WO2024083272A1

WO2024083272A1 - Layered soil settlement monitoring system and method based on machine vision

Info

Publication number: WO2024083272A1
Application number: PCT/CN2023/142392
Authority: WO
Inventors: 汪珂; 王立新; 令宜凡; 李储军; 喻忠
Original assignee: 中铁第一勘察设计院集团有限公司
Priority date: 2023-06-06
Filing date: 2023-12-27
Publication date: 2024-04-25
Also published as: CN116380016A; CN116380016B

Abstract

A layered soil (7) settlement monitoring system and method based on machine vision. The method comprises: arranging, at intervals, multiple layers of target media on the cross section of soil (7) to be subjected to monitoring; repeatedly collecting position information of the multiple layers of target media by means of a visual sensor (4); performing comparison to obtain position change parameters of each target medium at different moments; and performing calculation to obtain a relative layered settlement amount of each layer of the soil (7). The monitoring system comprises a measurement sleeve (1), a sliding rail system (2), a target medium injection system (3), a visual sensor (4), a power source module (5), a stabilized-pressure engagement and fixation device (8), etc. The visualized monitoring of a layered settlement condition of soil (7) is realized by means of a single measurement drill hole (11), thereby improving the monitoring accuracy and efficiency; and the monitoring system has relatively strong universality and adaptability, and is thus applicable to layered settlement monitoring of different types of soil (7).

Description

Soil layer settlement monitoring system and method based on machine vision

This application claims the priority of the Chinese patent application filed with the China Patent Office on June 6, 2023, with application number 202310659430.3, and application name “Soil stratification and settlement monitoring system and method based on machine vision”, all contents of which are incorporated by reference in this application.

Technical Field

The present application relates to the field of underground engineering technology detection, and specifically to a soil stratification settlement monitoring system and method based on machine vision.

Background technique

Soil stratification settlement refers to the different degrees of sinking movement of soil layers at different depths due to uneven force or external interference, which leads to changes in soil structure and properties. Soil stratification settlement is a common geological disaster that can cause serious damage and impact on underground projects, buildings, roads, bridges, etc., and even threaten human life and property. Therefore, in underground projects, it is often necessary to monitor the stratified settlement of the soil and grasp the settlement of soil layers at different depths and in different periods to predict the settlement trend after the completion of the project and determine the stability of the project.

At present, the deep mark point leveling method and electromagnetic settlement instrument method are mainly used for soil layer settlement monitoring. These methods all require the establishment of multiple measuring holes of different depths in the monitoring area for measurement. Specifically, the deep mark point leveling method is to use a drilling rig to drill a hole at a predetermined location, then put a measuring rod with a settlement plate into the hole, and use a casing to protect the measuring rod to lead the measuring rod out of the ground, and then manually observe it using the leveling height measurement method. The disadvantages of this method are: (1) Often only one deep mark point can be arranged in a borehole, which is only applicable to situations with fewer measuring points; (2) The measurement process is greatly affected by weather conditions, the monitoring data cannot be obtained and processed in real time, and abnormal situations cannot be discovered and warned in time, and the accuracy of the measurement results is limited; (3) When the distance between the monitoring point and the benchmark point is far, long-distance measurement is required, and the measurement process is time-consuming and labor-intensive; (4) The monitoring equipment is easily damaged or interfered with, and the maintenance cost is high.

The electromagnetic settlement instrument method is to bury the settlement tube in the soil by vertical drilling, set the settlement magnetic ring in the axial direction of the settlement tube according to the layered measurement interval, and the settlement magnetic ring has a spring claw on the outside that extends into the hole wall soil. The settlement magnetic ring settles with the hole wall soil, and the electromagnetic probe is used to measure the initial position and the position after settlement of the magnetic ring. The disadvantages of this method are: (1) Due to the weak anchoring force of the magnetic ring spring claw, it is difficult to tightly grasp the borehole wall. In addition, the gap between the settlement tube and the settlement magnetic ring is easily filled with soil, which will produce a large resistance to the settlement magnetic ring, often resulting in the settlement magnetic ring being difficult to settle synchronously with the soil layer; (2) There is a large human error. The measurement process requires regular calibration of the settlement tube mouth elevation, which requires a high level of operator skills; (3) It is impossible to observe the soil layer settlement in real time.

Summary of the invention

The present application proposes a soil stratification settlement monitoring system and method based on machine vision to solve the problems of high cost, low automation and low measurement accuracy of existing soil stratification settlement monitoring.

In order to achieve the above purpose, the technical solution adopted in this application is:

On the one hand, the present application provides a soil stratification settlement monitoring method based on machine vision, in which multiple layers of target media are set at intervals in the soil section to be monitored, and the position information of the multiple layers of target media is repeatedly collected through visual sensors. The position change parameters of each target medium at different times are compared to obtain the relative stratification settlement of each soil layer.

Further, the method comprises:

Along the upper end surface of the soil body to be monitored, drilling measurement holes downwards into the soil body to be monitored;

Fluorescent marker liquid is injected longitudinally at intervals along the inner wall of the measuring borehole as a fluorescent position target;

The visual sensor traverses all fluorescent position targets and collects the height information of each fluorescent position target as the first range measurement data;

At preset time intervals, the visual sensor repeatedly collects the height information of the target at each position to obtain multiple range measurement data;

The data from multiple measurement ranges are compared with the data from the first measurement range to calculate the relative stratified settlement of each soil layer at different time periods.

