A method and device for identifying tunnel face collapse based on digital images
Through digital image processing and binocular photography technology, the joint structure of the palm face is identified and analyzed, and the problem of the failure to accurately predict the collapse position and range in the existing technology is solved, and the refined analysis and prediction of the collapse of the palm face is realized, improving the improvement of the collapse of the palm face. The accuracy of construction decisions.
Patent Information
- Application Number
- CN202210162176.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-22
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2042-02-22
AI Technical Summary
The existing technology cannot accurately obtain the deformation characteristics and maximum deformation points of the full section of the palm surface, resulting in the collapse prediction results that are inconsistent with the actual situation, and the location and scope of the collapse cannot be predicted in detail, affecting emergency treatment decisions.
Using a digital image-based method, the palm surface image is obtained through binocular photography, preprocessing and image segmentation, three-dimensional spatial deformation is calculated, joint structure is identified, and the type, location and range of landslides are determined based on the integrity of surrounding rock and deformation data.
Accurate identification of landslide collapse on the palm surface, including prediction of type, direction and scope, and improve the reliability of landslide warning and the guiding significance of construction decisions.
Smart Images

Figure CN114565758B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of surrounding rock stability evaluation after tunnel excavation, and relates to a tunnel face surrounding rock stability evaluation method, and in particular relates to a tunnel face collapse identification method and device based on digital images. Background Art
[0002] At present, there are many methods for judging tunnel instability, such as plastic zone penetration judgment, plastic zone size judgment, plastic strain energy judgment, etc. However, in actual engineering, considering feasibility, on-site construction still mainly uses deformation data measured by construction monitoring as judgment indicators. At present, the deformation state of key parts such as the arch crown and arch foot of the face is mainly obtained through internal displacement meters, inclinometers, total stations and other instruments or visual inspection, which are regarded as the maximum deformation of the face, and whether it is unstable or collapsed is judged based on this. For example, patent application CN113074695A discloses a tunnel stability interpretation method based on three-dimensional deformation monitoring of the face, which monitors the arch crown settlement and horizontal convergence within an excavation cycle distance on the face in real time, and uses an inclinometer to monitor the extrusion displacement in front of the face, so as to realize three-dimensional monitoring of tunnel face deformation, and judge the stability of the face through three deformation indicators: displacement change rate of measuring point, displacement change of measuring point and extrusion deformation of the face.
[0003] The outstanding problems faced by the existing technology are:
[0004] (1) It is impossible to obtain the deformation characteristics of the entire cross-section of the tunnel face, the maximum deformation point, and the maximum deformation amount, which easily leads to the collapse prediction results not being consistent with the actual situation;
[0005] (2) It can only qualitatively predict whether a landslide will occur, but cannot accurately predict the location and scope of the landslide, which is not conducive to the formulation of emergency treatment decisions after the early warning. Summary of the invention
[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a fast and accurate method and device for identifying tunnel face collapse based on digital images.
[0007] The purpose of the present invention can be achieved by the following technical solutions:
[0008] A method for identifying tunnel face collapse based on digital images comprises the following steps:
[0009] Obtain the complete tunnel face image through binocular photography and perform preprocessing;
[0010] Perform image segmentation on the preprocessed tunnel face image, extract the tunnel face area, and calculate the three-dimensional spatial deformation of the tunnel face;
[0011] Identify and extract face joints from the face area, count the number of joint groups, obtain a joint skeleton using a skeleton processing algorithm, count the joint spacing, and determine the degree of integrity of the surrounding rock based on the number of joint groups and the joint spacing. The degree of integrity of the surrounding rock includes two categories: relatively complete to complete surrounding rock and relatively broken to broken surrounding rock;
[0012] The corresponding tunnel face collapse identification results are obtained according to the determined surrounding rock integrity, specifically:
[0013] For relatively complete to complete surrounding rock, digital image processing technology is used to obtain the key points of the cross-section arch top and arch bottom and connect them to represent the cross-section height. The extracted tunnel face area is divided into three equal parts according to the cross-section height, namely upper, middle and lower areas. The maximum bulging deformation of the tunnel face is used as the collapse judgment index, and the collapse type, location and scope are determined by combining the deformation data of different areas.
