target feature points, where the target feature points are the feature points whose coordinates have changed; determine the average change amount, sample standard deviation and sample size corresponding to the target feature points according to the coordinate data of the target feature points in each construction surface image; determine the test statistic of the target feature points according to the average change amount, sample standard deviation and sample size; determine the preset threshold corresponding to the test statistic according to the preset confidence level and the degree of freedom corresponding to the target feature points; determine that the concrete on the construction surface has no collapse risk when the test statistic is less than the preset threshold; and determine that the concrete on the construction surface has a collapse risk when the test statistic is greater than or equal to the preset threshold.

[0089] Further, the processor 220 can also be configured to: determine the transformation matrix between adjacent construction surface images according to the coordinate data of each feature point in each construction surface image through an image change estimation algorithm; map the feature points in the construction surface image to the same spatial coordinate system based on the transformation matrix; determine the displacement amount of each feature point in the spatial coordinate system; determine that the concrete on the construction surface has a collapse risk when the displacement amount of any feature point is greater than or equal to the preset displacement threshold; and determine that the concrete on the construction surface has no collapse risk when the displacement amounts of all feature points are less than the preset displacement threshold.

[0090] Further, before extracting the corner points of each construction surface image, the processor 220 can also be configured to: preprocess each construction surface image; perform image enhancement and dehazing processing on the preprocessed construction surface image through a preset dehazing algorithm; and adjust the contrast of the construction surface image after the image enhancement and dehazing processing.

[0091] Further, the processor 220 can also be configured to continuously output a warning signal until the concrete on the tunnel top construction surface where the construction worker is located has no collapse risk.

[0092] Through the above technical solution, during the operation of the wetting machine, the video data of the tunnel top construction surface where the construction worker is located is collected, and then the corner points of each frame of construction surface image in the video data are extracted. According to the corner points, the matching point pairs between all construction surface images are screened to obtain multiple common feature points of multiple frames of construction surface images and the coordinate data of each feature point in each construction surface image. Then, according to the coordinate data of each feature point in each construction surface image, it is judged whether the concrete on the construction surface has a collapse risk. Finally, when there is a collapse risk, a warning signal is output. This application can monitor the collapse situation of the concrete on the tunnel top construction surface where the construction worker is located in real time and output a warning signal in time, so that the construction worker can evacuate in time, improving the safety of concrete construction on the tunnel top.

[0093] An embodiment of the present application further provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to cause a machine to execute the above method for early warning of concrete collapse in a tunnel.

[0094] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0096] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the specified functions in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0098] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0099] The memory may include non - permanent memory in the form of computer - readable media, random access memory (RAM) and / or non - volatile memory such as read - only memory (ROM) or flash RAM. The memory is an example of computer - readable media.

[0100] Computer - readable media includes permanent and non - permanent, removable and non - removable media and can store information by any method or technology. The information can be computer - readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase - change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read - only memory (ROM), electrically erasable programmable read - only memory (EEPROM), flash memory or other memory technologies, compact disc read - only memory (CD - ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non - transitory media that can be used to store information that can be accessed by a computing device. As defined herein, computer - readable media does not include transitory media such as modulated data signals and carrier waves.

[0101] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non - exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0102] The above are only embodiments of the present application and are not used to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for early warning of concrete collapse in a tunnel, characterized in that: The method comprises: During the operation of the spray wet machine, video data of the tunnel top construction surface where the construction workers are located is collected, wherein the video data includes multiple frames of construction surface images; Extracting corner points of each construction surface image; According to the corner points, matching point pairs between all the construction surface images are screened to obtain a plurality of common feature points of the multiple frames of construction surface images, and coordinate data of each of the feature points in each of the construction surface images; Judging whether there is a risk of collapse of concrete on the construction surface according to the coordinate data of each feature point in each construction surface image; In case of collapse risk, an early warning signal is output.

2. The method according to claim 1, characterized in that: The video data of the tunnel top construction surface where the construction workers are located is collected, including: Acquiring an initial image containing the construction personnel through an image acquisition device; Based on the pre-trained target detection model, identifying the safety helmet worn by the construction worker in the initial image, and determining the position information of the safety helmet; The shooting angle of the image acquisition device is adjusted according to the position information of the safety helmet, so that the image acquisition device can collect video data of the tunnel top construction surface where the construction personnel are located.