Furthermore, the method further comprises:

An overall settlement measuring device is set on the upper end surface of the soil to be monitored to measure the overall settlement of the monitored soil at different time periods;

The overall settlement and relative stratified settlement are combined to obtain the relative stratified true settlement of each soil layer at different time periods.

Furthermore, the visual sensor traverses all position targets and collects the height information of each position target, including:

The visual sensor directly obtains the height information of the target at each position through the ruler;

or,

The height information of the target at each position is obtained by calculating the movement speed, acceleration and movement time of the visual sensor.

On the other hand, the present application provides a soil stratification settlement monitoring system based on machine vision, comprising:

A measuring sleeve, which is a transparent tube, is arranged in the measuring borehole, and a scale and a plurality of injection holes are arranged along the longitudinal direction of the transparent tube;

A slide rail system, wherein the slide rail system is arranged in the measuring sleeve, and comprises a measuring slide rail and a self-driving slider body arranged on the measuring slide rail, wherein the self-driving slider body is connected to a power module;

A target medium injection system, wherein the target medium injection system is arranged on the self-driving slider body and can move up and down along the measuring slide rail with the self-driving slider body;

The visual sensor is arranged on the self-driving slider body and can move up and down along the measuring slide rail along with the self-driving slider body.

Furthermore, a surface calibration settlement system is arranged on the upper end face of the measuring borehole, the surface calibration settlement system comprises a surface calibration sealing cover, a surface detection target is arranged on the upper end of the surface calibration sealing cover, and a machine vision instrument is arranged outside the surface calibration sealing cover.

Furthermore, the cross section of the measuring sleeve is elliptical, the measuring sleeve is closely attached to the inner wall of one side of the measuring borehole, the inner wall of the other side of the measuring borehole and the measuring sleeve form a gap, and a pressure stabilizing locking device is arranged in the gap.

Furthermore, the voltage-stabilizing locking device includes a pressure-stabilizing air bag, the pressure-stabilizing air bag is connected to a pressure-stabilizing air pump through an air pipe and a pressure gauge, and the pressure-stabilizing air pump is connected to a power module.

Furthermore, the voltage-stabilizing locking device includes a pressure-stabilizing liquid capsule, the pressure-stabilizing liquid capsule is connected to a pressure-stabilizing liquid pump through a liquid pipe and a pressure gauge, and the pressure-stabilizing air pump is connected to a power module.

Furthermore, the measuring slide rail comprises a rail body, a slide block slot and a power transmission slot are provided in the extending direction of the rail body, and a baffle is provided at the upper end of the rail body.

Furthermore, the self-driving slider body is inserted into the rail body, and a servo motor and a driving wheel connected to the servo motor are arranged inside the self-driving slider; the self-driving slider body is electrically connected to the power transmission slot through its power contact, and the power transmission slot is connected to the power module.

Furthermore, an illumination light source is arranged on the self-driving slider body.

Furthermore, an acceleration sensor or a high-precision pedometer is arranged on the main body of the self-driving slider.

Furthermore, the target medium injection system includes a liquid storage capsule, which is connected to a booster pump, which is connected to a one-way injection needle through a liquid tube, and the one-way injection needle is provided with an injection needle retractable mechanism, and the booster pump and the injection needle retractable mechanism are both connected to a power module.

Furthermore, a one-way check valve is provided at the orifice of the injection hole, and an injection needle retaining groove is provided on the inner wall of the measuring sleeve where the injection hole is located.

Furthermore, the visual sensor includes a plurality of high-definition cameras, which are arranged on the outer end surface of the self-driving slider body and connected to the power module, and are connected to the image processing terminal via wired or wireless transmission.

Compared with the prior art, the beneficial effects of this application are as follows:

1. An embodiment of the present application discloses a soil layer settlement monitoring system based on machine vision. The system, based on a single measurement borehole based on machine vision, can inject inert fluorescent marker liquid into soil at different layers, collect image data of fluorescent position targets, directly obtain the height information of each position target through its own scale, and the image processing terminal calculates the relative layer settlement of each soil layer at different time periods. The system can realize visual monitoring of soil layer settlement, improve the accuracy and efficiency of monitoring, and has strong versatility and adaptability, and can be applied to different types of soil and target media.

2. An embodiment of the present application discloses a soil layer settlement monitoring system based on machine vision. By using a slide rail system, the target medium injection system and the visual sensor system can be freely moved up and down inside the measuring casing, thereby conveniently Injecting target media into soil at different levels and collecting image data of targets at each position improves the efficiency and flexibility of monitoring.

3. An embodiment of the present application discloses a soil stratification settlement monitoring system based on machine vision, which utilizes the cooperation of an elliptical measuring sleeve and a stable locking device to achieve firm compression of the measuring sleeve in the measuring borehole, prevent the measuring device from moving or rotating in the measuring borehole, and ensure the stability and accuracy of monitoring.

4. An embodiment of the present application discloses a soil stratification settlement monitoring method based on machine vision. By directly measuring the displacement changes of the target medium at different time periods with a ruler, the relative stratification settlement of each soil layer at different time periods can be obtained, thereby reflecting the changing trend and law of soil stratification settlement, and providing a basis for the prediction and prevention of soil settlement.

