Tunnel face crack identification method and image extraction equipment

Abstract

The invention discloses a tunnel face crack identification method and an image extraction device, and relates to the technical field of tunnel face crack detection. A safe flat zone is selected at any position in front of a tunnel face to set a central shooting point and peripheral shooting points, and the central shooting point is arranged on a tunnel axis; a binocular camera is used for shooting pictures of different heights at all shooting points, an infrared distance measuring sensor and an electronic compass are carried on the binocular camera, and the shooting direction can be detected through the electronic compass so that the shooting direction can be kept to be in the hole axis direction every time; the infrared distance measuring sensor can detect the actual distance between the binocular camera and the tunnel face during shooting each time, pictures shot each time can be analyzed and calculated according to the actual distance and are converted into the real width and the real height corresponding to each picture, and the real width and the real height are used as reference splicing to achieve collection of the panoramic image of the tunnel face. The recognition range of the panoramic image of the tunnel face is wide, detail features are clear, and the accuracy of crack recognition is guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of tunnel face crack detection, and in particular to a tunnel face crack recognition method and image extraction equipment. Background Art

[0002] During the construction of highway and railway tunnels, hydraulic tunnels, underground mines, etc., face image extraction and crack identification are important means to evaluate the construction status and quality. The measurement and statistics of face cracks constitute the basis for rock mass quality evaluation and surrounding rock classification. However, crack identification is based on the extracted face image, and its accuracy is directly affected by the detailed features of the extracted face image and the identification range.

[0003] At present, in many engineering projects, when collecting tunnel face images, dedicated photographers usually directly take panoramic images of the tunnel face, and then further process the images. The detailed features of the tunnel face images taken in this way are not clear, and the images taken are not accurate due to the shooting angle. It is difficult to accurately obtain the true size of the image, and then it is difficult to obtain the actual size of the tunnel face cracks. The current solution is usually to estimate the tunnel face size by manually calibrating the more special signs or objects in the image, and then calculate the crack length. This method not only increases the workload, but may also produce large errors in the calibration process, resulting in "inaccurate measurement" and "inaccurate calculation" of the crack parameters. In addition, due to the severe environmental conditions of underground engineering, there is a lack of flat areas as shooting sites, and the tunnel face shooting lacks flexibility, which makes the image shooting work during construction subject to many influences.

[0004] There are also relatively intelligent image extraction devices in the prior art. The basic working principle is to set a distance in front of the face, so that the shooting device moves horizontally (in the horizontal direction and perpendicular to the axis of the tunnel) and vertically, and shoot different local positions of the face respectively, and then splice the pictures taken to form a panoramic image of the face with clear details. This method can overcome the defects of "uncertain measurement" and "uncertain calculation" in crack identification. However, according to the above-mentioned limitations of the severe environmental conditions of underground engineering, it is difficult to have a large flat area in front of the face for horizontal movement of the shooting device. The prior art usually sets up beams, slide rails and other forms to achieve horizontal sliding shooting, which makes the overall structure of the image extraction device relatively complex, large in size, and difficult to carry. Each time shooting, it is necessary to select a suitable position to install the beams and slide rails, which is troublesome to install and operate, and lacks flexibility; when installing the beams and slide rails, it is difficult to ensure the lateral and horizontal position accuracy during installation. This installation error will also cause inaccurate face image extraction to a certain extent. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for identifying cracks in a face of a tunnel and an image extraction device. The method can form a panoramic image with a wide recognition range and clear detailed features during image extraction. The device has a simple and portable structure and can achieve flexible shooting.

