A tunnel construction deformation monitoring system and monitoring method based on machine vision

Abstract

The present invention belongs to the technical field of tunnel construction monitoring. Aiming at the problem that the machine vision equipment used for tunnel deformation monitoring in the prior art is insufficient in device protection and monitoring safety, a machine vision-based tunnel construction deformation monitoring system and a monitoring method thereof are proposed. The tunnel construction deformation monitoring system comprises a machine vision deployment camera, a machine vision monitoring camera, a monitoring target, a data processing platform, a safety alarm, a protection device and a power supply device. The present invention avoids the damage to the camera caused by dust and humidity in the tunnel and reduces the influence of vibration interference on image acquisition through the design of the protection device. At the same time, through the design of the safety alarm, when the tunnel is deformed, the on-site staff can immediately know the abnormal section and abnormal position, and take support measures in time, which greatly reduces the occurrence of safety hazards.

Description

Technical Field

[0001] The present invention belongs to the technical field of tunnel construction monitoring, and in particular relates to a machine vision-based tunnel construction deformation monitoring system and a monitoring method thereof. Background Art

[0002] The current traditional manual deformation monitoring methods of tunnels generally have the disadvantages of long monitoring cycle, low efficiency and low automation, which are difficult to meet the monitoring needs of tunnel projects. The total station robot and distributed fiber optic sensing technology use remote data transmission, which have the advantages of high monitoring efficiency and accuracy, and can achieve long-distance rapid measurement, but they are expensive, the equipment is easily damaged, and requires professional maintenance. At present, machine vision monitoring has been applied to tunnel deformation monitoring, but factors such as dust, humidity, flying stones and vibration in the tunnel construction environment have a great impact on the stability and reliability of machine vision monitoring equipment. The machine vision equipment currently used for tunnel deformation monitoring often adopts a simple connection method to fix the connection with the tunnel wall, without any protection measures or imperfect protection measures, which makes the equipment itself easy to damage, and the monitoring data is unreliable, making it impossible to accurately make tunnel deformation safety warnings. Moreover, the monitoring targets are usually passive targets, which require additional light compensation and increase the power supply lines. At the same time, there is a lack of safety alarm systems and user platforms. When abnormal deformation occurs in the tunnel, the staff cannot be informed immediately, which poses a great safety hazard. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a tunnel construction deformation monitoring system and a monitoring method based on machine vision, which can avoid damage to the equipment caused by humidity, dust, etc. in the tunnel, and effectively solve the influence of vibration interference on the monitoring accuracy; machine vision cameras are used to lay out targets, and the monitoring targets are active light targets and are detachable, which can reduce the monitoring cost; the safety alarm can immediately notify on-site staff and users of abnormal deformation of the tunnel so that protective measures can be taken in time to avoid safety accidents.

[0004] The first object of the present invention is to provide a tunnel construction deformation monitoring system based on machine vision, comprising:

[0005] The machine vision camera is set in the center of the tunnel floor to determine the installation position of each monitoring target in the tunnel section;

[0006] A protective device, which is installed on the inner wall of the tunnel and is used to protect the machine vision monitoring camera;

[0007] The machine vision monitoring camera is installed inside the protective device and is used to monitor and collect clear image data of the monitoring target in real time and upload it to the data receiving platform;

[0008] Monitoring targets are set on the inner wall of the tunnel to monitor the deformation of the tunnel;

[0009] The data receiving platform is used to process the received clear image data of the monitoring target, detect abnormal data, send the detected abnormal data to the security alarm, and issue an alarm instruction;

[0010] Safety alarm, installed on the inner wall of the tunnel, is used to receive alarm instructions and notify abnormal data in the form of sound and light;

[0011] Power supply equipment, used to power machine vision monitoring cameras.

[0012] Preferably, the protective device includes a protective shell, a shock absorbing device and a camera tray, the camera tray and the shock absorbing device are both arranged inside the protective shell, and the camera tray is connected to the inner wall of the protective shell through the shock absorbing device, and the machine vision monitoring camera is placed on the camera tray.

[0013] Preferably, the shock absorbing device comprises a second shock absorbing mechanism, a third shock absorbing mechanism and two first shock absorbing mechanisms, the two first shock absorbing mechanisms are symmetrically arranged on the left and right sides of the camera support plate, and one end of each of the first shock absorbing mechanisms is connected to the second limiting groove arranged on the camera support plate, and the other end is connected to the inner wall of the protective shell;

[0014] One end of the second shock absorbing mechanism is connected to the second limiting groove provided on the rear side of the camera support plate, and the other end is connected to the inner wall of the rear side of the protective shell; one end of the third shock absorbing mechanism is connected to the bottom of the camera support plate, and the other end is connected to the inner wall of the bottom of the protective shell;

[0015] The first shock absorbing mechanism and the second shock absorbing mechanism have the same structure.

[0016] Preferably, the first shock absorbing mechanism comprises a plurality of shock absorbing units arranged in parallel and at intervals, one end of each shock absorbing unit is connected to the second limiting groove, and the other end is connected to the inner wall of the protective shell;

[0017] Each shock absorbing unit comprises a first spring damper, an adjusting mechanism and a T-shaped plate, one end of the first spring damper is connected to the second limiting groove, and the other end is connected to the adjusting mechanism, and the adjusting mechanism is connected to the inner wall of the protective housing through the T-shaped plate;

[0018] The adjustment mechanism includes a clamp, two rotating shafts, two first fixed plates and two first fixed nuts. The clamp includes a sliding plate and two clamping plates of a U-shaped structure. One end of the sliding plate is connected to the first spring damper, and the other end is slidably connected to the flange end of the T-plate. The two clamping plates are symmetrically arranged on the side surfaces of the flange end of the T-plate close to the inner wall of the protective shell, and the two clamping plates are rotatably connected to the two ends of the sliding plate through the rotating shaft. The two first fixed plates are vertically symmetrically arranged on the opposite side surfaces of the web of the T-plate, and the first fixed plate and the clamping plate on the same side are fixed by the first fixing nuts.

[0019] Preferably, the third damping mechanism comprises a weighing platform, a base, a plurality of second limit plates, a plurality of restoring springs, a plurality of fourth limit grooves and a plurality of electronic dampers, the weighing platform is arranged inside the third limit groove, the plurality of fourth limit grooves are respectively opened at the bottom of the weighing platform, the plurality of second limit plates are respectively embedded in the corresponding plurality of fourth limit grooves, and the base is embedded in the inner wall of the bottom of the protective housing;

[0020] Multiple electronic dampers are arranged between multiple second limit plates and the base, and the bottom end of each electronic damper is connected to the base, multiple recovery springs are respectively arranged on the corresponding electronic dampers, and one end of each recovery spring is connected to the corresponding electronic damper, and the other end is connected to the corresponding second limit plate above the recovery spring, and the bottom end is connected to the base, a gravity sensor is arranged inside the weighing platform, and a damping controller is arranged on the base, and the gravity sensor and multiple electronic dampers are electrically connected to the damping controller.