5. An embodiment of the present application discloses a soil stratification settlement monitoring method based on machine vision. When the scale is covered or damaged for some reason, the motion parameters of the visual sensor can be obtained through an acceleration sensor and a high-precision pedometer, and the missing relative stratification settlement amount can be calculated to ensure the continuity and integrity of the monitoring. At the same time, under normal measurement conditions, the relative stratification settlement amount calculated by the motion parameters can be compared with the relative stratification settlement amount directly measured by the scale to verify the correctness and accuracy of the monitoring.

Of course, the implementation of the various technical solutions of the present application does not necessarily require achieving all of the advantages described above at the same time.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying any creative work.

FIG1 is a block diagram of the system composition of Example 3 of the present application.

FIG. 2 is a schematic diagram of the overall structure of the monitoring system of Example 3 of the present application.

FIG. 3 is a schematic cross-sectional view of FIG. 2 .

FIG4 is a schematic diagram of the structure of Example 3 of the present application after the injection is completed and the measuring sleeve is rotated 180 degrees.

FIG. 5 is a schematic diagram showing the principle of calculating the relative stratified settlement amount when the scale is deformed in Example 5. FIG.

FIG. 6 is a schematic diagram showing the principle of calculating the relative stratified settlement amount when the scale is covered or defective in Example 5. FIG.

In the figure,

1- measuring sleeve, 101- injection hole, 102- ruler;

2-slide rail system, 201-measuring slide rail, 2011-rail body, 2012-baffle, 202-self-driving slide body;

3-target medium injection system, 301-one-way injection needle;

4- visual sensor, 401- high-definition camera, 402- image processing terminal;

5- Power module;

6-fluorescent position target;

7- Soil;

8-pressure stabilizing locking device, 801-pressure stabilizing air bag, 802-trachea, 803-pressure gauge, 804-pressure stabilizing air pump;

9-surface calibration settlement system, 901-surface calibration sealing cover, 902-surface detection target, 903-machine vision instrument;

10- Acceleration sensor;

11-Measure the drill hole.

Detailed ways

The present application is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are intended to enable the present application to be better understood. However, those skilled in the art can easily recognize that some of the features can be omitted in different situations, or can be replaced by other elements, materials, and methods. In some cases, some operations related to the present application are not shown or described in the specification, in order to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

The serial numbers assigned to the components herein, such as "S1", "S2", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in this application, unless otherwise specified, include direct and indirect connections (couplings).

Embodiment 1:

The present embodiment relates to a soil layer settlement monitoring system based on machine vision, which is arranged in a measuring borehole and includes: a measuring casing, a slide rail system, a target medium injection system, a visual sensor and a power module.

The measuring sleeve in this embodiment is an elliptical tube made of transparent material, specifically PVC transparent material; after the transparent measuring sleeve is set in the measuring borehole, the visual sensor set in the tube can clearly observe the fluorescent position target in the soil. The purpose of the elliptical structural design of the measuring sleeve is: first, to enable the measuring sleeve to adapt to and be fixed in measuring boreholes of different calibers through a voltage-stabilizing and fixing device; second, to facilitate the rotation and fixing of the measuring sleeve in the measuring borehole, thereby realizing different functions. Specifically, because the elliptical measuring sleeve is close to the inner wall of one side of the measuring borehole, the other side of the measuring borehole A gap is formed between the inner wall and the measuring sleeve, and a pressure stabilizing fixture is arranged in the gap. The pressure stabilizing fixture includes a plurality of pressure stabilizing liquid capsules, which are connected to a pressure stabilizing liquid pump through a liquid pipe and a pressure gauge, and the pressure stabilizing liquid pump is connected to a power module. The output of the pressure stabilizing liquid pump is adjusted to expand and tighten the pressure stabilizing liquid capsule in the gap, and to apply appropriate pressure to the outer wall of the measuring sleeve to ensure that the measuring sleeve is not easy to move.

A ruler is provided on the measuring sleeve along the longitudinal direction of the tube, and the scale of the ruler is used for visual sensors to intuitively measure the height of the fluorescent position target. Several injection holes are opened on the measuring sleeve at a certain interval along the longitudinal direction of the tube. The injection holes are directly opposite to the ruler and are 180 degrees apart. A one-way check valve is installed in each injection hole. The injection hole serves as a channel for the target medium injection system to inject into the soil to form the fluorescent position target. The one-way check valve can prevent the liquid of the fluorescent target from flowing back or leaking. An injection needle clamping groove is provided on the inner wall of the measuring sleeve and each injection hole. The injection needle clamping groove can fix and guide the one-way injection needle to ensure that the one-way injection needle is aligned with the injection hole and inserted, and can reduce the friction between the injection needle and the measuring sleeve.

