A method and device for quickly measuring deformation of tunnel excavation surface

Abstract

The present invention relates to a method and device for rapidly measuring tunnel excavation face deformation. The method comprises the following steps: obtaining a complete tunnel face image through binocular photography and performing preprocessing; performing tunnel face contour fitting on the preprocessed tunnel face image based on post-blast hole traces, extracting the tunnel face contour, and obtaining a tunnel face region; extracting key points of the tunnel cross-sectional contour based on the curvature variation characteristics of the tunnel face contour, wherein the key points of the tunnel cross-sectional contour include the arch crown and two arch feet; obtaining multiple reference points based on the tunnel cross-sectional contour key points, and calculating the three-dimensional spatial deformation of each reference point, wherein the deformation types of the three-dimensional spatial deformation include in-plane deformation and bulging deformation, and the reference points are selected based on the deformation type; and calculating the three-dimensional spatial deformation of any point within the tunnel face region based on the deformation of the three reference points. Compared with the existing technology, the present invention has the advantages of ensuring construction safety and reducing the amount of calculation.

Description

Technical Field

[0001] The present invention relates to the technical field of deformation identification after tunnel excavation, and in particular to a method and device for quickly measuring deformation of a tunnel excavation surface. Background Art

[0002] In actual tunnel construction, maintaining a constant understanding of the tunnel's stability is crucial for ensuring safe tunnel construction. Traditional face deformation monitoring techniques primarily rely on buried measuring instruments. Rock deformation is measured using levels, total stations, or by placing sliding micrometers within the rock. However, these methods all have drawbacks: First, deploying measuring instruments is labor-intensive, time-consuming, and inefficient. Second, measurement work interferes with on-site construction, and human factors significantly impact measurement accuracy, leading to unstable results. Finally, on-site operations require extended time and often involve harsh conditions, posing a threat to the safety of surveyors.

[0003] To avoid these drawbacks, non-contact measurement methods have been introduced in recent years, primarily photogrammetry based on digital photography. The application of digital photography in geotechnical engineering can improve the automation of monitoring, provide intuitive and detailed results, enhance measurement efficiency, and minimize disruption to construction. However, close-range photogrammetry still faces significant challenges:

[0004] (1) At present, all tunnel face close-range photogrammetry methods require the deployment of artificial markers on the tunnel face to be measured. For example, Chinese patent CN110849324B discloses a long-exposure oblique photography tunnel holographic measurement method, the implementation steps of which include marking image control points and measuring coordinates when drilling and staking out tunnel faces; collecting tunnel body images with different angles, high resolution, and rock stratum textures in a low-light environment through long-exposure photography; identifying feature points on multiple tunnel body images, calculating the spatial coordinates of the feature points based on the multi-image spatial forward intersection method, and developing the points to form a grid surface model, thereby obtaining a digital surface model carrying stratum information and the surface undulations of the tunnel body contour. However, most artificial markers are made of reflective materials and are mainly installed by pasting or drilling. In actual projects, laying out markers on the face is time-consuming and requires workers to frequently touch and be exposed to the face for a long time, which poses a great safety risk. For the area above the face, installation is difficult, takes a lot of time, and interferes with normal tunnel construction operations. This measurement method of installing artificial markers clearly violates the guiding principle of "timely closure" in tunnel construction.

[0005] (2) Research results of a few researchers in the field of photogrammetry without marker points. For example, the paper “Screen face deformation monitoring technology based on sub-pixel non-marked method” proposed a non-marked sub-pixel monitoring method based on the grayscale image correlation algorithm, designed an indoor experiment, set the corner points as feature points, and analyzed photos without light changes and no unexpected disturbances. However, the existing non-marked face deformation detection technology all adopts a single-camera fixed monitoring solution. This technical means is not practical in the actual tunnel construction process and will seriously interfere with the subsequent construction of the face. On the other hand, the tunnel face deformation obtained by using a single-camera technical solution is a two-dimensional plane deformation, and the tunnel face bulging deformation data is missing.