[0021] Preferably, the machine vision-based tunnel construction deformation monitoring system provided by the present invention further includes a target fixing device for fixing the monitoring target, the target fixing device includes a target mounting shell and a limiting cone tube, a notch is provided on the limiting cone tube, a cone opening is provided on the top of the target mounting shell, the bottom of the cone opening passes through the bottom surface of the target mounting shell, and the limiting cone tube is located in the cone opening;

[0022] The steel bar embedded part passes through the limiting cone tube and the target installation shell, and the steel bar embedded part is connected to the target installation shell through a fastening nut. The monitoring target is set on the surface of the target installation shell, and a power supply and an LED lamp are arranged inside the target installation shell.

[0023] The second object of the present invention is to provide a monitoring method of the tunnel construction deformation monitoring system based on machine vision, comprising the following steps:

[0024] Step 1, respectively install monitoring targets on the steel bar embedded parts on the left side, the steel bar embedded parts on the top, and the steel bar embedded parts on the right side of any section S1, and number the monitoring targets as 1, 2, ..., m, m+1, and determine the distance a between the center of each monitoring target and the surface of the concrete layer;

[0025] Step 2: Set up a machine vision layout camera, turn on infrared light compensation, and determine the installation positions of the monitoring targets on section S2, the monitoring targets on section S3, ..., and the monitoring targets on section Sn through the laser transmitter on the machine vision layout camera. Install the monitoring targets according to the installation positions and number them, recorded as 1, 2, ..., m, m+1;

[0026] Step 3: After the monitoring targets in step 1 and step 2 are installed, the machine vision deployment camera is put away, the protective device and the machine vision monitoring camera are installed on the inner wall of the tunnel, the LED lights inside the machine vision monitoring camera and the target installation shell are turned on, and the monitoring range of the machine vision monitoring camera is adjusted;

[0027] Step 4: Adjust the exposure parameters of the machine vision monitoring camera to adjust the imaging quality of all monitoring targets. Then the machine vision monitoring camera captures all monitoring targets and uploads the data to the receiving platform, and sets the target information through the data receiving platform.

[0028] Step 5: Set the shooting interval according to the monitoring needs, the machine vision monitoring camera takes images of all monitoring targets, and uploads the deformation data of all monitoring target positions to the data receiving platform;

[0029] Step 6: Set the cumulative displacement safety threshold of the monitoring target on the data receiving platform, number the safety alarm, and then fix the safety alarm on the corresponding section tunnel wall;

[0030] Step 7. When the deformation data of the monitoring target position received by the data receiving platform exceeds the cumulative displacement safety threshold of the monitoring target, the data receiving platform sends the abnormal deformation section and the monitoring position to the corresponding safety alarm and issues an alarm command. After the safety alarm receives the alarm command, the light turns on and the abnormal deformation position is broadcast.

[0031] Preferably, in step 1 and step 2, all monitoring targets are installed with the steel bar embedded parts by the following method:

[0032] Step 101, align the notch of the limiting cone tube with the steel bar embedded part, push the limiting cone tube, so that the steel bar embedded part is embedded in the limiting cone tube, and realize the connection between the limiting cone tube and the steel bar embedded part;

[0033] Step 102, setting the monitoring target on the surface of the target installation shell, passing the target installation shell through the cone opening from the bottom of the steel bar embedded part and sleeved on the steel bar embedded part, and making the inner surface of the cone opening and the outer surface of the limiting cone tube clamped;

[0034] Step 103, sleeve the fastening nut on the bottom of the steel bar embedded part, twist the fastening nut to move the fastening nut upward along the steel bar embedded part, squeeze the target installation shell, and complete the installation.

[0035] Preferably, step 2 comprises the following steps:

[0036] Step 201, set up a machine vision camera in the center of the tunnel floor, turn on infrared light compensation, and adjust the angle of the machine vision camera so that all monitoring targets of section S1 are within the field of view of the machine vision camera;

[0037] Step 202, the monitoring image quality is adjusted by adjusting the exposure parameters and infrared compensation of the machine vision camera, and then a spatial coordinate system is established based on the relationship between the image size and the shooting distance, with the intersection of the arch steel bar and the concrete layer surface of the section S1 as the coordinate origin O, the section S1 as the xy plane, and the tunnel axis direction as the z direction;

[0038] Step 203, obtaining the x value and y value of the intersection of the steel embedded parts and the concrete layer surface where the monitoring targets on the left side, the top and the right side of the section S1 are located, that is, determining the x value and y value of the intersection of the steel embedded parts and the concrete layer surface where each monitoring target on the sections S2, S3, ..., Sn is located, and determining the z value of the intersection of the steel embedded parts and the concrete layer surface where each monitoring target on the sections S2, S3, ..., Sn is located according to the monitoring section spacing, thereby determining the installation position coordinates of all monitoring targets;

[0039] Step 204, according to the installation position coordinates of all monitoring targets on sections S2, S3, ..., Sn, the laser transmitter on the machine vision camera shines light on the tunnel wall, and all monitoring targets are installed according to the distance a between the center of the monitoring target and the surface of the concrete layer.

[0040] Preferably, in step 5, the monitoring target position deformation data is calculated by the following method:

[0041] Step 501: The machine vision monitoring camera captures monitoring target No. m on section Sn at time i, captures an image and uploads the section information, number information, coordinate value and shooting time of monitoring target No. m;

[0042] Step 502: After a photo-taking interval, the machine vision monitoring camera captures the monitoring target m of the section Sn at the i+1th time, takes an image, and uploads the section information, number information, coordinate value, and shooting time of the monitoring target m;