The slide rail system is arranged inside the measuring casing, including a measuring slide rail and a self-driving slider body arranged on the measuring slide rail. The measuring slide rail includes a rail body, the upper end of the rail body is connected to a baffle, a power module and a pressure-stabilizing air pump are arranged on the upper end of the baffle, and a card slot and a power transmission slot are arranged in the extension direction of the rail body. The self-driving slider body is sleeved on the rail body and can move up and down along it to complete the target medium injection and image data acquisition of different layers of soil. The card slot is used to locate the position of the self-driving slider body to prevent it from rotating or deviating from the track; the power transmission slot is connected to the power module to provide power for the self-driving slider body, the target medium injection system carried by the self-driving slider body, and the visual sensor. The baffle is used to fix the position of the rail body in the measuring casing, and the baffle can also limit the upward movement range of the self-driving slider body to prevent it from leaving the rail body.

In this embodiment, the self-driving slider body includes a rectangular shell, in which a motor and a driving ball are arranged. The servo motor drives the driving wheel to rotate, thereby controlling the precise movement speed and direction of the self-driving slider body on the rail body. The self-driving slider body is electrically connected to the power transmission slot of the rail body through its power contacts.

The target medium injection system is arranged on the main body of the self-driving slider and can move synchronously with the main body of the self-driving slider. The target medium injection system is electrically connected to the power module through the power transmission slot. The target medium injection system includes a liquid storage capsule, which is connected to a booster pump, and the booster pump is connected to a one-way injection needle through a liquid pipe. The one-way injection needle is provided with an injection needle telescopic mechanism. In this embodiment, the injection needle telescopic mechanism uses a telescopic gear, which can control the telescopic movement of the one-way injection needle so that it can extend or retract from the injection hole of the measuring sleeve, and then penetrate the fluorescent target liquid injected into the soil. The liquid storage capsule is connected to the booster pump to provide sufficient hydraulic pressure for the one-way injection needle, so that the fluorescent target liquid can penetrate the soil and gather to form a fluorescent position target; in this embodiment, the liquid storage capsule stores an inert fluorescent target liquid, specifically a dispersion of phosphorescent pigments, and the inert fluorescent target liquid is a liquid that is not easy to react chemically with the soil or biodegrade, has a strong fluorescence effect, and can emit bright colors under the illumination of the illumination light source, which is convenient for image recognition of the visual sensor. Through the design of the target medium injection system of this embodiment, clearly visible fluorescent position targets can be formed in soil bodies of different layers, while reducing the interference and pollution of the soil body by the monitoring activities themselves.

In this embodiment, the visual sensor includes three high-definition cameras, one on the main body of the self-driving slider where the one-way injection needle is located. A high-definition camera is respectively arranged on the upper and lower sides of the body side and the one-way injection needle, and another high-definition camera is arranged in the center of the opposite side of its shell. The three high-definition cameras are connected to the power module 5 through the power transmission slot. The three high-definition cameras are connected to the image processing terminal. In this embodiment, the image processing terminal is a cloud server that can calculate image data. The three high-definition cameras communicate with the cloud server through wireless transmission. The two high-definition cameras above and below the one-way injection needle are used to determine the position of the injection hole, so that the one-way injection needle can accurately extend out of the injection hole; the other high-definition camera is used to collect image data of the fluorescent position target.

The monitoring system of this embodiment can inject inert fluorescent marker liquid into different layers of soil and collect image data of fluorescent position targets. The height information of each fluorescent position target can be directly obtained through its own scale. The image processing terminal calculates the relative layered settlement of each soil layer at different time periods.

Embodiment 2:

This embodiment provides a soil stratification settlement monitoring system based on machine vision. On the basis of Embodiment 1, this embodiment additionally adds a motion state measurement system to obtain the height information of the fluorescent agent target through motion parameter calculation.

Specifically, the self-driving slider body is provided with an acceleration sensor and a high-precision pedometer. The acceleration sensor is used to measure the acceleration of the self-driving slider body so as to calculate its moving distance and speed; the high-precision pedometer is used to record the position and moving distance of the self-driving slider body. The parameters such as the movement speed, acceleration and movement time of the self-driving slider body (i.e., the visual sensor) are obtained.

In addition, in this embodiment, an illumination light source is also provided on the main body of the self-driving slider, and the illumination light source is used to excite the fluorescent marked liquid and improve its visibility, so as to facilitate the image acquisition of the fluorescent position target by the visual sensor.

The monitoring system of this embodiment obtains the motion parameters of the self-driving slider body, and combines the image data of the fluorescent position target collected by the visual sensor to provide another algorithm calculation method for parameters in addition to directly obtaining the height parameters of the fluorescent position target by the ruler; the two methods can verify each other. In addition, in special circumstances, such as when the ruler scale is not clearly photographed or is blocked, the height parameters of the fluorescent position target can also be clearly obtained through algorithm calculation, and finally the relative stratified settlement of each soil layer can be obtained.

Embodiment 3:

Referring to Figure 1, this embodiment relates to a soil stratification settlement monitoring system based on machine vision, which is arranged in a measuring borehole and includes: a measuring casing 1, a slide rail system 2, a target medium injection system 3, a visual sensor 4, a power module 5, a motion state measurement system and a surface calibration settlement system 9.

2 to 4 , FIG. 2 is a schematic diagram of the overall structure of the monitoring system; FIG. 3 is a schematic diagram of the cross-section of FIG. 2 ; and FIG. 4 is a schematic diagram of the structure after the measuring sleeve is rotated 180 degrees after the injection is completed.