[0014] For relatively broken to broken surrounding rocks, the face area is first preliminarily zoned according to the joint skeleton, and then it is determined whether to merge the zones based on the continuity of the deformation of the surrounding rocks on both sides of the joints. The final face zones are determined, and the location and scope of the collapse are determined by comprehensively considering the deformation acceleration of each zone and the relative difference in deformation on both sides of the joints.
[0015] Furthermore, the calculation of the three-dimensional spatial deformation of the tunnel face is specifically as follows:
[0016] Based on two images obtained simultaneously by binocular photography, the SGBM algorithm is used to calculate the parallax of the two images. According to 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 tunnel face is obtained.
[0017] Furthermore, the camera for collecting the tunnel face image is arranged at a distance of 30m to 50m from the tunnel face, and fill light is provided during the camera shooting process.
[0018] Furthermore, the preprocessing includes image noise removal.
[0019] Furthermore, the combination of deformation data from different regions to determine the type, location and scope of the landslide is specifically as follows:
[0020] Obtain the surrounding rock grade based on the face joint parameters;
[0021] Based on the three-dimensional deformation of the tunnel face, the maximum deformation value of each row is extracted row by row according to the pixel points, the corresponding pixel point position is recorded, and the position is projected to the longitudinal section, the bulging deformation projection curve is drawn, and the position of the maximum bulging deformation point of the entire section is recorded;
[0022] Determine whether the deformation value of the maximum bulging deformation point of the entire section exceeds the maximum allowable deformation of the corresponding surrounding rock level. If so, determine whether the tunnel face has collapsed, the type of collapse and the corresponding range, and use the displacement vector direction at the maximum bulging deformation point as the predicted direction of the collapse.
[0023] Furthermore, the types of collapse include top overturning collapse, local collapse of the tunnel face and bottom shear sliding collapse.
[0024] Furthermore, the maximum allowable deformation after tunnel excavation is calculated according to the plane strain calculation mode, specifically:
[0025] For shallow tunnels, u=2.0·u r=r0 (θ=90°),
[0026] For deep tunnels, u=1.5·u r=r0 (θ=90°),
[0027] Where u is the maximum allowable deformation, r is the radius of the circular tunnel excavation section, and r 0 is 1 / 2 of the excavation section height, H is the tunnel burial depth, E 2 is the elastic modulus of surrounding rock, P 0 is the surrounding rock pressure borne by the tunnel structure, μ is the Poisson's ratio, and θ is the angle between the line connecting the perimeter points and the center of the circle and the positive direction of the x-axis.
[0028] Furthermore, the determination of whether to merge the partitions based on the continuity of deformation of the surrounding rocks on both sides of the joint is specifically as follows:
[0029] Comparing the deformations on both sides of the same joint, when the difference in deformation on both sides is less than or equal to 2 cm, the deformation is considered continuous and the adjacent partitions are merged into the same partition.
[0030] Furthermore, the comprehensive deformation acceleration of each subarea and the relative difference of deformation on both sides of the joints are used to determine the type, location and scope of the collapse:
[0031] Based on the deformation development trend of the partition, calculate the deformation acceleration a;
[0032] When a>0 appears in the deformation process, it is determined that the corresponding area will collapse. When the deformation process lasts for a<=0, the deformation difference of the areas on both sides of the joint is determined: when there is a weak interlayer, the deformation difference is greater than 2cm, and it is determined that the area will collapse; when there is no weak interlayer, the deformation difference is greater than 1cm, and it is determined that the area will collapse;
[0033] When landslide occurs, the displacement vector direction at the maximum deformation point in each area is used as the predicted direction of each landslide, and the chain landslide area is determined by the one-time retrieval principle to obtain the final landslide range.
[0034] The present invention also provides an electronic identification device, comprising:
[0035] one or more processors;
[0036] Memory; and
[0037] One or more programs stored in a memory, the one or more programs comprising instructions for executing the method for identifying tunnel face collapse based on digital images as described above.