[0006] The objective of the present invention is achieved through the following technical solutions: A method for identifying cracks in a tunnel face comprises the following steps: S1. Setting a central shooting point and a plurality of peripheral shooting points in front of the face. The central shooting point and the peripheral shooting points are set in a flat and safe area according to actual conditions. The central shooting point is set on the axis of the hole in front of the face. S2, taking pictures at different heights at the central shooting point with a binocular camera, keeping the binocular camera horizontal and facing the palm face during shooting, recording the shooting direction of the binocular camera during shooting, importing each taken picture into a computer and recording the actual distance between the binocular camera and the palm face, the focal length and magnification of the binocular camera during each shooting; S3, taking pictures at different heights at each of the peripheral shooting points using a binocular camera, keeping the binocular camera horizontal during shooting and keeping the shooting direction consistent with the direction when shooting at the central shooting point, importing each of the taken pictures into a computer and recording the actual distance between the binocular camera and the palm face, the focal length and magnification of the binocular camera during each shooting; S4, the computer analyzes and calculates the imported data, and calculates the real width and real height corresponding to the picture according to the actual distance between the binocular camera and the palm face corresponding to each picture, the focal length and magnification of the binocular camera, and the physical parameters of the binocular camera; S5. The computer uses an image stitching toolbox to scale the image according to the real width and real height corresponding to the captured image, and then completes the stitching of the panoramic image of the palm face, obtains the panoramic image of the palm face, and calculates the real width and real height corresponding to the panoramic image of the palm face.

[0007] Furthermore, after step S5, the following steps are further performed: S6, the computer identifies and extracts the crack trace information in the panoramic image of the face through a U-net network model, and realizes vectorization of the crack trace image; S7. The computer imports the vectorized image into the fracture occurrence extraction system to obtain fracture length, fracture dip and fracture tendency information.

[0008] Specifically, the calculation formula for obtaining the real width and real height corresponding to the picture in step S4 is as follows: When N≤N0: , , When N>N0: , , Among them, D is the real width corresponding to the picture; H is the real height corresponding to the picture; f is the focal length of the binocular camera when taking the picture; f0 is the critical focal length of the optical zoom and digital zoom of the binocular camera, which is a physical parameter of the binocular camera; N is the magnification of the binocular camera when taking the picture; N0 is the critical magnification of the optical zoom and digital zoom of the binocular camera, which is a physical parameter of the binocular camera; L is the actual distance between the binocular camera and the palm face when shooting; B is the center line of the two lens sensors in the binocular camera, which is a physical parameter of the binocular camera; h is the height of the lens sensor in the binocular camera, which is a physical parameter of the binocular camera; d is the width of the lens sensor in the binocular camera, which is a physical parameter of the binocular camera.

[0009] Specifically, obtaining the crack length information in step S7 refers to importing the vectorized image into the crack occurrence extraction system, extracting the coordinates of each pixel point in the image, using the 8-neighborhood tracking method to find the crack endpoints and track the crack path, thereby extracting the skeleton of the entire crack; then calculating the distance between consecutive points on the crack path according to the Euclidean distance formula, and accumulating the pixel length of the entire crack; finally, converting the pixel length of the crack into the actual length according to the real width and real height corresponding to the panoramic image of the face.

[0010] Specifically, obtaining the crack inclination information in step S7 refers to importing the vectorized image into the crack occurrence extraction system, extracting the coordinates of each pixel point in the image, performing linear fitting on all pixel coordinate points in a single crack skeleton according to a linear regression algorithm combined with a loop function, obtaining the slope of the straight line, converting the slope into an angle, and thus obtaining the crack apparent inclination; obtaining the crack tendency in step S7 refers to calculating the crack tendency according to the shooting direction of the binocular camera and the crack apparent inclination.

[0011] A tunnel face image extraction device, used for implementing the above-mentioned tunnel face crack identification method, comprises a computer and a binocular camera, the binocular camera is equipped with an infrared ranging sensor and an electronic compass, the infrared ranging sensor is installed at the front end of the binocular camera, the infrared ranging sensor is set at the midpoint of the center line between two lens sensors of the binocular camera, the infrared ranging sensor is used to detect the distance between the obstacle photographed in front of the binocular camera and the binocular camera, the electronic compass is used to detect the shooting direction of the binocular camera, and the binocular camera, the infrared sensor and the electronic compass are all electrically connected to the computer.