[0043] Step 503, establish the correspondence between the monitoring targets with the same number in the same section in the image at the i-th moment and the i+1-th moment, calculate the relative difference in the coordinates of the monitoring target number m captured at the i-th moment and the i+1-th moment and upload it, the relative difference being the position deformation data of the monitored position at the i+1-th moment compared to the monitoring target number m at the i-th moment.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The machine vision-based tunnel construction deformation monitoring system provided by the present invention avoids the damage to the camera caused by dust and humidity in the tunnel through the design of the protective device, and reduces the influence of vibration interference on image acquisition; and the present invention adopts machine vision to deploy the camera to determine the target deployment position, which greatly reduces the monitoring cost compared with the deployment of the total station; the monitoring target is an active light target and can be recycled, which is conducive to saving monitoring costs; at the same time, through the design of the safety alarm, when the tunnel is deformed, the on-site staff can immediately know the abnormal section and abnormal position, and take support measures in time, which greatly reduces the occurrence of safety hazards. The monitoring method provided by the present invention is simple, has high monitoring efficiency, low monitoring cost, and strong practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a framework diagram of a tunnel construction deformation monitoring system based on machine vision provided in Example 1 of the present invention;

[0047] Figure 2 This is an overall monitoring schematic diagram of a tunnel construction deformation monitoring system based on machine vision provided in Example 1 of the present invention;

[0048] Figure 3 is a schematic diagram of the connection structure of the protective device and the machine vision monitoring camera in Example 1 of the present invention;

[0049] Figure 4 is a schematic diagram of the internal structure of the protective housing in Embodiment 1 of the present invention;

[0050] Figure 5 is a top view of the camera support plate in Embodiment 1 of the present invention;

[0051] Figure 6 is a bottom view of the camera support plate in Embodiment 1 of the present invention;

[0052] Figure 7 is a top view of the internal structure of the protective housing in Embodiment 1 of the present invention;

[0053] Figure 8 is a front view of the internal structure of the protective housing in Embodiment 1 of the present invention;

[0054] Fig. 9 is a schematic diagram of the connection structure between the shock absorbing unit, the second limiting groove and the protective housing in Example 1 of the present invention;

[0055] Fig.10 is a schematic structural diagram of the third damping mechanism in Embodiment 1 of the present invention;

[0056] Fig.11 is a schematic structural diagram of a target fixing device in Example 1 of the present invention;

[0057] Fig.12 It is a schematic diagram of the overall process of the tunnel construction deformation monitoring method based on machine vision provided in Example 2 of the present invention;

[0058] Fig.13 is a schematic flow chart of step S2 in embodiment 2 of the present invention;

[0059] Fig.14 It is a structural schematic diagram of the machine vision camera layout target layout in Example 2 of the present invention;

[0060] Fig.15 is a schematic diagram of establishing a coordinate system in Embodiment 2 of the present invention;

[0061] In the figure: machine vision monitoring camera-1, protective housing-2, dust filter-21, heat dissipation hole-22, moisture-proof layer-23, dust-proof glass-24, bolt hole-25, shock absorbing device-3, first shock absorbing mechanism-31, return spring-311, damper-312, first limit plate-313, sliding plate-314, rotating shaft-315, card plate-316, second shock absorbing mechanism-32, third shock absorbing mechanism-33, electronic damper-331, second limit plate-332, return spring-333, T-plate -34, first fixing plate-35, first fixing nut-36, weighing platform-37, fourth limiting groove-372, base-38, damping controller-381, camera support plate-4, first limiting groove-41, second limiting groove-42, third limiting groove-43, buffer pad-44, machine vision layout camera-5, monitoring target-6, target fixing device-7, limiting cone tube-71, notch-711, target mounting shell-72, cone mouth-721, fastening nut-73, safety alarm-8. DETAILED DESCRIPTION

[0062] The following is a description of the embodiments of the present invention. Figures 1 to 15 , the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] Embodiment 1:

[0064] like Figures 1 to 11 As shown, an embodiment of the present invention provides a tunnel construction deformation monitoring system based on machine vision, comprising:

[0065] The machine vision camera 5 is arranged in the center of the tunnel floor and is used to determine the installation position of each monitoring target 6 in the tunnel section. The machine vision camera 5 in the embodiment of the present invention is equipped with a laser emitter and an infrared light compensator. Through the configuration of the laser emitter and the infrared light compensator, the position of the monitoring target 6 can be determined more accurately.

[0066] A machine vision monitoring camera 1 is arranged inside the protective device and is used to monitor and collect clear image data of the monitoring target 6 in real time and upload it to the data receiving platform;

[0067] The monitoring target 6 is arranged on the inner wall of the tunnel to monitor the deformation of the tunnel. The monitoring target 6 in the embodiment of the present invention is made of a light-transmitting material and is pasted on the surface of the target mounting shell 72 .

[0068] The data receiving platform is used to process the received clear image data of the monitoring target 6, detect abnormal data, send the detected abnormal data to the security alarm 8, and issue an alarm instruction;

[0069] The safety alarm 8 is arranged on the inner wall of the tunnel, and is used to receive alarm instructions and send out notifications of abnormal data in the form of sound and light. In the embodiment of the present invention, a wireless communication module is arranged inside the safety alarm 8, and the safety alarm 8 is fixedly installed on the tunnel wall of the monitoring section. The safety alarm 8 is provided with a sound alarm and a light alarm. Through the arrangement of the sound alarm and the light alarm, the on-site staff can be notified in the form of sound and light when abnormal deformation sections and monitoring positions appear. The machine vision monitoring camera 1 in the embodiment of the present invention is connected to the data receiving platform through a wired network to obtain stable data transmission. The safety alarm 8 is wirelessly connected to the data receiving platform. A wireless communication module is arranged inside the target installation shell 72, and the wireless communication module is wirelessly connected to the data receiving platform. The wireless communication information module is connected to the LED light arranged in the target installation shell 72. The data receiving platform can control the opening or closing of the LED light through the wireless communication information module according to the monitoring task requirements, thereby controlling the light or non-light of the monitoring target. The machine vision deployment camera 5 and the data receiving platform are connected to the power supply device.

[0070] A power supply device, used to supply power to the machine vision monitoring camera 1;

[0071] Specifically, the machine vision deployment camera 5 used for deploying the monitoring target 6 has a short usage time. After the spatial coordinate system is established to determine the deployment position of the monitoring target 6, the machine vision deployment camera 5 can be put away without the need for protective devices. Considering the dim tunnel environment, the machine vision deployment camera 5 needs to be equipped with an infrared light compensator for fill light, and a laser transmitter is needed to mark the deployment position of the monitoring target 6 after determining the deployment position. The machine vision monitoring camera 1 mainly performs long-term monitoring tasks, so it needs to be equipped with protective devices, while the monitoring target 6 is a target with a light source and does not need to be equipped with an infrared light compensator.

[0072] The protective device is arranged on the inner wall of the tunnel and is used to protect the machine vision monitoring camera 1.