[0038] Compared with the prior art, the present invention performs a refined analysis based on the instability mechanism of the tunnel face, thereby determining the type, direction and scope of the collapse, and has the following beneficial effects:
[0039] (1) According to the analysis of the instability mechanism, the tunnel face is divided into two categories: relatively intact to intact surrounding rock and relatively broken to broken surrounding rock. According to the differences in the deformation modes and collapse types of the two types of tunnel faces, targeted analysis is carried out, appropriate indicators are selected, and evaluation criteria for the corresponding modes are formulated. Compared with the traditional analysis method, the technical route of the present invention is obviously more in line with reality and has high recognition reliability.
[0040] (2) For relatively complete to complete surrounding rock faces, three common deformation modes and corresponding landslide types are proposed, and the landslide range is calculated from the perspective of the plastic zone of the surrounding rock; the bulging deformation of the face is used as the judgment index, and corresponding standards are formulated respectively. For relatively broken to broken faces, starting from the deformation trend, the judgment index combining deformation acceleration and deformation difference of adjacent areas is used to quickly predict landslides and their scale. Compared with traditional methods, the method of the present invention can not only accurately determine landslides, but also predict the direction and scale of landslides, and can timely guide the construction party to adopt response strategies, which has obvious guiding significance for engineering construction.
[0041] (3) Based on the plane strain solution of elastic mechanics, the present invention obtains the deformation law of tunnel surrounding rock, and combines the actual construction monitoring and measurement data to obtain the calculation formula for the tunnel collapse displacement limit value, providing a theoretical basis for tunnel collapse assessment. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 It is a schematic diagram of the process of the present invention;
[0043] Figure 2 Connect the maximum bulging deformed pixels in each row;
[0044] Figure 3 is the projection curve of bulging deformation;
[0045] Figure 4 The division of the tunnel face area for relatively complete to complete surrounding rock tunnel faces;
[0046] Figure 5 This is a schematic diagram of a top overturning landslide;
[0047] Figure 6 This is a schematic diagram of local collapse of the tunnel face;
[0048] Figure 7 This is a schematic diagram of bottom shear sliding collapse;
[0049] Figure 8 Plane strain calculation model for tunnel excavation;
[0050] Fig. 9 To calculate the external load of the surrounding rock circle;
[0051] Fig.10 A schematic diagram for determining the scope of top overturning landslide;
[0052] Fig.11 Determine the schematic diagram of the local collapse range of the tunnel face;
[0053] Fig.12 Provide a schematic diagram for determining the scope of bottom shear-slip collapse;
[0054] Fig.13 Preliminary zoning of the tunnel face for relatively broken to broken tunnel faces;
[0055] Fig.14 The final partition of the tunnel face for the relatively broken to broken tunnel face;
[0056] Fig.15 It is a schematic diagram of deformation stability curve of the tunnel face area;
[0057] Fig.16 It is a schematic diagram of the deformation instability curve of the tunnel face area;
[0058] Fig.17 A search diagram. DETAILED DESCRIPTION
[0059] The present invention is described in detail below in conjunction with 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.
[0060] Explanation of terms
[0061] 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 in tunnel excavation (coal mining, mining or tunnel engineering).
[0062] Structural plane: refers to a discontinuous plane with extremely low or no tensile strength. That is, various structural planes formed under the action of tectonic forces after the formation of the rock mass, including faults, joints, cleavages and interlayer dislocations.
[0063] Joints: Joints, also known as fissures, are small fracture structures in which the rock blocks on both sides do not move significantly after the rock mass breaks under stress. Joints are a very common structural geological phenomenon, which are the cracks we see on rock outcrops, or rock cracks.
[0064] Landslide: refers to the collapse of earth and rock. Steep slopes beside roads and dams suddenly collapse due to weathering, flooding, vibration, etc., or the tops of tunnels, mines, etc. collapse due to soft soil and rock layers.
[0065] Example 1
[0066] like Figure 1 As shown, this embodiment provides a method for identifying tunnel face collapse based on digital images, comprising the following steps:
[0067] S1. Camera setup.