[0006] (3) Conventional calculations of the three-dimensional spatial deformation of the tunnel face are all based on traditional binocular photography technology. It is necessary to solve the three-dimensional coordinates of each pixel point on the tunnel face and compare each image one by one to solve the spatial deformation of the pixel point. However, this calculation method is computationally intensive, time-consuming, and has poor timeliness. Summary of the Invention

[0007] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a method and device for quickly measuring the deformation of a tunnel excavation surface, which ensures construction safety and has a small amount of calculation.

[0008] The purpose of the present invention can be achieved by the following technical solutions:

[0009] A method for quickly measuring deformation of a tunnel excavation face comprises the following steps:

[0010] Obtain complete tunnel face images through binocular photography and perform preprocessing;

[0011] Based on the blast hole traces after blasting, the pre-processed tunnel face image is fitted with the tunnel face contour, the tunnel face contour is extracted, and the tunnel face area is obtained;

[0012] Extracting key points of the tunnel cross-section profile based on the curvature change characteristics of the tunnel face profile, wherein the key points of the tunnel cross-section profile include the arch crown and two arch feet;

[0013] A plurality of reference points are obtained based on the key points of the tunnel cross-section profile, and a three-dimensional spatial deformation of each reference point is calculated, wherein the deformation type of the three-dimensional spatial deformation includes in-plane deformation and bulging deformation, and the reference point is selected based on the deformation type;

[0014] The three-dimensional deformation of any point in the tunnel face area is calculated based on the deformation of three reference points.

[0015] Furthermore, the preprocessing includes image denoising and image size unification.

[0016] Furthermore, performing tunnel face contour fitting on the pre-processed tunnel face image based on the blast hole traces after blasting specifically includes:

[0017] The K3M sequential iterative processing algorithm is used to extract the blasthole skeleton by gradually performing an erosion operation from the periphery of the target image to the center of the target until the erosion reaches the width of a single layer of pixels.

[0018] Extract the endpoints of each blasthole skeleton one by one and filter to obtain the skeleton endpoints;

[0019] Based on the curvature change characteristics of the fitting curve, outliers are deleted from the skeleton endpoints obtained by screening to obtain contour points;

[0020] The contour points are used as fitting data and the tunnel face contour is fitted based on the least square method.

[0021] Furthermore, the screening to obtain the skeleton endpoints is specifically:

[0022] The blasthole skeleton is binarized to obtain the two end points of the blasthole skeleton. The center pixel of the shooting picture is used as the reference pixel. The Euclidean distances from the two end points to the reference pixel are calculated respectively, and the endpoints with the smaller distance are selected as the skeleton endpoints.

[0023] Furthermore, the abnormal point is obtained based on the vector angle judgment, specifically:

[0024] A rectangular coordinate system is established based on the point B to be determined and two points A and C adjacent to the point to be determined, with the AC direction as the x-axis and the y-axis established with the midpoint of AC as the coordinate origin O;

[0025] Determine whether the angle between vector AB and the positive direction of the y-axis is greater than 90 degrees. If so, the point B to be determined is an abnormal point.

[0026] Furthermore, the extraction method of the vault is:

[0027] Determine the vertical coordinate value of each contour pixel point, and take the pixel point with the largest vertical coordinate value as the arch top;

[0028] The extraction method of the arch foot is:

[0029] For step excavation, the pixel point where the curvature of the contour pixel suddenly changes to 0 is the arch foot;

[0030] For full-section excavation, the pixel point where the curvature of the contour pixel suddenly changes to below the curvature threshold is the arch foot.

[0031] Furthermore, for the in-plane deformation, the reference points are selected as follows:

[0032] Based on the arch crown and arch foot, the tunnel face area is divided into a middle triangular area and two side arc areas;

[0033] For the middle triangular area, the arch top and two arch feet are used as three reference points;

[0034] For the arc-shaped areas on both sides, the arc segment midpoint, the arch top and an arch foot closer to the arc segment midpoint of the arc segment are used as three reference points.