[0073] The protective device includes a protective shell 2, a shock absorbing device 3 and a camera support plate 4. The camera support plate 4 and the shock absorbing device 3 are both arranged inside the protective shell 2, and the camera support plate 4 is connected to the inner wall of the protective shell 2 through the shock absorbing device 3. The machine vision monitoring camera 1 is placed on the camera support plate 4. The top of the protective shell 2 in the embodiment of the present invention is detachably provided with a top cover, which is specifically covered on the top of the protective shell 2. Bolt holes are arranged at the four corners of the top of the protective shell 2. The top cover is fixed to the bolt holes 25 on the protective shell 2 by bolts; by opening the top cover, the height of the first spring damper and the second spring damper is adjusted, and the machine vision monitoring camera 1 is put in or taken out.

[0074] In the embodiment of the present invention, the protective shell 2 includes a dustproof glass 24, two dust filters 21, two heat dissipation holes 22 and two moisture-proof layers 23. An opening is provided on the front side of the protective shell 2, and the dustproof glass 24 is arranged at the opening. The dustproof glass 24 is transparent glass. Since the dust is large during the tunnel construction period, the dustproof glass 24 is arranged on the front side of the protective shell 2 so that the dustproof glass 24 is located directly in front of the viewing angle of the machine vision monitoring camera 1. In this way, not only will the machine vision monitoring camera 1 not be blocked, so that the tunnel deformation monitoring can be carried out smoothly, but also dust can be prevented from accumulating in the protective shell 2 and affecting the line of sight; the two heat dissipation holes 22 are symmetrically opened on the protective shell 2, and the dustproof glass 24 is arranged on the front side of the protective shell 2. On the two opposite sides of the housing 2, the heat generated by the machine vision monitoring camera 1 during operation can be dissipated through the design of the heat dissipation holes 22, so as to avoid heat accumulation and overheating and damage to the machine vision monitoring camera 1; two dust filters 21 are respectively arranged in the corresponding heat dissipation holes 22. Since the dust concentration is relatively high during tunnel construction, dust will enter the protective housing 2 from the heat dissipation holes 22, which may cause damage to the machine vision monitoring camera 1 for a long time, so the embodiment of the present invention further adds dust filters 21 in the heat dissipation holes 22 to prevent dust from entering; two moisture-proof layers 23 are respectively symmetrically arranged inside the protective housing 2, and the position of each moisture-proof layer 23 is directly opposite to the corresponding heat dissipation holes 22. Since the air is humid during tunnel construction, water vapor will enter the protective housing 2 from the heat dissipation holes 22, which may cause damage to the machine vision monitoring camera 1 for a long time, so the moisture-proof layer 23 is designed. In the specific design of this embodiment, the moisture-proof layer 23 and the protective housing 2 are detachably connected. Through such a design, when the moisture-proof layer 23 has a high humidity and poor water absorption effect, it is convenient to disassemble and replace the moisture-proof layer 23, thereby avoiding damage to the machine vision monitoring camera 1 caused by water vapor. At the same time, when setting the position of the moisture-proof layer, not only should the position of each moisture-proof layer 23 be aligned with the corresponding heat dissipation hole 22, but there should also be a certain distance between the moisture-proof layer 23 and the heat dissipation hole 22, so that it can absorb water vapor without affecting the heat dissipation of the machine vision monitoring camera 1.

[0075] In the embodiment of the present invention, the upper surface and the lower surface of the camera support plate 4 are respectively provided with a first limiting groove 41 and a third limiting groove 43, the first limiting groove 41 is used to place the machine vision monitoring camera 1, and the left side, right side and rear side of the camera support plate 4 are respectively provided with a plurality of second limiting grooves 42, and a buffer layer 44 is arranged inside the first limiting groove 41 and the third limiting groove 43. Through the design of the first limiting groove 41 and the third limiting groove 43, the displacement of the machine vision monitoring camera 1 and the weighing platform 37 can be limited respectively, and the design of the buffer layer 44 can reduce the vibration interference from the protective housing 2; through the cooperation of the second limiting groove 42 and the first limiting plate 313, the camera support plate 4 can be prevented from overturning.

[0076] In the embodiment of the present invention, the shock absorbing device 3 includes a second shock absorbing mechanism 32, a third shock absorbing mechanism 33 and two first shock absorbing mechanisms 31, the two first shock absorbing mechanisms 31 are symmetrically arranged on the left and right sides of the camera support plate 4, and one end of each of the first shock absorbing mechanisms 31 is connected to the corresponding second limiting groove 42, and the other end is connected to the inner wall of the protective shell 2; one end of the second shock absorbing mechanism 32 is connected to the second limiting groove 42 on the rear side of the camera support plate 4, and the other end is connected to the inner wall of the rear side of the protective shell 2; one end of the third shock absorbing mechanism 33 is connected to the bottom of the camera support plate 4, and the other end is connected to the bottom inner wall of the protective shell 2; the first shock absorbing mechanism 31 and the second shock absorbing mechanism 32 have the same structure.

[0077] In the embodiment of the present invention, the first damping mechanism 31 includes a plurality of damping units arranged in parallel and spaced apart from each other, one end of each damping unit is connected to the corresponding second limiting groove 42, and the other end is connected to the inner wall of the protective housing 2;

[0078] Each shock absorbing unit includes a first spring damper, an adjusting mechanism and a T-shaped plate 34 . One end of the first spring damper is connected to the second limiting groove 42 , and the other end is connected to the adjusting mechanism. The adjusting mechanism is connected to the inner wall of the protective shell 2 through the T-shaped plate 34 .

[0079] In the embodiment of the present invention, the first spring damper includes a return spring 311, a damper 312 and a first limit plate 313. The return spring 311 is arranged on the damper 312, and one end of the return spring 311 is connected to the damper 312, and the other end is connected to the first limit plate 313. The first limit plate 313 is inserted and arranged inside the second limit groove 42, and one end of the damper 312 away from the first limit plate 313 is connected to the adjustment mechanism. In this embodiment, the return spring 311 is specifically wound around the damper 312, and one end of the return spring 311 is connected to the base of the damper 312, and the other end is connected to the first limit plate 313.