[0068] This embodiment adopts binocular photography technology, and a camera is deployed on both sides of the tunnel for synchronous collection; the camera is optimally located within a range of 30m to 50m from the tunnel face to ensure that a complete tunnel face image can be collected; the fill light is turned on during the shooting process, and a fill light device is provided when necessary to ensure that the entire tunnel face is clearly visible and free of shadows.
[0069] S2. Obtain a complete image of the palm face through binocular photography.
[0070] S3. Image preprocessing and deformation calculation.
[0071] The collected tunnel face images are first preprocessed to eliminate image noise caused by the dark and dusty environment of the tunnel. The preprocessed tunnel face images are segmented to extract the tunnel face area and calculate the three-dimensional spatial deformation of the tunnel face. Specifically:
[0072] Based on the images taken simultaneously by the left and right cameras, the SGBM algorithm is used to calculate the parallax of the two images. According to 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 palm face is calculated.
[0073] S4. Determination of the integrity of the surrounding rock at the face.
[0074] A combination of expansion and corrosion methods is used to identify and extract face joints from the face area, count the number of joint groups, identify dominant joints, and use a skeleton processing algorithm to obtain a joint skeleton. The pixel spacing between adjacent skeleton feature points is counted to obtain the joint spacing. Based on the joint group number and joint spacing, the degree of integrity of the surrounding rock is determined. The degree of integrity of the surrounding rock includes two categories: relatively complete to complete surrounding rock and relatively broken to broken surrounding rock. The specific determination criteria are shown in Table 1.
[0075] Table 1 Criteria for judging the integrity of surrounding rock at the tunnel face
[0076]
[0077] S5. Division of the face area.
[0078] For relatively complete to complete surrounding rock faces, digital image processing technology is used to obtain the key points of the cross-section arch top and arch bottom and connect them to represent the cross-section height. The face is divided into three equal parts according to the cross-section height, namely upper, middle and lower areas.
[0079] For the tunnel face with relatively broken to broken surrounding rock, the tunnel face is first preliminarily divided into zones according to the joint skeleton, and then it is determined whether to merge the zones based on the continuity of the deformation of the surrounding rock on both sides of the joints to determine the final tunnel face zones.
[0080] S6. Identification of face collapse.
[0081] Based on the deformation data of the tunnel face surrounding rock obtained in step S3 and the tunnel face partition result determined in step S5, identify whether the tunnel face has collapsed, and determine the location and range of the collapse:
[0082] For relatively intact to complete surrounding rock, the maximum bulging deformation of the tunnel face is used as the indicator for determining collapse, and the type, location and scope of collapse are determined in combination with the deformation data of different areas;
[0083] For relatively broken to broken surrounding rocks, the deformation acceleration of each zone and the relative difference of deformation on both sides of the joint surface are combined to determine the type, location and scope of the collapse;
[0084] Based on the above steps S5 and S6, the specific landslide identification process for the two types of surrounding rocks (2) relatively intact to intact surrounding rocks and relatively broken to broken surrounding rocks is described as follows.
[0085] (1) Identification of collapse at the tunnel face with relatively intact to complete surrounding rock
[0086] For relatively intact to complete surrounding rock, the tunnel face after excavation often shows bulging deformation. The steps for identifying collapse of this type of tunnel face include:
[0087] 1) Based on the above-mentioned extraction of face joint attitude parameters, the surrounding rock grade is calculated using the calculation formula of the Highway Tunnel Design Code.
[0088] 2) Based on the full-section deformation perception results, extract the maximum deformation value of each row by pixel point, and record the corresponding pixel point position, such as Figure 2 , and project it to the longitudinal section to draw the bulging deformation projection curve, such as Figure 3 , and record the position of the maximum bulging deformation point of the entire section.
[0089] 3) Divide the tunnel face into three equal parts along the height direction, namely upper, middle and lower areas, specifically:
[0090] Based on the segmented tunnel face image, digital image processing is used to extract and mark the top and bottom points of the arch. Connect the top and bottom of the arch and divide the line segment into three equal parts. Draw a horizontal line through the equal division points to divide the tunnel face into three areas: upper, middle, and lower. Figure 4 shown.