[0035] Furthermore, for the bulging deformation, extracting the reference points of the tunnel face area specifically includes:

[0036] The midpoint of the core area of ​​the tunnel face is identified and used as a reference point. Any two adjacent points among the midpoints of the arch crown, the two arch feet, and the two arc segments of the arc area between the arch crown and the two arch feet are used as the other two reference points to form three reference points.

[0037] Furthermore, the specific calculation process of the three-dimensional spatial deformation of the reference point includes:

[0038] Based on two images taken simultaneously by binocular photography, the SGBM algorithm is used to calculate the disparity between the two images. Based on the geometric relationship of parallel binocular vision, the three-dimensional spatial information of the image is calculated.

[0039] Based on the comparison of multiple images, the three-dimensional spatial deformation of the reference point is calculated.

[0040] The present invention also provides a device for rapid detection of tunnel excavation face deformation, comprising a binocular camera and an industrial computer, wherein the industrial computer stores one or more executable programs, and the one or more programs include instructions for executing the above-mentioned method for rapid measurement of tunnel excavation face deformation.

[0041] Compared with traditional photogrammetry technology, this method does not require the deployment of manual markers. Instead, it uses digital image technology to automatically identify stable geometric feature points on the tunnel face, thereby quickly identifying tunnel face deformation. This method has the following beneficial effects:

[0042] (1) The present invention uses binocular photography technology to identify geometric feature points on the tunnel face as tracking and calculation objects, without the need to manually set landmarks, thus avoiding frequent contact and prolonged exposure of workers to the tunnel face, and reducing measurement and personnel safety risks. At the same time, it also solves the problems of difficult installation work in the area above the tunnel face, which requires a lot of time and interferes with normal tunnel construction operations. It can ensure construction safety and greatly improve work efficiency.

[0043] (2) The present invention adopts a dual-camera shooting scheme, with stations freely set up within a fixed distance in front of the tunnel face. Compared with fixed-station and timed shooting schemes, the present invention will not interfere with the normal construction of the tunnel project. On the other hand, compared with the currently commonly used single-camera measurement scheme, binocular photography technology can solve the three-dimensional spatial coordinates of the tunnel face feature points, and then accurately calculate the three-dimensional spatial deformation of the tunnel face. This provides tunnel face bulging deformation data for subsequent evaluation of the surrounding rock geological conditions in front of the tunnel face and establishment of tunnel face stability criteria, thus establishing a more complete and reasonable tunnel face stability judgment system.

[0044] (3) Compared with the traditional binocular camera method for calculating the deformation of the palm face, the present invention first calculates the deformation of the reference point, and then calculates the deformation of any part based on the three-point coordinates and the deformation value, which can avoid a lot of calculations and greatly improve the measurement efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of a flow chart of the present invention;

[0046] Figure 2 Schematic diagram of the connected domain of 8 endpoint pixels of the blasthole skeleton (taking the horizontal skeleton as an example), where (2a) is the left endpoint, (2b) is the middle point, and (2c) is the right endpoint;

[0047] Figure 3 This is a schematic diagram of the normal endpoint cross-section profile fitting effect;

[0048] Figure 4 This is a schematic diagram of the fitting effect of the abnormal endpoint cross-section contour;

[0049] Figure 5 Schematic diagram of the outlier determination model;

[0050] Figure 6 Schematic diagram of tunnel face contour fitting and image segmentation extraction, where (6a) is the blast hole trace, (6b) is the blast hole skeleton extraction, (6c) is the tunnel face contour fitting, and (6d) is the tunnel face image region segmentation;

[0051] Figure 7 This is a schematic diagram for angle calculation;

[0052] Figure 8 Schematic diagram of the coordinate relationship of each point for calculating the deformation in the plane;

[0053] Figure 9 This is a schematic diagram of the bulging deformation calculation partition;

[0054] Figure 10 Schematic diagram of tunnel face bulging deformation calculation. DETAILED DESCRIPTION

[0055] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0056] Explanation of terms

[0057] Tunnel face: Tunnel face, also known as tunnel face, is a term used in tunnel construction. It refers to the working face that continuously advances during tunnel excavation (coal mining, mining or tunnel engineering).