3. The method according to claim 1, characterized in that The step of extracting corner points of each construction surface image comprises: Performing Gaussian blur processing of different scales on the construction surface image to generate images of multiple scales; Calculating the gradient of the image in a set direction on each of the scale images to obtain a gradient image; Determine the gradient covariance matrix of each pixel in the image according to the gradient image; Determine the corner point response value of each pixel point according to the gradient covariance matrix to obtain a corner point response map; The local maximum points in the corner point response map are determined by a non-maximum suppression method to obtain the corner points of the construction surface image.

4. The method according to claim 1, characterized in that The step of screening the matching point pairs between all the construction surface images according to the corner points to obtain a plurality of common feature points of the multiple frames of construction surface images and the coordinate data of each feature point in each of the construction surface images includes: Determine the feature vector corresponding to each corner point by using a preset feature description algorithm; By using a preset feature matching algorithm, feature adjacency is matched between each pair of adjacent construction surface images to obtain an initial matching point pair; Selecting stable and correct target matching point pairs from the initial matching point pairs based on a random sampling consistency algorithm; Performing intersection operations on target matching point pairs between all construction surface images to obtain a plurality of common feature points of the multiple frames of construction surface images; Determine the coordinate data of each of the feature points in each of the construction surface images.

5. The method according to claim 4, characterized in that The determining of the coordinate data of each of the feature points in each of the construction surface images comprises: The coordinate data of each of the feature points in each of the construction surface images are determined according to the target matching point pairs and a predetermined transformation model.

6. The method according to claim 1, characterized in that The step of judging whether there is a risk of collapse of concrete on the construction surface according to the coordinate data of each feature point in each construction surface image includes: Determine, according to the coordinate data of each feature point in each construction surface image, the coordinate change of each feature point to screen the target feature point, wherein the target feature point is the feature point whose coordinate has changed; Determine the average change, sample standard deviation and sample size corresponding to the target feature point according to the coordinate data of the target feature point in each of the construction surface images; Determining a test statistic of the target feature point according to the average change, the sample standard deviation and the sample size; Determining a preset threshold corresponding to the test statistic according to a preset confidence level and the degree of freedom corresponding to the target feature point; When the test statistic is less than the preset threshold, it is determined that there is no risk of collapse of the concrete on the construction surface; When the test statistic is greater than or equal to the preset threshold, it is determined that there is a risk of collapse of the concrete on the construction surface.

7. The method according to claim 1, characterized in that The step of judging whether there is a risk of collapse of concrete on the construction surface according to the coordinate data of each feature point in each construction surface image includes: Determining the transformation matrix between adjacent construction surface images by an image change estimation algorithm according to the coordinate data of each feature point in each construction surface image; Based on the transformation matrix, the feature points in the construction surface image are mapped to the same spatial coordinate system; Determining the displacement of each of the feature points in the spatial coordinate system; When the displacement of any feature point is greater than or equal to a preset displacement threshold, it is determined that there is a risk of collapse of the concrete on the construction surface; When the displacements of all feature points are smaller than the preset displacement threshold, it is determined that there is no risk of collapse of the concrete on the construction surface.

8. The method according to claim 1, characterized in that: Before extracting the corner points of each of the construction surface images, the method further comprises: Preprocessing each of the construction surface images; Perform image enhancement and defogging processing on the pre-processed construction surface image using a preset defogging algorithm; Adjust the contrast of the construction surface image after image enhancement and dehazing.

9. The method according to claim 1, characterized in that: The method further comprises: The warning signal is continuously outputted until there is no risk of collapse of concrete on the top construction surface of the tunnel where the construction personnel are located.

10. A controller, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and implement the method for early warning of concrete collapse in a tunnel according to any one of claims 1 to 9 when executing the instructions.

11. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for causing a machine to execute the method for early warning of concrete collapse in a tunnel according to any one of claims 1 to 9.

Images