[0080] In the embodiment of the present invention, the adjustment mechanism includes a clamping plate, two rotating shafts 315, two first fixing plates 35 and two first fixing nuts 36. The clamping plate includes a sliding plate 314 and two clamping plates 316 of a U-shaped structure. One end of the sliding plate 314 is connected to the first spring damper, and the other end is slidably connected to the flange end of the T-shaped plate 34 (specifically, the groove portion of the sliding plate 314 is slidably clamped with the flange end of the T-shaped plate 34). The sliding plate 314 is slidably connected to the flange end of the T-shaped plate 34, so that the sliding plate 314 can slide up and down along the flange end of the T-shaped plate 34, thereby driving the first spring damper to move up and down. Therefore, it can match the heights of different camera trays 4. The two clamping plates 316 are symmetrically arranged on one side of the flange end of the T-shaped plate 34 close to the inner wall of the protective shell 2. Specifically, there is a certain gap between the clamping plate 316 and the surface of the flange end of the T-shaped plate 34, and the two clamping plates 316 are rotatably connected to the two ends of the sliding plate 314 through the rotating shaft 315. The two first fixed plates 35 are vertically symmetrically arranged on the opposite side surfaces of the web of the T-shaped plate 34. The first fixed plate 35 and the clamping plate 316 on the same side are fixed by the first fixing nut 36. The first fixed plate 35 is reserved with a threaded hole for the first fixing nut 36 to pass through. By rotating the card plate 316 to the outside of the T-shaped plate 34 through the rotating shaft 315, the sliding plate 314 can slide up and down along the flange end of the T-shaped plate 34 to adjust the height of the first spring damper and the second spring damper. When the height is adjusted, the card plate 316 is rotated in the opposite direction to the inside of the flange of the T-shaped plate 34, and the first fixing nut 36 is screwed to make it pass through the first fixing plate 35, thereby squeezing the card plate 316 and pushing the card plate 316 to move forward until it is tightly against the reset spring 311, thereby fixing the card plate 316 and the T-shaped plate. At this time, the card plate 316 moves forward, driving the first limiting plate 313 to be inserted into the inside of the second limiting groove 42.

[0081] In the embodiment of the present invention, the third shock absorbing mechanism 33 includes a weighing platform 37, a base 38, a plurality of restoring springs 333, a plurality of second limiting plates 332, a plurality of fourth limiting grooves 372 and a plurality of electronic dampers 331, the weighing platform 37 is arranged inside the third limiting groove 43, a plurality of fourth limiting grooves 372 are respectively opened at the bottom of the weighing platform 37, a plurality of second limiting plates 332 are respectively embedded in the corresponding plurality of fourth limiting grooves 372, the base 38 is embedded on the inner wall of the bottom of the protective housing 2, and a plurality of electronic dampers 331 are arranged on the plurality of second limiting grooves 372. The positioning plate 332 and the base 38 are connected, and the bottom end of each electronic damper 331 is connected to the base 38. Multiple restoring springs 333 are respectively arranged on the corresponding multiple electronic dampers 331, and one end of each restoring spring 333 is connected to the corresponding electronic damper 331, and the other end is connected to the corresponding second limiting plate 332 above the restoring spring 333. A gravity sensor is arranged inside the weighing platform 37, and a damping controller 381 is arranged on the base 38. The gravity sensor and the multiple electronic dampers 331 are respectively electrically connected to the damping controller 381.

[0082] After the machine vision monitoring camera 1 is placed on the camera support plate 4, the restoring spring 333 on the electronic damper 331 is compressed due to gravity, and its spring length is shortened, and the shock absorption capacity is weakened. In order to solve this problem, a gravity sensor is set in the weighing platform 37. After obtaining the gravity information of the machine vision monitoring camera 1, the gravity sensor transmits an electrical signal to the damping controller 381 set in the base 38. The damping controller 381 performs damping compensation on the electronic damper 331 according to the electrical signal, so that the shock absorption capacity of the electronic damper 331 is the same as that of the first spring damper and the second spring damper.

[0083] After the machine vision monitoring camera 1 is placed on the camera support plate 4, the restoring spring 333 of the electronic damper 331 is compressed due to gravity, and its spring length is shortened, and the camera support plate 4 drops. At this time, the first spring damper and the second spring damper are adjusted so that the height of the first limit plate 313 is consistent with the height of the second limit groove 42 on the surface of the camera support plate 4, and then the end of the card plate 316 is rotated to the inner side of the flange of the T-shaped plate 34 through the rotating shaft 315, and the first fixing nut 36 is screwed to pass through the first fixing plate 35, thereby squeezing the card plate 316 to play a fixing role. At the same time, due to the thrust, the surface of the first limit plate 313 of the first spring damper and the second spring damper is in contact with the groove surface of the second limit groove 42. When the embodiment of the present invention is specifically designed, the distance from the rotating shaft 315 to the end of the clamping plate 316 should be smaller than the distance from the end of the clamping plate 316 to the first fixed plate 35 after the end of the clamping plate 316 rotates to the inner side of the flange of the T-shaped plate 34, so as to prevent the end of the clamping plate 316 from being unable to rotate to the inner side of the flange of the T-shaped plate 34.

[0084] This embodiment can perform damping compensation in the direction of gravity according to the weight of the monitoring device, and the heights of the first spring damper and the second spring damper are adjustable, so the present invention is applicable to different monitoring devices.

[0085] The machine vision-based tunnel construction deformation monitoring system provided by an embodiment of the present invention also includes a target fixing device 7 for fixing the monitoring target, the target fixing device 7 includes a target mounting shell 72 and a limiting cone tube 71, a notch 711 is provided on the limiting cone tube 71, a cone opening 721 is provided on the top of the target mounting shell 72, the bottom of the cone opening 721 passes through the bottom surface of the target mounting shell 72, that is, the bottom end of the cone opening 721 is connected to a through hole passing through the bottom of the target mounting shell 72, and the limiting cone tube 71 is located in the cone opening 721; the steel bar embedded part passes through the limiting cone tube 71 and the target mounting shell 72, and the steel bar embedded part is connected to the target mounting shell 72 through a fastening nut 73, the monitoring target 6 is arranged on the surface of the target mounting shell 72, and the target mounting shell 72 is provided with a power supply, an LED lamp and a wireless communication module.

[0086] The inner diameter of the limiting cone tube 71 at each position in the embodiment of the present invention is equal and equal to the diameter of the embedded steel bar; the inner wall of the limiting cone tube 71 is provided with an internal thread, and the limiting cone tube 71 is connected to the external thread on the embedded steel bar through the internal thread. In the embodiment of the present invention, the interval between the inner grooves of the internal thread is consistent with the interval between the external threads on the embedded steel bar, that is, after the embedded steel bar enters the interior of the limiting cone tube 71 through the notch 711, the external thread of the embedded steel bar and the internal thread of the limiting cone tube 71 will abut against each other, so that the limiting cone tube 71 will not move arbitrarily on the embedded steel bar.