[0058] Binocular photogrammetry: Positioning is performed using two cameras. For a specific point on an object, two cameras fixed at different locations capture the object's image and obtain the coordinates of that point on the image planes of the two cameras. As long as the precise relative position of the two cameras is known, geometric methods can be used to determine the coordinates of the feature point in the coordinate system of one fixed camera, thus determining the location of the feature point.

[0059] Marking points: They are placed on the surface of the object being measured and serve as control points or measured points during the measurement process to improve measurement accuracy and reliability.

[0060] Example

[0061] The present invention is based on binocular photogrammetry technology. It does not require manual markers or scales, but uses image processing algorithms to automatically identify and track key points on the tunnel face. Based on the spatial deformation of a small number of key points, three-point deformation data is used to calculate the plane deformation and bulging deformation of any part within the tunnel face area, thus enabling rapid measurement of the three-dimensional spatial deformation of the tunnel face.

[0062] like Figure 1 As shown, this embodiment provides a method for quickly measuring deformation of a tunnel excavation face, comprising the following steps:

[0063] S1. Obtain a complete tunnel face image through binocular photography.

[0064] Two cameras are deployed on both sides of the tunnel for synchronous data collection. The best distance between the cameras and the tunnel face is 30m to 50m to ensure that the complete tunnel face image can be captured. During the shooting process, the fill light is turned on, and fill light equipment is equipped when necessary to ensure that the entire tunnel face is clearly visible and free of shadows.

[0065] Binocular photography technology can be used to calculate the three-dimensional spatial coordinates of the characteristic points of the tunnel face, and then calculate the three-dimensional spatial deformation of the tunnel face; it provides tunnel face bulging deformation data for subsequent evaluation of the surrounding rock geological conditions in front of the tunnel face and establishment of tunnel face stability criteria, and establishes a more complete and reasonable tunnel face stability judgment system.

[0066] S2. Tunnel face image preprocessing.

[0067] Image preprocessing primarily involves image denoising and image resizing. Image noise can be categorized as Gaussian and impulse noise due to the harsh tunnel construction environment, including dust, light source interference, and interference from the camera itself. A combination of Gaussian low-pass filtering and median filtering is employed. First, Gaussian low-pass filtering is used to eliminate Gaussian noise, followed by median filtering to eliminate impulse noise. Finally, the filtered tunnel face image undergoes pixel transformation, image resizing, and format standardization.

[0068] S3. Extraction of the tunnel face contour.

[0069] Taking the columnar blasthole traces after blasting as the identification object, the skeleton of each columnar blasthole trace is extracted using the skeleton extraction algorithm, and the skeleton endpoints located on the tunnel face contour are extracted; based on the curvature change characteristics of the fitting curve, a screening judgment criterion is proposed on the basis of experiments, and the extracted endpoints are judged one by one to exclude abnormal points such as sharp points and mutation points caused by falling blocks and danger removal, and the tunnel face contour is fitted after screening.

[0070] like Figure 6 As shown in the figure, the tunnel face contour extraction specifically includes:

[0071] (1) Blasthole skeleton extraction

[0072] The K3M sequential iterative processing algorithm is used to gradually perform an erosion operation from the periphery of the target image to the center of the target until the erosion reaches the width of a single layer of pixels, thereby extracting the blasthole skeleton. The main operation process is as follows:

[0073] 1) For a single blasthole operation, extract the target contour points and record them;

[0074] 2) Check the 8-pixel neighborhood of the contour point in turn to see if it contains only 3 connected pixels. If so, delete this point from the contour point and erode the corresponding point in the target image;

[0075] 3) Check the 8-pixel neighborhood of the remaining contour points in 2) to see if they contain only 3 or 4 connected pixels. If so, delete the point from the contour points and erode the corresponding point in the target image.

[0076] 4) Check the 8-pixel neighborhood of the remaining contour points in 3) to see if they contain only 3, 4, or 5 connected pixels. If so, delete this point from the contour points and delete the corresponding points in the target image.