[0012] Furthermore, it also includes a leveling seat, which includes a bottom plate and a top plate, the bottom plate and the top plate are connected by three foot screws, a leveler is installed on the top plate, the binocular camera is arranged on the top plate, and the shooting direction of the binocular camera is parallel to the top plate.

[0013] Furthermore, it also includes a rotating seat, which is connected to the top plate, and the binocular camera is rotatably connected to the rotating seat.

[0014] Furthermore, it also includes a spiral elevator, the shell of the spiral elevator is fixedly connected to the top plate, the screw of the spiral elevator is perpendicular to the top plate, and the swivel seat is fixedly installed on the top of the screw of the spiral elevator.

[0015] Furthermore, it also includes a tripod, wherein the head of the tripod is detachably connected to the base plate.

[0016] The beneficial effects of the present invention are: The method for identifying cracks in a tunnel face of the present invention can realize the acquisition of a panoramic image of the tunnel face. During the acquisition, the shooting and splicing of multiple shooting points ensure that the panoramic image of the tunnel face has a wide recognition range and clear detailed features. During each shooting, a distance measuring instrument is used to record the actual distance between the binocular camera and the tunnel face when the image is taken, so as to convert the distance into the real width and real height corresponding to each image as a scaling reference before splicing. Therefore, the shooting process is not restricted by the severe environmental conditions of the underground project, and any flat and safe position can be selected for flexible shooting. During each shooting, the shooting direction is measured and calibrated by an azimuth instrument such as a compass to ensure that the direction of each shooting is consistent along the axis of the tunnel, so as to ensure the accuracy of the panoramic image of the tunnel face after splicing. At the same time, the measured shooting direction also provides a basis for subsequent identification of crack tendency information.

[0017] The tunnel face image extraction device of the present invention is used to implement the above method. Its overall structure is simple and portable. The shooting position is not restricted by the severe environmental conditions of underground engineering. The shooting points can be selected arbitrarily to reduce safety hazards. The distance parameters are directly obtained during shooting, so that the overall picture of the tunnel face after splicing is more detailed, which is conducive to improving the accuracy of crack identification. In addition, the shooting direction can also be recorded to provide a basis for subsequent identification of crack tendencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of the principle of image analysis and calculation in step S4 of a method for identifying cracks in a tunnel face according to the present invention; Figure 2 A schematic diagram of the image stitching process in step S5 of a method for identifying cracks in a tunnel face according to the present invention; Figure 3 This is an example of the vectorization operation of the crack trace image in step S6 of the method for identifying cracks in a tunnel face of the present invention; Figure 4 It is a schematic diagram of the overall structure of a tunnel face image extraction device of the present invention; Figure 5 for Figure 4 Schematic diagram of the structure after removing the tripod. DETAILED DESCRIPTION

[0019] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0020] A method for identifying cracks in a tunnel face comprises the following steps: S1. Set a central shooting point and multiple peripheral shooting points in front of the palm face. When setting, the distance between each shooting point and the palm face is not restricted, and any flat and safe area in front of the palm face can be selected for setting; the central shooting point is set at the hole axis position in front of the palm face, and the peripheral shooting points are set arbitrarily according to actual needs. The number of peripheral shooting points and the position of each peripheral shooting point in the horizontal direction must ensure that the palm face images taken by all shooting points can be spliced ​​into a complete palm face panoramic image with a wide recognition range.

[0021] S2. Use a binocular camera to shoot the palm face at the central shooting point. When shooting, the binocular camera is kept horizontal and facing the palm face, and the shooting direction of the binocular camera is recorded by a compass or other azimuth instruments. After shooting, the captured image is imported into a computer, and the actual distance between the binocular camera and the palm face when the image is taken is recorded by a distance measuring instrument. At the same time, the focal length, magnification and physical parameter information of the binocular camera when the image is taken are recorded (the parameters of the binocular camera itself, including the critical focal length of the binocular camera's optical zoom and digital zoom, the critical magnification of the binocular camera's optical zoom and digital zoom, the center line of the two lens sensors in the binocular camera, the height of the lens sensor in the binocular camera, the width of the lens sensor in the binocular camera, etc.). When shooting at the central shooting point, the binocular camera takes pictures at different heights in the vertical direction in the above manner so that the taken pictures contain all the information of the position of the tunnel face on the tunnel axis in the height direction. At the same time, each shooting records the actual distance between the binocular camera and the tunnel face corresponding to each picture, the focal length of the binocular camera, the magnification and the physical parameter information in the above manner.