[0087] The top inner diameter of the conical opening 721 is smaller than the top outer diameter of the limiting cone tube 71, and the bottom inner diameter of the conical opening 721 is equal to the bottom outer diameter of the limiting cone tube 71. In this embodiment, the outer surface of the limiting cone tube 71 matches the inner surface of the conical opening 721, so that when the limiting cone tube 71 enters the conical opening 721, it can be in close contact with the inner surface of the conical opening 721 and initially limit the limiting cone tube 71.

[0088] Specifically, the monitoring target 6 in the embodiment of the present invention is not a disposable target, but the monitoring target 6 is pasted on the target installation shell 72, and then the limiting cone tube with a notch 711 and a threaded groove is installed on the embedded steel bar with a thread, so that the position of the limiting cone tube 71 is fixed, and then the target installation shell 72 is installed, and then the fastening nut is screwed to fix the target installation shell. The target installation shell 72 is equipped with a wireless communication module, a power supply and an LED light, and the wireless communication module can receive the command to turn on or off the LED light from the data receiving platform.

[0089] Embodiment 2:

[0090] like Figures 12 to 15 As shown, the embodiment of the present invention provides a tunnel construction deformation monitoring method based on machine vision, which is implemented using the tunnel construction deformation monitoring system based on machine vision described in Example 1, and includes the following steps:

[0091] Step 1. After the concrete spraying is completed, monitoring targets 6 are respectively installed on the steel embedded parts on the left side, the top and the right side of any section S1, and the monitoring targets 6 are numbered as 1, 2, ..., m, m+1, and the distance between the center of each monitoring target 6 and the surface of the concrete layer is determined to be a; the direction of the monitoring target 6 is opposite to the tunnel excavation direction, and the number of monitoring targets 6 is determined by the section size.

[0092] Step 2: Set up the machine vision layout camera 5, turn on the infrared light compensation, and use the laser transmitter on the machine vision layout camera 5 to mark the installation positions of the monitoring target 6 on the section S2, the monitoring target 6 on the section S3, ..., and the monitoring target 6 on the section Sn. Install the monitoring target 6 according to the installation position, and number the monitoring target 6, record it as 1, 2, ..., m, m+1, which specifically includes the following steps:

[0093] Step 201, set up a machine vision camera 5 in the center of the tunnel floor, turn on the infrared light compensator, and adjust the angle of the machine vision camera 5 so that all monitoring targets 6 of the section S1 are within the field of view of the machine vision camera 5;

[0094] Step 202, the monitoring image quality is adjusted by adjusting the exposure parameters and infrared compensation of the machine vision camera 5 to obtain a higher monitoring image quality, and then a spatial coordinate system is established based on the relationship between the image size and the shooting distance, with the intersection of the arch steel bar and the concrete layer surface of the section S1 as the coordinate origin O, the section S1 as the xy plane, and the tunnel axis direction as the z direction;

[0095] Step 203, obtaining the x value and y value of the intersection of the embedded steel bar parts and the concrete layer surface where the monitoring targets 6 on the left side, the top and the right side of the section S1 are located. Since the embedded steel bars on each section are set at the same position, when the x value and y value of the intersection of the embedded steel bar parts and the concrete layer surface where each monitoring target 6 on the section S1 is located are obtained, the x value and y value of the intersection of the embedded steel bar parts and the concrete layer surface where all the monitoring targets 6 on the sections S2, S3, ..., Sn are located are determined, and the monitoring section spacing is determined according to the existing construction specifications for railway tunnels. The z value of the intersection of the embedded steel bar parts and the concrete layer surface where each monitoring target 6 on the sections S2, S3, ..., Sn is located is determined based on the monitoring section spacing, thereby determining the installation position coordinates of all the monitoring targets 6;

[0096] Step 204, according to the installation position coordinates of all monitoring targets 6 on sections S2, S3, ..., Sn, the laser emitter on the machine vision camera 5 shines light on the tunnel wall, and all monitoring targets 6 are installed according to the distance a between the center of the monitoring target 6 and the surface of the concrete layer.

[0097] In step 1 and step 2, the monitoring target 6 is installed with the steel bar embedded part by the following method:

[0098] Step 101, align the notch 711 of the limiting cone tube 71 with the steel bar embedded part, push the limiting cone tube 71, so that the steel bar embedded part is embedded in the limiting cone tube 71, and the limiting cone tube 71 is connected with the steel bar embedded part;

[0099] Step 102, stick the monitoring target 6 to the surface of the target installation shell 72, pass the target installation shell 72 through the cone opening 721 from the bottom of the steel bar embedded part and put it on the steel bar embedded part, and make the inner surface of the cone opening 721 clamped with the outer surface of the limiting cone tube 71;

[0100] Step 103, put the fastening nut 73 on the bottom of the steel bar embedded part, screw the fastening nut 73 to move the fastening nut 73 up along the steel bar embedded part, squeeze the target mounting shell 72, complete the installation, wait for the section monitoring task to be completed, remove the target mounting shell for the next section monitoring.

[0101] After the monitoring targets 6 in step 3 and step 1 and the monitoring targets 6 in step 2 are installed, the machine vision deployment camera 5 is put away, the protective device and the machine vision monitoring camera 1 are installed on the inner wall of the tunnel, the machine vision monitoring camera 1 and the LED lights inside the target installation shell 72 are turned on, and the monitoring range of the machine vision monitoring camera 1 is adjusted, that is, the monitoring targets 6 of each section are within the monitoring range by adjusting the angle of the machine vision monitoring camera 1;

[0102] Step 4: adjust the exposure parameters of the machine vision monitoring camera 1, and adjust the imaging quality of all monitoring targets 6, so as to obtain clear images of each monitoring target in each section. Then the machine vision monitoring camera 1 automatically captures all monitoring targets 6 and uploads them to the data receiving platform. The section in which the captured monitoring targets are located is determined according to the different sizes of the captured monitoring targets, and the section number and target number of each monitoring target are set through the data receiving platform.

[0103] Step 5: Set the shooting interval of the machine vision monitoring camera 1 according to the monitoring needs. The machine vision monitoring camera 1 takes images of all the monitoring targets 6. Each time the machine vision monitoring camera 1 takes a picture, the captured image and the shooting time are uploaded to the data receiving platform, and the deformation data of all the monitoring target positions are uploaded to the data receiving platform. The data receiving platform presents the received information in the form of charts.