[0077] 5) Check the 8-pixel neighborhood of the remaining contour points in 4) to see if they contain only 3, 4, 5, or 6 connected pixels. If so, delete the point from the contour points and erode the corresponding point in the target image.

[0078] 6) Check the 8-pixel neighborhood of the remaining contour points in 5) to see if they contain only 3, 4, 5, 6, or 7 connected pixels. If so, delete the point from the contour points and erode the corresponding point in the target image.

[0079] 7) Repeat the above operation until there is no object that can be corroded, then end the operation and obtain the blasthole skeleton.

[0080] Using the above processing method, morphological pixel corrosion operations are performed on all identified blastholes one by one to obtain and extract the skeleton of each blasthole.

[0081] (2) Extraction of blasthole skeleton endpoints

[0082] Based on the above blasthole skeleton extraction results, the skeletons are operated one by one to extract the endpoints in preparation for the subsequent cross-section contour fitting.

[0083] 1) Binarize the blasthole skeleton image, with the skeleton in black and the background in white; perform 8-connected domain judgment on each skeleton pixel: if there are two black pixels, it is an endpoint; if the number of black pixels exceeds two, it is an intermediate point; obtain the endpoints of the skeleton, such as Figure 2 shown.

[0084] 2) Taking the center pixel of the captured image as the reference, calculate the Euclidean distances from the two endpoints to the reference pixel, as shown in formula (1). Compare the distance values, eliminate the endpoints with larger distances, and retain the endpoints with smaller distances as the subsequent contour fitting points.

[0085]

[0086] Where: D - Euclidean distance between two pixels;

[0087] p——endpoint, pixel coordinates are (x, y);

[0088] q - reference, pixel coordinates are (s, t).

[0089] (3) Outlier screening method based on curvature change characteristics

[0090] Based on the geometric characteristics of highway tunnel cross-sections, the designed cross-section is a standard three-center circle combination. After blasting, the tunnel cross-section has no time to deform, and the initial contour remains essentially circular. Due to the uncertainty of blasting quality control, the blasthole endpoints may be of poor quality. Furthermore, due to the presence of joints, the tunnel roof may experience a tendency to fall off. These factors can all lead to sharp points and sudden changes in the initial cross-section fitting process. By eliminating these outliers, fitting accuracy can be effectively improved.

[0091] Assume that B' is an abnormal point caused by the top block falling, which will be disturbed and slide down as a whole during the subsequent construction process. It needs to be eliminated when fitting the initial section.

[0092] Depend on Figure 3 It can be seen that the normal cross-section profile has a smooth curve from A to B to C, and the overall curvature changes little; the fitting curve with abnormal points has large curvature changes at points A, B', and C, as shown in Figure 2. Figure 4 shown.

[0093] Therefore, the abnormal point can be determined based on the degree of curvature change. On the other hand, direct curvature calculation is relatively complex and the judgment standard is difficult to determine. Therefore, the present invention converts it into a method of judging the vector angle and provides a quantitative judgment standard.

[0094] like Figure 5 As shown, A, B, and C are three adjacent points, and the arc ABC is the color trace contour curve. A coordinate system is established as shown in the figure, with the AC direction as the x-axis and the midpoint of AC as the coordinate origin O.

[0095] For the designed contour, the angle between vector AB and the positive y-axis is less than 90 degrees. Due to construction factors, geological factors, and image processing technology, the position of the actual extracted blasthole endpoint B' may vary, the curvature of the fitted arc AB'C may change, and the angle between vector AB' and the positive y-axis may also change.

[0096] Based on a large number of fitting experiments, it is believed that when point B' is located in the 1st and 2nd quadrants of the coordinate system, the curve fitting accuracy is not greatly affected, and the vector angle θ1 is less than 90 degrees. When point B' is located in the 3rd and 4th quadrants of the coordinate system, the fitting result is poor and can be judged as an abnormal point. At this time, the vector angle θ2 is greater than 90 degrees.