The measuring sleeve 1 in this embodiment is an oval tube made of transparent material, such as PET transparent material; after the transparent measuring sleeve 1 is set in the measuring borehole 11, the visual sensor 4 set in the tube can clearly observe the situation of the fluorescent position target 6 in the soil. The oval structure design of the measuring sleeve 1 aims to: first, enable the measuring sleeve 1 to pass through the voltage stabilizing card The fixing device 8 is adapted to and fixed in the measuring boreholes 11 of different calibers; secondly, it is convenient for the measuring sleeve 1 to rotate and fix in the measuring borehole 11, so as to realize different functions. Specifically, since the elliptical measuring sleeve 1 is close to the inner wall of one side of the measuring borehole 11, a gap is formed between the inner wall of the other side of the measuring borehole 11 and the measuring sleeve, and a pressure-stabilizing fixing device 8 is arranged in the gap. The pressure-stabilizing fixing device 8 includes a plurality of pressure-stabilizing airbags 801, and the pressure-stabilizing airbags 801 are connected to the pressure-stabilizing air pump 804 through the air pipe 802 and the pressure gauge 803, and the pressure-stabilizing air pump 804 is connected to the power module 5. The output of the pressure-stabilizing air pump 804 is adjusted to make the pressure-stabilizing airbag 801 expand and press tightly in the gap, and apply appropriate pressure to the outer wall of the measuring sleeve 1 to ensure that the measuring sleeve 1 is not easy to move.

A scale 102 is provided on the measuring sleeve 1 along the longitudinal direction of the tube, and the scale 102 is used for the visual sensor to directly measure the height of the fluorescent position target 6. A number of injection holes 101 are provided on the measuring sleeve 1 at a certain interval along the longitudinal direction of the tube, and the injection holes 101 are directly opposite to the scale 102 and are 180 degrees apart. A one-way check valve is installed in each injection hole 101. The injection hole 101 is used as a target medium injection system to inject into the soil to form a channel for the fluorescent position target 6. The one-way check valve can prevent the liquid of the fluorescent mark from flowing back or leaking. An injection needle clamping groove is provided on the inner wall of the measuring sleeve 1 and each injection hole 101. The injection needle clamping groove can fix and guide the one-way injection needle 301, ensure that the one-way injection needle 301 is aligned and inserted with the injection hole 101, and can reduce the friction between the injection needle and the measuring sleeve.

The slide rail system 2 is arranged inside the measuring sleeve 1, and includes a measuring slide rail 201 and a self-driving slider body 202 arranged on the measuring slide rail 2. The measuring slide rail 201 includes a rail body 2011, the upper end of the rail body 2011 is connected to a baffle 2012, a power module 5 and a pressure-stabilizing air pump 804 are arranged on the upper end of the baffle 2012, and a card slot and a power transmission slot are arranged in the extension direction of the rail body 2011. The self-driving slider body 202 is sleeved on the rail body 2011 and can move up and down along it to complete the target medium injection and image data acquisition at different soil positions. The card slot is used to locate the position of the self-driving slider body 202 to prevent it from rotating or deviating from the track; the power transmission slot is connected to the power module 5 to provide power for the self-driving slider body 202, the target medium injection system 3 and the visual sensor 4 carried by the self-driving slider body 202. The baffle 2012 is used to fix the position of the rail body 2011 in the measuring sleeve 1 . The baffle 2012 can also limit the upward movement range of the self-driving slider body 202 to prevent it from separating from the rail body 2011 .

In this embodiment, the self-driving slider body 202 includes a rectangular shell, in which a servo motor and a driving wheel are arranged. The servo motor drives the driving wheel to rotate, thereby controlling the precise movement speed and direction of the self-driving slider body 202 on the rail body 2011. The self-driving slider body 202 is electrically connected to the power transmission slot of the rail body 2011 through its power contact. In other embodiments, the servo motor and the driving wheel can also be replaced by other driving systems, such as a stepping motor and a driving ball, which can complete the precise movement position and direction control. An illumination light source is also arranged on the self-driving slider body 202. The illumination light source is used to excite the fluorescent target liquid and improve its visibility, so as to facilitate the image acquisition of the fluorescent position target 6 by the visual sensor 4.

The target medium injection system 3 is arranged on the self-driving slider body 202 and can move synchronously with the self-driving slider body 202. The target medium injection system 3 is electrically connected to the power module 5 through the power transmission slot. The target medium injection system 3 includes a liquid storage capsule, which is connected to a booster pump, and the booster pump is connected to a one-way injection needle 301 through a liquid pipe. The one-way injection needle 301 is provided with an injection needle telescopic mechanism. In this embodiment, the injection needle telescopic mechanism uses a telescopic gear, which can control the extension of the one-way injection needle 301. The liquid reservoir is connected to the booster pump to provide sufficient hydraulic pressure for the one-way injection needle 301, so that the fluorescent liquid can penetrate the soil and gather to form a fluorescent position target 6; in this embodiment, the liquid reservoir stores an inert fluorescent liquid, which can be a dispersion of an inert pigment (such as a fluorescent pigment, a phosphorescent pigment, etc.) or a nanoparticle solution of an inert metal. The inert fluorescent liquid is a liquid that is not easy to react chemically or biodegrade with the soil, has a strong fluorescence effect, and can emit bright colors under the illumination of an illumination light source, which is convenient for image recognition by a visual sensor. Through the design of the target medium injection system of this embodiment, a clearly visible fluorescent position target 6 can be formed in soils of different layers, while reducing the interference and pollution of the monitoring activity itself to the soil 7.