[0104] In step 5, the monitoring target position deformation data is calculated by the following method:

[0105] Step 501: The machine vision monitoring camera 1 captures the monitoring target 6 No. m of the section Sn at the i-th moment, captures an image and uploads the section information, number information, coordinate value and shooting time of the monitoring target 6 No. m;

[0106] Step 502: After a photo-taking interval, the machine vision monitoring camera 1 captures the monitoring target 6 of section Sn at the i+1th time, takes an image and uploads the section information, number information, coordinate value and shooting time of the monitoring target 6;

[0107] Step 503, establish the correspondence between the monitoring target 6 with the same number in the same section in the image at the i-th moment and the i+1-th moment, calculate the relative difference in the coordinates of the monitoring target 6 with number m captured at the i-th moment and the i+1-th moment and upload it, the relative difference being the position deformation data of the monitoring target 6 with number m at the monitored position at the i+1-th moment compared to the i-th moment.

[0108] Step 6: Set the cumulative displacement safety threshold of the monitoring target 6 on the data receiving platform, number the safety alarm 8 in advance, and the number is consistent with the monitoring section number. After the safety alarm 8 is numbered, fix the safety alarm 8 on the corresponding section tunnel wall;

[0109] Step 7: When the deformation data of the monitoring target position received by the data receiving platform exceeds the monitoring target cumulative displacement safety threshold, the data receiving platform sends the abnormal deformation section and monitoring position to the corresponding safety alarm 8 and issues an alarm command. After the safety alarm 8 receives the alarm command, the light alarm lights up and the abnormal deformation position is broadcast through the sound alarm. After the staff checks the abnormal deformation position and takes measures, the cumulative displacement value returns to the safety threshold, and the data receiving platform sends a shutdown alarm command to the safety alarm 8.

[0110] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A tunnel construction deformation monitoring system based on machine vision, characterized in that: include: A machine vision camera (5) is arranged at the center of the tunnel floor and is used to determine the installation position of each monitoring target (6) on the tunnel section; A protective device, arranged on the inner wall of the tunnel, for protecting the machine vision monitoring camera (1); The protective device comprises a protective shell (2), a shock absorbing device (3) and a camera support plate (4); the camera support plate (4) and the shock absorbing device (3) are both arranged inside the protective shell (2); the camera support plate (4) is connected to the inner wall of the protective shell (2) via the shock absorbing device (3); and the machine vision monitoring camera (1) is placed on the camera support plate (4); The protective shell (2) comprises a dustproof glass (24), two dust filters (21), two heat dissipation holes (22) and two moisture-proof layers (23); an opening is provided on the front side of the protective shell (2); the dustproof glass (24) is arranged at the opening; the two heat dissipation holes (22) are symmetrically provided on two opposite sides of the protective shell (2); the two dust filters (21) are respectively arranged in the corresponding heat dissipation holes (22); the two moisture-proof layers (23) are respectively symmetrically provided inside the protective shell (2), and each moisture-proof layer (23) is located directly opposite to the corresponding heat dissipation hole (22); The shock absorbing device (3) comprises a second shock absorbing mechanism (32), a third shock absorbing mechanism (33) and two first shock absorbing mechanisms (31), the two first shock absorbing mechanisms (31) being symmetrically arranged on the left and right sides of the camera support plate (4), respectively, and one end of each of the first shock absorbing mechanisms (31) is connected to a second limiting groove (42) arranged on the camera support plate (4), and the other end is connected to the inner wall of the protective housing (2); One end of the second shock absorbing mechanism (32) is connected to a second limiting groove (42) provided on the rear side surface of the camera support plate (4), and the other end is connected to the inner wall of the rear side surface of the protective shell (2); one end of the third shock absorbing mechanism (33) is connected to the bottom of the camera support plate (4), and the other end is connected to the inner wall of the bottom of the protective shell (2); The first shock absorbing mechanism (31) and the second shock absorbing mechanism (32) have the same structure; The first shock absorbing mechanism (31) comprises a plurality of shock absorbing units arranged in parallel and at intervals, one end of each shock absorbing unit being connected to the second limiting groove (42) and the other end being connected to the inner wall of the protective housing (2); Each shock absorbing unit comprises a first spring damper, an adjustment mechanism and a T-shaped plate (34); one end of the first spring damper is connected to the second limiting groove (42), and the other end is connected to the adjustment mechanism; the adjustment mechanism is connected to the inner wall of the protective housing (2) via the T-shaped plate (34); The adjustment mechanism comprises a clamping plate, two rotating shafts (315), two first fixing plates (35) and two first fixing nuts (36); the clamping plate comprises a sliding plate (314) of a U-shaped structure and two clamping plates (316); one end of the sliding plate (314) is connected to the first spring damper, and the other end is slidably connected to the flange end of the T-shaped plate (34); the two clamping plates (316) are symmetrically arranged on the side of the flange end of the T-shaped plate (34) close to the inner wall of the protective shell (2), and the two clamping plates (316) are rotatably connected to the two ends of the sliding plate (314) through the rotating shafts (315); the two first fixing plates (35) are vertically symmetrically arranged on the opposite side surfaces of the web of the T-shaped plate (34); the first fixing plate (35) and the clamping plate (316) on the same side are fixed by the first fixing nuts (36); The third damping mechanism (33) comprises a weighing platform (37), a base (38), a plurality of second limiting plates (332), a plurality of fourth limiting grooves (372), a plurality of restoring springs (333) and a plurality of electronic dampers (331); the weighing platform (37) is arranged inside the third limiting groove (43); the plurality of fourth limiting grooves (372) are respectively opened at the bottom of the weighing platform (37); the plurality of second limiting plates (332) are respectively embedded in the corresponding plurality of fourth limiting grooves (372); and the base (38) is embedded in the bottom inner wall of the protective housing (2); A plurality of electronic dampers (331) are arranged between a plurality of second limit plates (332) and a base (38), and the bottom end of each electronic damper (331) is connected to the base (38); a plurality of restoring springs (333) are respectively arranged on the corresponding plurality of electronic dampers (331), and one end of each restoring spring (333) is connected to the corresponding electronic damper (331), and the other end is connected to the corresponding second limit plate (332) above the restoring spring (333); a gravity sensor is arranged inside the weighing platform (37), a damping controller (381) is arranged on the base (38), and the gravity sensor and the plurality of electronic dampers (331) are electrically connected to the damping controller (381); A machine vision monitoring camera (1) is arranged inside the protective device and is used to monitor and collect clear image data of the monitoring target (6) in real time and upload it to a data receiving platform; A monitoring target (6) is arranged on the inner wall of the tunnel and is used to monitor the deformation of the tunnel; A data receiving platform, used to process the received clear image data of the monitoring target (6), detect abnormal data, send the detected abnormal data to the security alarm (8), and issue an alarm instruction; A safety alarm (8) is arranged on the inner wall of the tunnel and is used to receive an alarm instruction and send out notifications of abnormal data in the form of sound or light; A power supply device is used to supply power to a machine vision monitoring camera (1).