[0097] The following formula can be used to calculate the angle:

[0098]

[0099] (4) Tunnel face profile fitting based on least squares method

[0100] Based on the above-screened blasthole endpoints as fitting data, the least squares method can be used to perform piecewise circular fitting for the initial cross-sectional profile, ultimately combining the three-center circle tunnel cross-sectional profile. For the subsequent deformed cross-section, piecewise polynomial fitting can be used.

[0101] Regarding the least squares fitting of the circle, the following algebraic distance square sum and its correction formula are used:

[0102]

[0103] Where: xi ——The horizontal coordinate of the fitting data point;

[0104] y i ——Yordinal coordinate of fitting data point;

[0105] x c ——The horizontal coordinate of the center of the fitting circle;

[0106] y c ——The vertical coordinate of the center of the fitting circle.

[0107] S4. Extraction of the reference points of the tunnel face.

[0108] Based on the extracted curvature variation characteristics of the tunnel face profile, key points of the tunnel cross-section profile, including the crown and two arch feet, are extracted. Multiple reference points are then derived from these key points. The selection of these reference points is determined by the type of 3D deformation, including in-plane and bulging deformation. Different reference points are selected for each deformation type.

[0109] In order to understand the convergence and deformation of the tunnel face, it is necessary to extract the arch crown and arch foot points of the tunnel face. For different construction conditions, the following regulations are made:

[0110] 1) For bench excavation, only part of the tunnel face can be photographed, and the intersection of the side wall and the ground surface is used as the arch foot for deformation monitoring.

[0111] 2) For full-section excavation, the entire section can be photographed, and the connection between the invert arch and the reinforcement can be used as the arch foot for deformation monitoring.

[0112] Accordingly, the determination of the arch crown and arch foot is as follows:

[0113] (1) Vault determination

[0114] The vault is at the top of the section, and this feature can be used directly to determine the vertical coordinate value of each contour pixel. The pixel with the largest vertical coordinate value is the vertex, which is used for subsequent section convergence calculations.

[0115] (2) Judgment of arched feet

[0116] For the segmented tunnel face image, the curvature change characteristics are used as the evaluation criteria to identify the arch top and arch foot.

[0117] For bench excavation, the arch foot is the intersection of the arc and the ground. For full-face excavation, the arch foot is the intersection of the arc and the inverted arch. Both have significant curvature changes. Using the boundary pixel curvature value calculation method, the curvature value mutation point is recorded and marked as the arch foot.

[0118] The steps are as follows: Figure 7As shown, at any point, take two vectors with the same magnitude ω before and after the point and calculate the angle θ between them. That is, take the ωth point before and after the point and form two vectors v1 and v2 with the point respectively. Then the angle between v1 and v2 is:

[0119]

[0120] The method for determining the arch foot is as follows:

[0121] 1) For the step method, the ground is a straight line with a curvature of 0. When the curvature suddenly changes to 0, it is marked as an arch foot.

[0122] 2) For the full-section method, the inverted arch is significantly flatter than the vault, spandrel, and sidewall, and its curvature is significantly smaller. When the curvature suddenly changes to a smaller value and the subsequent curvature continues to remain at a smaller value, this point is marked as the arch foot.

[0123] S5. Based on image matching and coordinate solution, calculate the three-dimensional spatial deformation of the reference point.

[0124] Based on two simultaneous images obtained through binocular photography, the SGBM algorithm is used to calculate the parallax of the two images. Based on the geometric relationship of parallel binocular vision, the three-dimensional spatial information of the images is calculated. Based on the comparison of multiple images, the three-dimensional spatial deformation of the reference point is calculated.

[0125] S6. Calculation of tunnel face deformation. Calculating the three-dimensional coordinates of each point on the tunnel face is computationally intensive and time-consuming, which cannot meet the requirements for timely monitoring in engineering projects. This embodiment uses key points such as the crown, arch foot, and tunnel face center as calculation reference points, divides the tunnel face into regions, and uses deformation data from these three reference points to calculate deformation values ​​for each part of the tunnel face. This avoids extensive computation and allows for rapid determination of tunnel face deformation.