[0022] S3. Use the binocular camera to take pictures at different heights at each peripheral shooting point. Keep the binocular camera horizontal during each shooting, and adjust the shooting direction of the binocular camera using the aforementioned compass and other azimuth instruments to keep the direction consistent with that at the central shooting point during each shooting. Similarly, import each shot into the computer and record the actual distance between the binocular camera and the palm face, the focal length of the binocular camera, the magnification and physical parameter information during each shooting.

[0023] S4. The computer analyzes and calculates the imported data, and calculates the real width and real height of the picture according to the actual distance between the binocular camera and the face, the focal length and magnification of the binocular camera, and the physical parameters of the binocular camera at each shooting. For specific calculations, please refer to Figure 1 , the calculation formula for obtaining the actual width and height corresponding to the picture is as follows: When N≤N0: , , When N>N0: , , in, D is the actual width corresponding to the picture screen; H is the actual height corresponding to the picture; f is the focal length of the binocular camera when taking the picture; f0 is the critical focal length of the binocular camera’s optical zoom and digital zoom, and is a physical parameter of the binocular camera; N is the magnification of the binocular camera when taking the picture; N0 is the critical magnification of the optical zoom and digital zoom of the binocular camera, which is a physical parameter of the binocular camera; L is the actual distance between the binocular camera and the palm face during shooting; B is the center line of the two lens sensors in the binocular camera, which is the physical parameter of the binocular camera; h is the height of the lens sensor in the binocular camera, which is a physical parameter of the binocular camera; d is the width of the lens sensor in the binocular camera, which is a physical parameter of the binocular camera.

[0024] S5. The computer uses the image stitching toolbox to scale the images taken at each shooting point and height according to the actual width and actual height of the image. Figure 2 As shown, based on the overlapping parts of the detailed features in the scaled images, the scaled images are stitched into a complete panoramic image of the tunnel face, and the real width and real height corresponding to the panoramic image of the tunnel face are calculated. For example, when stitching images A and B of similar shooting points, the sizes of images A and B directly taken by the same binocular camera are consistent. Suppose the real width corresponding to the length of image A is 3m, the real height corresponding to the width of image A is 2m, the real width corresponding to the length of image B is 4.5m, and the real height corresponding to the width of image B is 3m. Before stitching, the real length or real width corresponding to image A can be used as a reference, and image B can be enlarged 1.5 times before stitching.

[0025] Thus, through steps S1 to S5, the acquisition of the panoramic image of the tunnel face can be realized. During the acquisition, the detailed images are taken at multiple shooting points and then stitched together as a whole, ensuring that the panoramic image of the tunnel face has a wide recognition range and clear detailed features. During each shooting, the actual distance between the binocular camera and the tunnel face when the image is taken is recorded by a distance measuring instrument, and the distance is converted into the real width and real height corresponding to each image as the scaling reference before stitching. Therefore, the shooting process is not restricted by the severe environmental conditions of the underground project, and any flat and safe position can be selected for flexible shooting. During each shooting, the shooting direction is measured and calibrated by an azimuth instrument such as a compass, ensuring that the direction of each shooting is consistent along the direction of the tunnel axis, and ensuring the accuracy of the panoramic image of the tunnel face after stitching. At the same time, since the binocular camera is kept horizontal and facing the tunnel face during shooting in step S2, the shooting direction determined at this time is the direction of the tunnel axis, and the measured shooting direction also provides a basis for the subsequent identification of crack tendency information.