2. The machine vision-based tunnel construction deformation monitoring system according to claim 1, characterized in that: Also included is a target fixing device (7) for fixing the monitoring target, the target fixing device (7) comprising a target mounting shell (72) and a limiting cone tube (71), the limiting cone tube (71) being provided with a notch (711), the top of the target mounting shell (72) being provided with a cone opening (721), the bottom of the cone opening (721) penetrating the bottom surface of the target mounting shell (72), and the limiting cone tube (71) being located within the cone opening (721); The steel bar embedded part penetrates the limiting cone tube (71) and the target installation shell (72), and the steel bar embedded part is connected to the target installation shell (72) via a fastening nut (73). The monitoring target (6) is arranged on the surface of the target installation shell (72), and a power supply and an LED lamp are arranged inside the target installation shell (72).

3. A tunnel construction deformation monitoring method based on machine vision, characterized in that: The tunnel construction deformation monitoring system based on machine vision described in claim 2 is implemented, comprising the following steps: Step 1, respectively install monitoring targets (6) on the steel bar embedded parts on the left side, the steel bar embedded parts on the top, and the steel bar embedded parts on the right side of any section S1, and number the monitoring targets (6) as 1, 2, ..., m, m+1, and determine the distance a between the center of the monitoring target (6) and the surface of the concrete layer; Step 2, set up a machine vision camera (5), turn on infrared light compensation, determine the installation positions of the monitoring targets (6) on section S2, the monitoring targets (6) on section S3, ..., and the monitoring targets (6) on section Sn through the laser transmitter on the machine vision camera (5), install the monitoring targets (6) according to the installation positions, and number the monitoring targets (6), record them as 1, 2, ..., m, m+1; Step 2 contains the following steps: Step 201, set up a machine vision camera (5) in the center of the tunnel floor, turn on infrared light compensation, and adjust the angle of the machine vision camera (5) so that all monitoring targets (6) of the section S1 are within the field of view of the machine vision camera (5); Step 202, the quality of the monitoring image is adjusted by adjusting the exposure parameters and infrared compensation of the machine vision camera (5), and then a spatial coordinate system is established based on the relationship between the image size and the shooting distance, with the intersection of the arch steel bar and the concrete layer surface of the section S1 as the coordinate origin O, the section S1 as the xy plane, and the tunnel axis direction as the z direction; Step 203, obtaining the x value and y value of the intersection of the steel embedded parts and the concrete layer surface where the monitoring targets (6) on the left side, the top and the right side of the section S1 are located, that is, determining the x value and y value of the intersection of the steel embedded parts and the concrete layer surface where all the monitoring targets (6) on the sections S2, S3, ..., Sn are located, and determining the z value of the intersection of the steel embedded parts and the concrete layer surface where each monitoring target (6) on the sections S2, S3, ..., Sn is located according to the monitoring section spacing, thereby determining the installation position coordinates of all the monitoring targets (6); Step 204, according to the installation position coordinates of all monitoring targets (6) on the sections S2, S3, ..., Sn, the laser emitter on the machine vision camera (5) shines light on the tunnel wall, and all monitoring targets (6) are installed according to the distance a between the center of the monitoring target (6) and the surface of the concrete layer; After the monitoring targets (6) in step 3 and step 2 are installed, the machine vision deployment camera (5) is folded, a protective device and a machine vision monitoring camera (1) are installed on the inner wall of the tunnel, the LED lights inside the machine vision monitoring camera (1) and the target installation shell (72) are turned on, and the monitoring range of the machine vision monitoring camera (1) is adjusted; Step 4: adjusting the exposure parameters of the machine vision monitoring camera (1) to adjust the imaging quality of all monitoring targets (6); then the machine vision monitoring camera (1) captures all monitoring targets (6) and uploads them to the data receiving platform, and sets the target information through the data receiving platform; Step 5: The photographing interval is set according to the monitoring needs, and the machine vision monitoring camera (1) photographs the images of all the monitoring targets (6), and uploads the deformation data of the positions of all the monitoring targets (6) to the data receiving platform; Step 6: setting the cumulative displacement safety threshold of the monitoring target (6) on the data receiving platform, numbering the safety alarm (8), and fixing the safety alarm (8) on the corresponding cross-section tunnel wall; Step 7: When the deformation data of the monitoring target position received by the data receiving platform exceeds the monitoring target cumulative displacement safety threshold, the data receiving platform sends the abnormal deformation section and the monitoring position to the corresponding safety alarm (8) and issues an alarm command. When the safety alarm (8) receives the alarm command, the light turns on and the abnormal deformation position is broadcast.

4. The method for monitoring tunnel construction deformation based on machine vision according to claim 3, characterized in that: In step 1 and step 2, all monitoring targets (6) are installed with the steel bar embedded parts by the following method: Step 101, aligning the notch (711) of the limiting cone tube (71) with the steel bar embedded part, pushing the limiting cone tube (71) so that the steel bar embedded part is embedded in the limiting cone tube (71), thereby achieving connection between the limiting cone tube (71) and the steel bar embedded part; Step 102, setting the monitoring target (6) on the surface of the target mounting shell (72), passing the target mounting shell (72) through the conical opening (721) from the bottom of the steel bar embedded part and sleeved on the steel bar embedded part, and making the inner surface of the conical opening (721) engage with the outer surface of the limiting conical tube (71); Step 103, sleeve the fastening nut (73) on the bottom of the steel bar embedded part, screw the fastening nut (73) so that the fastening nut (73) moves upward along the steel bar embedded part, squeezes the target installation shell (72), and completes the installation.

5. The method for monitoring tunnel construction deformation based on machine vision according to claim 3, characterized in that: In step 5, the deformation data of the monitoring target position is calculated by the following method: Step 501: the machine vision monitoring camera (1) captures the monitoring target (6) No. m on the section Sn at the i-th moment, captures an image and uploads the section information, number information, coordinate value and shooting time of the monitoring target (6) No. m; Step 502: after a photographing interval, the machine vision monitoring camera (1) captures the monitoring target (6) No. m on the section Sn at the i+1th time, takes an image and uploads the section information, number information, coordinate value and photographing time of the monitoring target (6) No. m; Step 503, establish a corresponding relationship between the monitoring target (6) with the same number in the same section in the image at the i-th moment and the i+1-th moment, calculate the relative difference in the coordinates of the monitoring target (6) with number m captured at the i-th moment and the i+1-th moment and upload it, the relative difference being the position deformation data of the monitored position at the i+1-th moment compared to the monitoring target (6) with number m at the i-th moment.

Images