[0091] 4) Based on the maximum bulging deformation value of the entire section and the area where it is located on the projection curve, determine whether the tunnel face has collapsed, as well as the type, location and scope of the collapse. The specific evaluation method is as follows:
[0092] Determination of landslide type:
[0093] The maximum deformation is used to determine whether the tunnel face has collapsed. Specifically, the maximum allowable deformation after tunnel excavation is calculated according to the plane strain calculation mode. If the deformation exceeds this value, the tunnel will become unstable. Figure 8 and Fig. 9 shown.
[0094] The calculation formula and standard are:
[0095] For shallow tunnels, u=2.0·u r=r0 (θ=90°)(H<2.5r 0 )
[0096] For deep tunnels, u=1.5·u r=r0 (θ=90°)(H>=2.5r 0 )
[0097] in,
[0098]
[0099]
[0100] Where: r is the radius of the circular tunnel excavation section. For common highway tunnels, r 0 1 / 2 of the height of the excavation section;
[0101] H——tunnel burial depth;
[0102] E 2 ——elastic modulus of surrounding rock;
[0103] P 0 - surrounding rock pressure on the tunnel structure;
[0104] μ——Poisson’s ratio;
[0105] θ——The angle between the line connecting the hole perimeter point and the circle center and the positive direction of the x-axis.
[0106] For the surrounding rock pressure P 0 , the calculation formula is as follows:
[0107]
[0108] When the engineering object is a deep-buried two-lane highway tunnel, the standards in Table 2 can be used to directly determine whether there is a landslide and the type of landslide. The types of landslides under different deformation modes are as follows: Figure 5-Figure 7 shown.
[0109] Table 2 Landslide judgment criteria based on maximum bulging deformation and deformation position (unit: cm)
[0110]
[0111] Note: 1) “—” in the table means that this deformation mode generally does not occur under the current surrounding rock level.
[0112] 2) When the bulging deformation is greater than or equal to the above-mentioned warning standard, it is considered that landslide will occur at the tunnel face.
[0113] Determination of landslide direction: The direction of the displacement vector at the maximum deformation point is used as the predicted direction of landslide.
[0114] Determination of landslide scale: The scope and method of determination are as follows.
[0115] For top overturning landslide, the landslide range is as follows Fig.10 The OABCDO area in the figure. Point O is the top of the arch, point C is the bottom of the arch, and point D is the point of maximum bulging deformation. is the internal friction angle of the surrounding rock; the straight line CB is the bottom shear slip line, and the horizontal angle The specific determination method is: draw a perpendicular line from point O to straight line CB, which intersects at point B; draw an arc with point O as the center and OB as the radius, which intersects with the top of the tunnel at point A, and the landslide range can be obtained.
[0116] For local collapse of the tunnel face, the collapse range is as follows: Fig.11 The ABC area in the figure. Point A is the arch top, point B is the arch bottom, and point C is the maximum bulging deformation point.
[0117] For bottom shear sliding collapse, the collapse range is as follows Fig.12 The OABCDO area in the figure. Point O is the top of the arch, point C is the bottom of the arch, and point D is the point of maximum bulging deformation. is the internal friction angle of the surrounding rock, the straight line CB is the bottom shear slip line, and the horizontal angle Line OA is the top shear slip line, and the horizontal angle The specific determination method is: draw a perpendicular line from point O to straight line CB, which intersects at point B; draw an arc with point O as the center and OB as the radius, which intersects with the upper slip line OA at point A, and the landslide range can be obtained.
[0118] (2) Identification method of relatively broken to broken surrounding rock face collapse
[0119] For the tunnel face with relatively broken to broken surrounding rock, the deformation of the tunnel face after excavation is mainly controlled by the structural surface, which is mainly manifested as the sliding of blocks along the structural surface. For the identification of such tunnel face collapse, the main processing steps are:
[0120] 1) Based on the identified tunnel face joints, the tunnel face is initially divided into independent areas, such as Fig.13 .