[0026] S6. The computer uses the U-net network model to identify and extract the crack trace information in the panoramic image of the face, and realizes the vectorization of the crack trace image. The operation process is as follows: Figure 3 shown.

[0027] S7. The computer imports the vectorized image into the fracture occurrence extraction system to obtain fracture length, fracture dip and fracture tendency information. Specifically: Obtaining the crack length information means importing the vectorized image into the crack occurrence extraction system, extracting the coordinates of each pixel in the image, using the 8-neighborhood tracking method to find the crack endpoints and track the crack path, thereby extracting the skeleton of the entire crack; then calculating the distance between consecutive points on the crack path according to the Euclidean distance formula, and accumulating to obtain the length of the entire crack; finally, converting the pixel length of the crack into the actual length according to the true width and true height corresponding to the panoramic image of the face; the calculation formula is as follows: , Where l is the actual length of the crack, l i is the pixel distance between two adjacent points on the crack, D is the actual length of the image, b is the pixel length of the image, (x i ,y i ) is the coordinate of a pixel point in the crack skeleton.

[0028] Obtaining fracture dip information means importing the vectorized image into the fracture occurrence extraction system, extracting the coordinates of each pixel in the image, performing linear fitting on all pixel coordinate points in a single fracture skeleton according to the linear regression algorithm combined with the loop function, obtaining the slope of the straight line, converting the slope into an angle, and thus obtaining the fracture apparent dip. The specific calculation formula is as follows: , Among them, k is the slope of the straight line, c is the y-intercept of the straight line, is the apparent inclination angle of the fracture, , , (x i ,y i ) is the coordinate of a pixel point in the crack skeleton.

[0029] Obtaining the fracture inclination means calculating the fracture inclination based on the shooting direction of the binocular camera (i.e., the direction of the hole axis) and the fracture apparent inclination angle. The calculation formula is as follows: , in, is the hole axis direction, is the tunnel face azimuth, is the crack inclination angle, is the fracture tendency; if the calculated , then the crack tendency ; If the calculated , then the crack tendency .

[0030] like Figure 4 , Figure 5 As shown, a palm face image extraction device includes a computer and a binocular camera 1. The binocular camera 1 is equipped with an infrared ranging sensor 2 and an electronic compass 3. The infrared ranging sensor 2 is installed at the front end of the binocular camera 1. The infrared ranging sensor 2 is set at the midpoint of the center line between the two lens sensors of the binocular camera 1. The infrared ranging sensor 2 is used to detect the distance between the obstacle photographed in front of the binocular camera 1 and the binocular camera 1. The electronic compass 3 is used to detect the shooting direction of the binocular camera 1. The binocular camera 1, the infrared sensor 2, and the electronic compass 3 are all electrically connected to the computer. The tunnel face image extraction device is used to implement the above-mentioned tunnel face crack identification method. The electronic compass 3 can detect and calibrate the shooting direction of the binocular camera 1 at each shooting, and the infrared distance sensor 2 can measure the actual distance between the binocular camera 1 and the tunnel face at each shooting. The computer can be a portable computer, which is used to receive the shooting direction detected by the electronic compass 3 and compare it at each shooting, guide the operator to debug, and ensure that the shooting direction is consistent. The computer is also used to receive the pictures taken at each shooting, the relevant parameters of the binocular camera 1 at each shooting, and the distance value detected by the infrared sensor 2 at each shooting, and complete the above-mentioned S4 to S7 analysis and calculation, picture splicing, crack identification and other processes. In general, the tunnel face image extraction device has a simple and portable overall structure, and the shooting position is not limited by the severe environmental conditions of the underground project. The shooting point can be selected arbitrarily to reduce safety hazards. The distance parameter is directly obtained during shooting, so that the overall picture of the tunnel face after splicing is more detailed, which is conducive to improving the accuracy of crack identification. In addition, the shooting direction can also be recorded to provide a basis for subsequent crack identification.