[0126] According to different deformation types, the tunnel face deformation calculation is realized based on the three selected reference points, specifically:

[0127] (1) In-plane deformation calculation method

[0128] 1) Divide the entire tunnel face into a central triangular area and two arc-shaped areas on both sides, such as Figure 8 As shown;

[0129] 2) For the middle triangular area, the deformation of the arch top and the arch feet on both sides can be used as reference points, denoted as A, B, and C respectively, to calculate the deformation value of any point P inside;

[0130] 3) For the arc-shaped areas on both sides, the midpoints of the arc segments are added as reference points, denoted as D and E. The arch crown, arch foot, and arc segment midpoints are used as reference points to calculate the deformation of the arc-shaped areas.

[0131] Taking the middle triangle area as an example, the lateral displacement u of point A is known. A , longitudinal displacement v A , lateral displacement u of point B B , longitudinal displacement v B , lateral displacement u of point C C , longitudinal displacement v C The displacement (u, v) of any point P on the tunnel face can be expressed as a polynomial combination of coordinates:

[0132]

[0133] For three reference points:

[0134]

[0135] v C =β4+β5x C +β6y C (6-2)

[0136] The horizontal and vertical displacements are organized into matrix form and transformed and solved to obtain the displacement of point P:

[0137]

[0138]

[0139] Where:

[0140] Under the condition that the coordinates and deformation values ​​of the reference points A, B, and C are known, the plane deformation value of any point P in the ABC area can be quickly solved according to the above formula.

[0141] For the arc-shaped areas on both sides, the above formula can be used to solve the problem based on the reference points A, B, D or A, C, E.

[0142] (2) Calculation method of bulging deformation

[0143] For the bulging deformation of the tunnel face, since the tunnel face size is large, the bulging deformation is small. Although the bulging surface is arc-shaped, the curvature radius is extremely large and approximates a straight line. The interpolation method is used for approximate calculation.

[0144] The division of the tunnel face area is as follows Figure 9 As shown in the figure, image processing is used to identify the midpoint of the core area of ​​the tunnel face, marked as O, and set as a reference point. In addition, any two adjacent points among the midpoints of the arch crown, the two arch feet, and the two arc segments of the arc area between the arch crown and the two arch feet are used as the other two reference points to form three reference points.

[0145] The specific calculation method is as follows:

[0146] Take regional OBD as an example, Figure 10 As shown in the figure, plane ODB is the initial tunnel face, and plane O′D′B′ is the deformed tunnel face. Under the conditions that the coordinates of the three reference points O, D, and B and the bulging deformation are known, the deformation is calculated using the principle of similar triangles.

[0147] Inside the right trapezoid B′BDD′:

[0148]

[0149] Inside the right trapezoid H′HOO′:

[0150]

[0151] Through the above formula, the length of PP′ can be calculated, which is the bulging deformation of point P.

[0152] If the above method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0153] Example 2

[0154] This embodiment provides a device for rapidly detecting tunnel excavation face deformation, comprising a binocular camera and an industrial computer. The industrial computer stores one or more executable programs, each of which includes instructions for executing the method for rapidly measuring tunnel excavation face deformation described in Example 1. The binocular cameras are mounted on brackets at appropriate locations on either side of the tunnel. Preferably, two cameras are positioned within 30-50 meters of the tunnel's front. The cameras are connected to the industrial computer to enable local, real-time transmission and processing of captured images.

[0155] In a preferred embodiment, the device further includes a fill light device to provide fill light to the tunnel face.