In this embodiment, the visual sensor 4 includes three high-definition cameras 401. One high-definition camera 401 is respectively arranged on the shell side of the self-driving slider body 202 where the one-way injection needle 301 is located, and above and below the one-way injection needle 301. Another high-definition camera 401 is arranged in the center of the opposite side of its shell. The three high-definition cameras 401 are connected to the power module 5 through the power transmission slot. The three high-definition cameras 401 are connected to the image processing terminal 402. In this embodiment, the image processing terminal 402 is a cloud server that can calculate image data. The three high-definition cameras 401 communicate with the cloud server through wireless transmission. The two high-definition cameras 401 above and below the one-way injection needle 301 are used to determine the position of the injection hole 101, so that the one-way injection needle 301 can accurately extend out of the injection hole 101; the other high-definition camera 401 is used to collect image data of the fluorescent position target 6.

The motion state measurement system includes an acceleration sensor 10 and a high-precision pedometer arranged on the self-driving slider body 202. The acceleration sensor 9 is used to measure the acceleration of the self-driving slider body 202 so as to calculate its moving distance and speed; the high-precision pedometer is used to record the position and moving distance of the self-driving slider body 202. The parameters such as the motion speed, acceleration and motion time of the self-driving slider body 202 (i.e., the visual sensor 4) are obtained.

The monitoring system of this embodiment can inject inert fluorescent marker liquid into different layers of soil, collect image data of fluorescent position targets, and directly obtain the height information of each fluorescent position target through its own ruler. The image processing terminal calculates the relative stratified settlement of each soil layer at different time periods; in addition, the motion state measurement system obtains the motion parameters of the self-driving slider body, combined with the image data of the fluorescent position target collected by the visual sensor, to provide another algorithm calculation parameter method in addition to directly obtaining the fluorescent position target height parameters by the ruler; the two methods can verify each other. In addition, in special circumstances, such as when the ruler scale is not clearly photographed or is blocked, the fluorescent position target height parameters can also be clearly obtained through algorithm calculation, and finally the relative stratified settlement of each soil layer is obtained.

In addition, this embodiment also includes a surface calibration settlement system 9, which is arranged on the upper end surface of the measuring borehole 11, including a surface calibration sealing cover 901, a surface detection target 902 is arranged on the upper end of the surface calibration sealing cover 901, and a machine vision instrument 903 is arranged outside the surface calibration sealing cover 901.

Surface calibration settlement system 9 The target points on the surface calibration sealing cover are measured to determine the overall settlement of the top of the soil. The overall settlement obtained is combined with the relative stratified settlement to obtain the relative stratified settlement of each soil layer at different time periods. The actual amount of sedimentation.

Embodiment 4:

This embodiment provides a soil layer settlement monitoring method based on machine vision, and the detection method is as follows:

S1. On the cross section of the soil to be monitored, multiple layers of position targets are set at certain intervals. The position targets are a material that can be recognized by visual sensors;

S2, collecting the position information of each position target through the visual sensor, and transmitting the position information to the image processing terminal;

S3, the image processing terminal calculates the height information of the target at each position according to the position information, and stores the height information as a single range data;

S4, repeating steps S2 and S3 at preset time intervals to obtain a plurality of range measurement data;

S5. Compare the data from multiple measurement ranges and calculate the relative stratified settlement of each soil layer at different time periods.

It should be noted that the position target can be any material that can be recognized by a visual sensor, for example, it can be a liquid, powder or granule with a fluorescent effect, or a metal sheet, paper or plastic sheet with a reflective effect, or an object with a specific color or shape.

The method of setting the position target can be determined according to the properties of the soil and the shape of the target medium. For example, if the soil is relatively loose, a liquid or powdered target medium can be injected into the soil through an injection needle to form a position target; if the soil is relatively hard, a solid target medium can be inserted into the soil by drilling or cutting to form a position target; if the soil is relatively flat, a material with a specific color can be directly pasted or painted on the soil surface as a position target.

This embodiment sets multiple layers of position targets on the cross section of the soil to be monitored, and repeatedly collects the position information of the multiple layers of position targets through visual sensors, compares the position change parameters of each position target at different times, and calculates the relative layered settlement of each soil layer. This method can realize real-time monitoring of soil layered settlement, improves the accuracy and efficiency of monitoring, has strong versatility and adaptability, and can be applied to different types of soil and target media.

Embodiment 5:

This embodiment provides a soil layer settlement monitoring method based on machine vision. The detection method is based on the detection device of Example 3, and the specific steps are as follows:

S1. Use a drilling rig to drill a measuring hole on the soil to be monitored, assemble the measuring casing according to the joints, and then place it into the hole close to one side of the measuring hole.

S2. Place the pressure-stabilizing airbag of the pressure-stabilizing locking device along the other side of the measuring borehole, turn on the pressure-stabilizing air pump, and allow the pressure-stabilizing airbag to expand and tighten in the gap, applying appropriate pressure to the outer wall of the measuring casing to ensure that the measuring casing does not move.

S3. Fill the reservoir of the target medium injection system with inert fluorescent marker liquid.