[0031] Furthermore, it also includes a leveling seat 4. The structure of the leveling seat 4 is similar to the structure of the adjustment platform of the engineering theodolite, level meter, etc. It includes a bottom plate 5 and a top plate 6. The bottom plate 5 and the top plate 6 are connected by three foot screws 7. A leveler 8 is installed on the top plate 6. After the bottom plate 5 is fixed on the bracket, the top plate 6 can be adjusted to a horizontal state by the foot screws 7. The leveler 8 can be a bubble leveler, which is used to observe the leveling state during leveling. The binocular camera 1 is set on the top plate 6. The shooting direction of the binocular camera 1 is parallel to the top plate 6. When in use, the bottom plate 5 can be installed on any bracket for support. The top plate 6 is leveled by the leveling seat 4, so that the binocular camera 1 can accurately shoot in the horizontal direction and maintain the consistency of the shooting direction of each shooting point. In specific implementation, it also includes a tripod 9 commonly used in engineering inspections. The bottom plate 5 is directly and detachably connected to the head of the tripod 9.

[0032] Furthermore, it also includes a swivel seat 10, which is connected to the top plate 6. The binocular camera 1 is rotatably connected to the swivel seat 10. The shooting direction of the binocular camera 1 can be adjusted by rotation. The binocular camera 1 can be rotated at the central shooting point to shoot directly at the palm face, and the binocular camera 1 can be rotated to make the shooting direction of each shooting point consistent after the tripod is placed and leveled at each peripheral shooting point. A locking screw is set on the swivel seat 10, and the binocular camera 1 can be locked by the locking screw after the adjustment is completed.

[0033] Furthermore, it also includes a screw elevator 11, which is a commercially available device. The worm drives the worm wheel to rotate through the worm, and the worm wheel and the screw constitute a screw nut pair, which can drive the screw to rise and fall when the worm wheel rotates. The housing of the screw elevator 11 is fixedly connected to the top plate 6, and the screw of the screw elevator 11 is perpendicular to the top plate 6. The swivel seat 10 is fixedly installed on the top of the screw of the screw elevator 11. The worm of the screw elevator 11 can be rotated to adjust the height of the swivel seat 10 and the binocular camera 1, so as to obtain the images of the tunnel face at different heights at each shooting point, so as to form a complete image of the tunnel face after splicing.

[0034] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for identifying cracks in a tunnel face, characterized in that: The steps include: S1. Setting a central shooting point and a plurality of peripheral shooting points in front of the face. The central shooting point and the peripheral shooting points are set in a flat and safe area according to actual conditions. The central shooting point is set on the axis of the hole in front of the face. S2, taking pictures at different heights at the central shooting point with a binocular camera, keeping the binocular camera horizontal and facing the palm face during shooting, recording the shooting direction of the binocular camera during shooting, importing each taken picture into a computer and recording the actual distance between the binocular camera and the palm face, the focal length and magnification of the binocular camera during each shooting; S3, taking pictures at different heights at each of the peripheral shooting points using a binocular camera, keeping the binocular camera horizontal during shooting and keeping the shooting direction consistent with the direction when shooting at the central shooting point, importing each of the taken pictures into a computer and recording the actual distance between the binocular camera and the palm face, the focal length and magnification of the binocular camera during each shooting; S4, the computer analyzes and calculates the imported data, and calculates the real width and real height corresponding to the picture according to the actual distance between the binocular camera and the palm face corresponding to each picture, the focal length and magnification of the binocular camera, and the physical parameters of the binocular camera; S5. The computer uses an image stitching toolbox to scale the image according to the real width and real height corresponding to the captured image, and then completes the stitching of the panoramic image of the palm face, obtains the panoramic image of the palm face, and calculates the real width and real height corresponding to the panoramic image of the palm face.