[0156] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A method for rapid measurement of tunnel excavation face deformation, characterized in that: The following steps are involved: Obtain complete tunnel face images through binocular photography and perform preprocessing; Based on the blast hole traces after blasting, the pre-processed tunnel face image is fitted with the tunnel face contour, the tunnel face contour is extracted, and the tunnel face area is obtained; Extracting key points of the tunnel cross-section profile based on the curvature change characteristics of the tunnel face profile, wherein the key points of the tunnel cross-section profile include the arch crown and two arch feet; A plurality of reference points are obtained based on the key points of the tunnel cross-section profile, and a three-dimensional spatial deformation of each reference point is calculated, wherein the deformation type of the three-dimensional spatial deformation includes in-plane deformation and bulging deformation, and the reference point is selected based on the deformation type; Calculate the three-dimensional deformation of any point in the tunnel face area based on the deformation of three reference points; The step of performing tunnel face contour fitting on the pre-processed tunnel face image based on the blast hole traces after blasting specifically includes: The K3M sequential iterative processing algorithm is used to extract the blasthole skeleton by gradually performing an erosion operation from the periphery of the target image to the center of the target until the erosion reaches the width of a single layer of pixels. Extract the endpoints of each blasthole skeleton one by one and filter to obtain the skeleton endpoints; Based on the curvature change characteristics of the fitting curve, outliers are deleted from the skeleton endpoints obtained by screening to obtain contour points; Using the contour points as fitting data, performing tunnel face contour fitting based on the least squares method; For the in-plane deformation, the reference points are selected as follows: Based on the arch crown and arch foot, the tunnel face area is divided into a middle triangular area and two side arc areas; For the middle triangular area, the arch top and two arch feet are used as three reference points; For the arc-shaped areas on both sides, the arc segment midpoint, the arch top, and an arch foot closer to the arc segment midpoint are used as three reference points; For the bulging deformation, extracting the reference points of the tunnel face area specifically includes: The midpoint of the core area of ​​the tunnel face is identified and used as a reference point. Any two adjacent points among the midpoints of the arch crown, the two arch feet, and the two arc segments of the arc area between the arch crown and the two arch feet are used as the other two reference points to form three reference points.

2. The method for rapid measurement of tunnel excavation face deformation according to claim 1, characterized in that: The preprocessing includes image denoising and image size unification.

3. The method for rapid measurement of tunnel excavation face deformation according to claim 1, characterized in that: The screening to obtain the skeleton endpoints is specifically: The blasthole skeleton is binarized to obtain the two end points of the blasthole skeleton. The center pixel of the shooting picture is used as the reference pixel. The Euclidean distances from the two end points to the reference pixel are calculated respectively, and the endpoints with the smaller distance are selected as the skeleton endpoints.

4. The method for rapid measurement of tunnel excavation face deformation according to claim 1, characterized in that: The abnormal point is obtained based on the vector angle judgment, specifically: A rectangular coordinate system is established based on the point B to be determined and two points A and C adjacent to the point to be determined, with the AC direction as the x-axis and the y-axis established with the midpoint of AC as the coordinate origin O; Determine whether the angle between vector AB and the positive direction of the y-axis is greater than 90 degrees. If so, the point B to be determined is an abnormal point.

5. The method for rapid measurement of tunnel excavation face deformation according to claim 1, characterized in that: The extraction method of the vault is: Determine the vertical coordinate value of each contour pixel point, and take the pixel point with the largest vertical coordinate value as the arch top; The extraction method of the arch foot is: For step excavation, the pixel point where the curvature of the contour pixel suddenly changes to 0 is the arch foot; For full-section excavation, the pixel point where the curvature of the contour pixel suddenly changes to below the curvature threshold is the arch foot.

6. The method for rapid measurement of tunnel excavation face deformation according to claim 1, characterized in that: The specific calculation process of the three-dimensional spatial deformation of the reference point includes: Based on two images taken simultaneously by binocular photography, the SGBM algorithm is used to calculate the disparity between the two images. Based on the geometric relationship of parallel binocular vision, the three-dimensional spatial information of the image is calculated. Based on the comparison of multiple images, the three-dimensional spatial deformation of the reference point is calculated.

7. A tunnel excavation face deformation rapid detection device, characterized in that: It comprises a binocular camera and an industrial computer, wherein the industrial computer stores one or more executable programs, and the one or more programs include instructions for executing the tunnel excavation face deformation rapid measurement method according to any one of claims 1-6.

Images