S4. Lower the slide rail system into the measuring casing.

S5. The self-driving slider body drives the target medium injection system to inject the fluorescent target liquid into the soil layer from top to bottom or from bottom to top. Specifically, the high-definition camera detects that when the self-driving slider body reaches the height of each injection hole of the measuring sleeve, the one-way injection needle extends from the injection hole to inject a certain amount of fluorescent mark liquid into the soil layer; the fluorescent mark liquid at all the injection hole positions is injected, forming a fluorescent position target for the entire soil body.

S6. The one-way injection needle is fully retracted into the self-driving slider body, and the illumination light source is turned on. The illumination light source illuminates the fluorescent position target at a uniform speed from bottom to top or from top to bottom to excite the fluorescent position target and improve its visibility.

S7. The pressure-stabilizing locking device turns off the pressure-stabilizing air pump, deflates the pressure-stabilizing airbag, and rotates the measuring sleeve together with the slide rail system by 180 degrees. At this time, the scale of the measuring sleeve is aligned with the fluorescent position target.

S8. Turn on the pressure-stabilizing air pump of the pressure-stabilizing fixing device to apply appropriate pressure to the outer wall of the measuring sleeve to fix the measuring sleeve so that it does not move.

S9. Set up a surface calibration settlement system at the upper end of the measuring borehole.

S10. The high-definition camera of the visual sensor is turned on, and the self-driving slider body drives the high-definition camera to move a measuring range from top to bottom (or from bottom to top). The high-definition camera shoots the corresponding positions of the fluorescent position targets and the scale of each soil layer through the transparent pipe wall and transmits them to the image processing terminal, which is stored as the initial measurement image and calculates the initial measuring range data (initial fluorescent position target height parameters).

S11. At preset time intervals, the visual sensor repeatedly collects the corresponding positions of each fluorescent position target and the scale and transmits them to the image processing terminal to obtain multiple subsequent measurement images, namely subsequent range measurement data (subsequent fluorescent position target height parameters).

S12. Compare the multiple subsequent measurement data with the initial measurement data respectively. Specifically, the image processing terminal compares the subsequent measurement images with the initial measurement images one by one, calculates the vertical position difference of the corresponding fluorescent position target, and calculates the relative layer settlement Δd of each soil layer in different time periods.

S13. At the same time as steps S10 and S11, the surface calibration settlement system synchronously collects the overall settlement ΔD of the soil at different time periods.

S14. The image processing terminal combines the relative stratified settlement Δd of the soil layer with the overall settlement ΔD of the soil body, and finally calculates the relative stratified true settlement D of each soil layer at different time periods, where D=Δd+ΔD.

When the scale of the measurement image is clear and definite, the relative layered settlement Δd of each soil layer at different time periods can be accurately obtained through the above S1-S12.

If the image captured by the machine vision instrument is deformed, the real relative stratification settlement amount Δd can be obtained by converting the image pixels, see Figure 5. The specific calculation method is as follows:

Assuming that x pixels represent a scale unit z, the scale unit of the image measured by the machine vision instrument is x’ pixels, and the actual relative stratification sedimentation amount measured by the machine vision instrument is y’ pixels, then the y pixels represented by Δd can be calculated through y/y’=x/x’, y=x/x’×y.

Therefore, Δd = (x/x’*y) × z.

If the scale is covered or damaged for some reason, the Δd at that position cannot be directly obtained. This embodiment provides another solution. A solution method is proposed. This method needs to obtain the motion speed, acceleration and motion time parameters of the visual sensor through the acceleration sensor installed on the main body of the self-driving slider, and calculate the missing relative stratified settlement amount Δd, see Figure 6. The specific calculation method is as follows:

d＝v ₀ Δt+(1/2)at ²

Where d is the position of the target, _v0 is the starting velocity, Δt is the time interval, a is the acceleration, and t is the time.

Starting from the initial moment, v ₀ = 0;

Initial position target position:

d ₀ ＝(1/2)a ₁ t ₁₀ ² +(1/2)a ₂ t ₂₀ ² +(1/2)a ₃ t ₃₀ ² +......+(1/2)a _n t _n0 ² ;

Target position after settlement:

d _t =(1/2)a ₁ t _1t ² +(1/2)a ₂ t _2t ² +(1/2)a ₃ t _3t ² +......+(1/2)a _n t _nt ² ;

Then the relative stratified settlement is: Δd = d _t - d ₀ ;

As shown in Figure 6, a _n is obtained by analyzing and intercepting the computer statistical acceleration and time curve. When the time interval is small enough, a _n is approximately unchanged.

Of course, the above algorithm is also used to verify the correctness of the relative stratified settlement Δd directly measured by the ruler.

The above specific examples are used to illustrate the present application, which is only used to help understand the present application and is not intended to limit the present application. For technicians in the technical field to which the present application belongs, they can also make some simple deductions, deformations or substitutions based on the ideas of the present application.