2. A method for identifying cracks in a tunnel face according to claim 1, characterized in that: After step S5, the following steps are further performed: S6, the computer identifies and extracts the crack trace information in the panoramic image of the face through a U-net network model, and realizes vectorization of the crack trace image; S7. The computer imports the vectorized image into the fracture occurrence extraction system to obtain fracture length, fracture dip and fracture tendency information.

3. A method for identifying cracks in a tunnel face according to claim 2, characterized in that: The calculation formula for the real width and real height corresponding to the picture obtained in step S4 is as follows: When N≤N0: ， ， When N>N0: ， ， Among them, D is the real width corresponding to the picture; H is the real height corresponding to the picture; f is the focal length of the binocular camera when taking the picture; f0 is the critical focal length of the optical zoom and digital zoom of the binocular camera, which is a physical parameter of the binocular camera; N is the magnification of the binocular camera when taking the picture; N0 is the critical magnification of the optical zoom and digital zoom of the binocular camera, which is a physical parameter of the binocular camera; L is the actual distance between the binocular camera and the palm face when shooting; B is the center line of the two lens sensors in the binocular camera, which is a physical parameter of the binocular camera; h is the height of the lens sensor in the binocular camera, which is a physical parameter of the binocular camera; d is the width of the lens sensor in the binocular camera, which is a physical parameter of the binocular camera.

4. A method for identifying cracks in a tunnel face according to claim 2, characterized in that: Obtaining the crack length information in step S7 refers to importing the vectorized image into the crack occurrence extraction system, extracting the coordinates of each pixel in the image, using the 8-neighborhood tracking method to find the crack endpoints and track the crack path, thereby extracting the skeleton of the entire crack; The distances between consecutive points on the crack path are calculated according to the Euclidean distance formula, and the pixel length of the entire crack is accumulated. Finally, the pixel length of the crack is converted into the actual length according to the true width and true height corresponding to the panoramic image of the face.

5. A method for identifying cracks in a tunnel face according to claim 2, characterized in that: Obtaining the crack inclination information in step S7 refers to importing the vectorized image into the crack occurrence extraction system, extracting the coordinates of each pixel point in the image, performing linear fitting on all pixel coordinate points in a single crack skeleton according to a linear regression algorithm combined with a loop function, obtaining the slope of the straight line, converting the slope into an angle, and thus obtaining the crack apparent inclination; obtaining the crack tendency in step S7 refers to calculating the crack tendency according to the shooting direction of the binocular camera and the crack apparent inclination.

6. A tunnel face image extraction device, characterized in that: A method for identifying cracks on a face of a tunnel for implementing claim 1, comprising a computer and a binocular camera, wherein the binocular camera is provided with an infrared ranging sensor and an electronic compass, wherein the infrared ranging sensor is provided at the front end of the binocular camera, wherein the infrared ranging sensor is provided at the midpoint of the center line between two lens sensors of the binocular camera, wherein the infrared ranging sensor is used to detect the distance between an obstacle photographed in front of the binocular camera and the binocular camera, wherein the electronic compass is used to detect the shooting direction of the binocular camera, and wherein the binocular camera, the infrared sensor and the electronic compass are all electrically connected to the computer.

7. The tunnel face image extraction device according to claim 6, characterized in that: It also includes a leveling seat, which includes a bottom plate and a top plate, the bottom plate and the top plate are connected by three foot screws, a leveler is installed on the top plate, the binocular camera is arranged on the top plate, and the shooting direction of the binocular camera is parallel to the top plate.

8. The tunnel face image extraction device according to claim 7, characterized in that: It also includes a rotating seat, which is connected to the top plate, and the binocular camera is rotatably connected to the rotating seat.

9. The tunnel face image extraction device according to claim 8, characterized in that: It also includes a spiral elevator, the shell of the spiral elevator is fixedly connected to the top plate, the screw of the spiral elevator is perpendicular to the top plate, and the swivel seat is fixedly installed on the top of the screw of the spiral elevator.

10. The tunnel face image extraction device according to claim 7, characterized in that: It also includes a tripod, wherein the head of the tripod is detachably connected to the base plate.

Images