[0121] 2) Search and compare the deformation on both sides of the same joint. When the deformation difference on both sides is 2mm, the deformation is considered discontinuous and the deformation of the two adjacent blocks is determined to be independent of each other. When the deformation difference on both sides of the joint is within 2mm, the deformation is considered continuous and the adjacent areas are merged into the same area. Fig.14 .
[0122] 3) Based on the division of the tunnel face area, the deformation at the maximum deformation point in each area is continuously monitored and the deformation development curve is drawn, such as Fig.15 and Fig.16 shown.
[0123] 4) Determine the tunnel face collapse and collapse range based on the above deformation development curve. The specific evaluation method is as follows:
[0124] Landslide judgment: Based on the regional deformation development trend, calculate the deformation acceleration a; use the deformation acceleration a as the indicator for determining landslide.
[0125] When a>0 appears in the deformation process, it is determined that the corresponding area will collapse;
[0126] When the deformation process lasts for a<=0, the deformation difference of the areas on both sides of the joint is determined. When there is a weak interlayer, the deformation difference is greater than 2cm, and it is determined that the area will collapse; when there is no weak interlayer, the deformation difference is greater than 1cm, and it is determined that the area will collapse.
[0127] Determination of landslide direction: The direction of the displacement vector at the maximum deformation point in each area is used as the predicted direction of each landslide.
[0128] Determination of landslide scale: First, landslides are determined in each area based on the above determination method to determine the landslide area, and then the chain landslide area is determined using the one-time retrieval principle.
[0129] In one search, only the upper blocks directly adjacent to the landslide area will be considered, which may be affected and cause chain collapse.
[0130] Direct landslide and chain landslide together constitute the landslide judgment range.
[0131] like Fig.17 As shown, area A is the landslide judgment area, areas B and C are directly adjacent and upper areas with a common structural surface, and area D is not; therefore, areas B and C are judged to be chain landslide areas.
[0132] The above method uses binocular photogrammetry to grasp the three-dimensional spatial deformation of the entire cross-section of the tunnel face. According to the quality of the surrounding rock, the tunnel face is divided into two categories: relatively complete to complete surrounding rock and relatively broken to broken surrounding rock. This method can obtain the deformation of the entire cross-section of the tunnel face, and thereby finely evaluate the stability of the surrounding rock, and predict whether the tunnel face will collapse and the location and range of the collapse.
[0133] 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, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform 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 codes.
[0134] Example 2
[0135] This embodiment provides an electronic identification device, including one or more processors, a memory and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the method for identifying tunnel face collapse based on digital images as described in Example 1.
[0136] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.
Claims
1. A method for identifying tunnel face collapse based on digital images, characterized in that: The following steps are involved: Obtain the complete tunnel face image through binocular photography and perform preprocessing; Perform image segmentation on the preprocessed tunnel face image, extract the tunnel face area, and calculate the three-dimensional spatial deformation of the tunnel face; Identify and extract face joints from the face area, count the number of joint groups, obtain a joint skeleton using a skeleton processing algorithm, count the joint spacing, and determine the degree of integrity of the surrounding rock based on the number of joint groups and the joint spacing. The degree of integrity of the surrounding rock includes two categories: relatively complete to complete surrounding rock and relatively broken to broken surrounding rock; The corresponding tunnel face collapse identification results are obtained according to the determined surrounding rock integrity, specifically: For relatively complete to complete surrounding rock, digital image processing technology is used to obtain the key points of the cross-section arch top and arch bottom and connect them to represent the cross-section height. The extracted tunnel face area is divided into three equal parts according to the cross-section height, namely upper, middle and lower areas. The maximum bulging deformation of the tunnel face is used as the collapse judgment index, and the collapse type, location and scope are determined by combining the deformation data of different areas. For relatively broken to broken surrounding rocks, the face area is first preliminarily zoned according to the joint skeleton, and then it is determined whether to merge the zones based on the continuity of the deformation of the surrounding rocks on both sides of the joints. The final face zones are determined, and the location and scope of the collapse are determined by comprehensively considering the deformation acceleration of each zone and the relative difference in deformation on both sides of the joints.