Claims

The soil stratification settlement monitoring method based on machine vision is characterized in that multiple layers of target media are set at intervals in the soil section to be monitored, the position information of the multiple layers of target media is repeatedly collected by visual sensors, the position change parameters of each target medium at different times are obtained by comparison, and the relative stratification settlement of each soil layer is calculated.
The soil layer settlement monitoring method based on machine vision according to claim 1 is characterized in that the method comprises:

Along the upper end surface of the soil body to be monitored, drilling measurement holes downwards into the soil body to be monitored;

Fluorescent marker liquid is injected at intervals along the inner wall of the measuring borehole to serve as a fluorescent position target;

The visual sensor traverses all fluorescent position targets and collects the height information of each fluorescent position target as the first range measurement data;

At preset time intervals, the visual sensor repeatedly collects the height information of the target at each position to obtain multiple range measurement data;

The data from multiple measurement ranges are compared with the data from the first measurement range to calculate the relative stratified settlement of each soil layer at different time periods.
The soil layer settlement monitoring method based on machine vision according to claim 1 or 2 is characterized in that the method further comprises:

An overall settlement measuring device is set on the upper end surface of the soil to be monitored to measure the overall settlement of the monitored soil at different time periods;

The overall settlement and relative stratified settlement are combined to obtain the relative stratified true settlement of each soil layer at different time periods.
The soil layer settlement monitoring method based on machine vision according to claim 2 is characterized in that the visual sensor traverses all position targets and collects height information of each position target, including:

The visual sensor directly obtains the height information of the target at each position through the ruler;

or,

The height information of the target at each position is obtained by calculating the movement speed, acceleration and movement time of the visual sensor.
The soil layer settlement monitoring system based on machine vision is characterized by comprising:

A measuring sleeve, which is a transparent tube, is arranged in the measuring borehole, and a scale and a plurality of injection holes are arranged along the longitudinal direction of the transparent tube;

A slide rail system, wherein the slide rail system is arranged in the measuring sleeve, and comprises a measuring slide rail and a self-driving slider body arranged on the measuring slide rail, wherein the self-driving slider body is connected to a power module;

A target medium injection system, wherein the target medium injection system is arranged on the self-driving slider body and can move up and down along the measuring slide rail with the self-driving slider body;

The visual sensor is arranged on the self-driving slider body and can move up and down along the measuring slide rail along with the self-driving slider body.
According to the soil stratification settlement monitoring system based on machine vision in claim 5, it is characterized in that a surface calibration settlement system is arranged on the upper end face of the measuring borehole, the surface calibration settlement system includes a surface calibration sealing cover, a surface detection target is arranged on the upper end of the surface calibration sealing cover, and a machine vision instrument is arranged outside the surface calibration sealing cover.
According to the machine vision-based soil stratification settlement monitoring system according to claim 5 or 6, it is characterized in that the cross-section of the measuring sleeve is elliptical, the measuring sleeve is tightly attached to the inner wall of one side of the measuring borehole, and the inner wall of the other side of the measuring borehole forms a gap with the measuring sleeve, and a voltage stabilizing locking device is arranged in the gap.
According to the machine vision-based soil stratification settlement monitoring system of claim 7, it is characterized in that the pressure-stabilizing locking device includes a pressure-stabilizing air bag, the pressure-stabilizing air bag is connected to a pressure-stabilizing air pump through an air pipe and a pressure gauge, and the pressure-stabilizing air pump is connected to a power module.
According to the machine vision-based soil stratification settlement monitoring system of claim 7, it is characterized in that the pressure stabilizing locking device includes a pressure stabilizing liquid bag, the pressure stabilizing liquid bag is connected to the pressure stabilizing liquid pump through a liquid pipe and a pressure gauge, and the pressure stabilizing air pump is connected to the power module.
According to the machine vision-based soil stratification settlement monitoring system of claim 8, it is characterized in that the measuring slide rail includes a rail body, a slider slot and a power transmission slot are provided in the extension direction of the rail body, and a baffle is provided at the upper end of the rail body.
According to the machine vision-based soil stratification settlement monitoring system of claim 10, it is characterized in that the self-driving slider body is passed through the rail body, and a servo motor and a driving wheel connected to the servo motor are arranged in the self-driving slider; the self-driving slider body is electrically connected to the power transmission trough through its power contact, and the power transmission trough is connected to the power module.
According to the machine vision-based soil stratification settlement monitoring system of claim 10, it is characterized in that an illumination light source is provided on the main body of the self-driving slider.
According to the machine vision-based soil stratification settlement monitoring system of claim 10, it is characterized in that an acceleration sensor or a high-precision pedometer is provided on the main body of the self-driving slider.
According to the machine vision-based soil stratification settlement monitoring system of claim 8, it is characterized in that the target medium injection system includes a liquid storage bag, the liquid storage bag is connected to the booster pump, the booster pump is connected to the one-way injection needle through a liquid pipe, the one-way injection needle is provided with an injection needle retractable mechanism, and the booster pump and the injection needle retractable mechanism are both connected to the power module.
According to the machine vision-based soil stratification settlement monitoring system of claim 8, it is characterized in that a one-way check valve is provided at the orifice of the injection hole, and an injection needle positioning groove is provided on the inner wall of the measuring sleeve where the injection hole is located.
According to the machine vision-based soil stratification settlement monitoring system of claim 8, it is characterized in that the visual sensor includes a plurality of high-definition cameras, the plurality of high-definition cameras are arranged on the outer end surface of the self-driving slider body and connected to the power module, and the plurality of high-definition cameras are connected to the image processing terminal via wired or wireless transmission.