2. The method for identifying tunnel face collapse based on digital images according to claim 1 is characterized in that: The calculation of the three-dimensional deformation of the tunnel face is specifically as follows: Based on two images obtained simultaneously by binocular photography, the SGBM algorithm is used to calculate the parallax of the two images. According to 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 tunnel face is obtained.
3. The method for identifying tunnel face collapse based on digital images according to claim 1, characterized in that: The camera for collecting the tunnel face image is arranged at a distance of 30m to 50m from the tunnel face, and fill light is provided during the camera shooting process.
4. The method for identifying tunnel face collapse based on digital images according to claim 1, characterized in that: The pre-processing includes image noise removal.
5. The method for identifying tunnel face collapse based on digital images according to claim 1, characterized in that: The method of combining deformation data from different regions to determine the type, location and scope of the landslide is specifically as follows: Obtain the surrounding rock grade based on the face joint parameters; Based on the three-dimensional deformation of the tunnel face, the maximum deformation value of each row is extracted row by row according to the pixel points, the corresponding pixel point position is recorded, and the position is projected to the longitudinal section, the bulging deformation projection curve is drawn, and the position of the maximum bulging deformation point of the entire section is recorded; Determine whether the deformation value of the maximum bulging deformation point of the entire section exceeds the maximum allowable deformation of the corresponding surrounding rock level. If so, determine whether the tunnel face has collapsed, the type of collapse and the corresponding range, and use the displacement vector direction at the maximum bulging deformation point as the predicted direction of the collapse.
6. The method for identifying tunnel face collapse based on digital images according to claim 1 or 5, characterized in that: The types of collapse include top overturning collapse, local collapse of the tunnel face and bottom shear sliding collapse.
7. The method for identifying tunnel face collapse based on digital images according to claim 5 is characterized in that: The maximum allowable deformation after tunnel excavation is calculated according to the plane strain calculation mode, specifically: For shallow tunnels, u=2.0·u r=r0 (θ=90°), For deep tunnels, u=1.5·u r=r0 (θ=90°), Among them, u is the maximum allowable deformation, r is the radius of the circular tunnel excavation section, r0 is 1 / 2 of the excavation section height, H is the tunnel burial depth, E2 is the elastic modulus of the surrounding rock, P0 is the surrounding rock pressure borne by the tunnel structure, μ is the Poisson's ratio, and θ is the angle between the line connecting the tunnel perimeter points and the center of the circle and the positive direction of the x-axis.
8. The method for identifying tunnel face collapse based on digital images according to claim 1, characterized in that: The determination of whether to merge the zones based on the continuity of deformation of the surrounding rocks on both sides of the joint is specifically as follows: Comparing the deformations on both sides of the same joint, when the difference in deformation on both sides is less than or equal to 2 cm, the deformation is considered continuous and the adjacent partitions are merged into the same partition.
9. The method for identifying tunnel face collapse based on digital images according to claim 1, characterized in that: The type, location and scope of the collapse are determined by combining the deformation acceleration of each subarea and the relative difference of deformation on both sides of the joints: Based on the deformation development trend of the partition, calculate the deformation acceleration a; When a>0 appears in the deformation process, it is determined that the corresponding area will collapse. When the deformation process lasts for a<=0, the deformation difference of the areas on both sides of the joint is determined: when there is a weak interlayer, the deformation difference is greater than 2cm, and it is determined that the area will collapse; when there is no weak interlayer, the deformation difference is greater than 1cm, and it is determined that the area will collapse; When landslide occurs, the displacement vector direction at the maximum deformation point in each area is used as the predicted direction of each landslide, and the chain landslide area is determined by the one-time retrieval principle to obtain the final landslide range.
10. An electronic identification device, characterized in that: include: one or more processors; Memory; and One or more programs stored in a memory, wherein the one or more programs include instructions for executing the method for identifying tunnel face collapse based on digital images as claimed in any one of claims 1 to 9.
Citation Information
Patent Citations
Tunnel stability interpretation method based on tunnel face three-dimensional deformation monitoring
CN113074695A
Method of shallow-buried-tunnel surrounding-rock collapse mode recognition and support structure calculation
CN107526873A
Intelligent geological sketching and geological evaluation method for tunnel face
CN113256082A
