An automatic monitoring method and system for subway tunnel deformation

Abstract

The present invention relates to the technical field of subway tunnel monitoring, and specifically discloses a method and system for automatically monitoring subway tunnel deformation. The method comprises creating a tunnel model based on building data and visual information; broadcasting a data acquisition instruction based on a preset frequency, and receiving tunnel data containing position and time acquired by a pre-installed rangefinder; when any rangefinder receives the data acquisition instruction, acquiring data based on an independent application probability; the application probability is obtained by the rangefinder after local processing of the monitoring data; updating and displaying the tunnel model based on the tunnel data containing position and time; and installing a local recognition module on the rangefinder in the present invention, which determines the data upload frequency and provides an intermittent data upload architecture, converting a real-time process into a timed process, thereby greatly reducing resource consumption while ensuring data validity as much as possible.

Description

Technical Field

[0001] The present invention relates to the technical field of subway tunnel monitoring, and in particular to an automatic monitoring method and system for subway tunnel deformation. Background Art

[0002] As a vital component of urban underground transportation systems, subway tunnels are crucial for their stability and safety. During subway construction and operation, tunnels may deform to varying degrees due to factors such as geological conditions, construction techniques, and environmental changes. To ensure the structural and operational safety of subway tunnels, real-time, accurate deformation monitoring is crucial. With advances in sensor technology, automated control techniques, and data analysis methods, subway tunnel deformation monitoring has evolved from traditional manual inspections to automated, intelligent monitoring systems.

[0003] Intelligent monitoring systems basically require rangefinders. There are many types of rangefinders, which can obtain distances with different precisions and then calculate deformation. The existing measurement process is generally completed by a unified computer device. The rangefinder needs to upload data in real time, and then the computer device performs statistical processing. In this process, the real-time uploading of data will occupy a large amount of data processing resources. How to reduce this consumption is the technical problem that the technical solution of the present invention aims to solve. Summary of the Invention

[0004] The object of the present invention is to provide a method and system for automatically monitoring deformation of a subway tunnel, so as to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method and system for automatically monitoring deformation of a subway tunnel, the method comprising:

[0007] Obtain architectural data of the subway tunnel, obtain visual information of the subway tunnel, and create a tunnel model based on the architectural data and visual information;

[0008] Based on a preset frequency broadcast data acquisition instruction, receive tunnel data containing location and time acquired by pre-installed distance meters; when any distance meter receives the data acquisition instruction, it acquires data based on independent application probability; the application probability is obtained by the distance meter after local processing of the monitoring data;

[0009] Update and display the tunnel model based on tunnel data including position and time;

[0010] The data upload frequency of each rangefinder within a preset time range is calculated based on the tunnel data containing position and time. Abnormal areas are marked in the tunnel model according to the data upload frequency, and the marked abnormal areas are enhanced and displayed.

[0011] Furthermore, the steps of obtaining architectural data of the subway tunnel, obtaining visual information of the subway tunnel, and creating a tunnel model based on the architectural data and the visual information include:

[0012] Obtain the BIM model of the subway tunnel as the base model;

[0013] Determine visual acquisition parameters based on the basic model; the visual acquisition parameters are parameter groups, each parameter group includes an acquisition position, an acquisition direction, and an acquisition wide angle, each parameter group corresponds to a acquisition area, and the union of the acquisition areas is not less than the inner surface of the basic model;

[0014] Acquiring visual information based on visual acquisition parameters, and identifying the visual information;

[0015] Fill the recognition result into the acquisition area corresponding to the visual acquisition parameter; the recognition result at least includes the object recognition result.

[0016] Furthermore, the step of broadcasting a data acquisition instruction based on a preset frequency and receiving tunnel data containing position and time acquired by a pre-installed rangefinder includes:

[0017] receiving a time interval input by a management party, and broadcasting a data acquisition instruction to all pre-installed rangefinders every time the time interval elapses;

[0018] Receive tunnel data containing position and time fed back by the rangefinder;

[0019] When any rangefinder receives a data acquisition instruction, it determines whether to apply the data acquisition instruction based on the application probability; the process of determining the application probability is as follows:

[0020] Receive the lookback time pre-set by the management party, obtain monitoring data in real time, calculate the mean and standard deviation of the monitoring data within the lookback time, calculate the difference between the latest monitoring data and the mean, and determine the application probability by combining the difference and standard deviation.

[0021] Furthermore, the step of updating and displaying the tunnel model according to the tunnel data including the position and time includes:

[0022] Select the latest tunnel data at each location based on time;

[0023] Query the coordinates of the distance meter corresponding to the latest tunnel data;

[0024] locating a correction point in the tunnel model based on the coordinates and the position of the tunnel data;

[0025] updating the coordinates of the correction points according to the tunnel data; wherein the updating process is a fitting process;

[0026] When the coordinates of a certain correction point are updated, the tunnel model is refreshed and displayed.

[0027] Furthermore, the steps of calculating the data upload frequency of each rangefinder within a preset time range based on the tunnel data containing the position and time, marking abnormal areas in the tunnel model based on the data upload frequency, and enhancing the display of the marked abnormal areas include:

[0028] The timeframe for receiving input from the management party;

[0029] The tunnel data is classified according to its source tag and sorted according to its time to obtain a data sequence; wherein the source tag is used to identify which distance meter uploaded the tunnel data and is inserted when the distance meter uploads the data;

[0030] Starting from the current time, retrieve the data in the data sequence within the time range in reverse order, and calculate the data upload frequency based on the number of data retrieved;

[0031] Comparing the data upload frequency with a preset frequency threshold, and when the data upload frequency reaches the preset frequency threshold, reading the position corresponding to the data within the time range;

[0032] The read positions are marked as abnormal positions, the abnormal positions are counted, the abnormal areas are obtained, and the marked abnormal areas are enhanced and displayed.

[0033] Furthermore, the distance meter includes a fixed distance meter and a mobile distance meter, and each distance meter is arranged on a base containing a steering component, and the steering component is used to adjust the center direction of the distance meter in the tunnel section.

[0034] Furthermore, the method further comprises:

[0035] Images are acquired periodically by a visual inspection device installed on the roof of the subway, and the accuracy of the tunnel model is verified based on the images. The verification method is as follows:

[0036] When acquiring images, the acquisition time, acquisition position, and acquisition direction are recorded synchronously. A simulated image is captured from the tunnel model based on the acquisition position and acquisition direction. The model image and the acquired image are compared, and the accuracy of the tunnel model is verified based on the comparison results. The acquisition time is used as a label for accuracy.

[0037] The technical solution of the present invention also provides an automatic monitoring system for subway tunnel deformation, the system comprising:

[0038] A tunnel model creation module is used to obtain architectural data of the subway tunnel, obtain visual information of the subway tunnel, and create a tunnel model based on the architectural data and visual information;

[0039] The tunnel data acquisition module is configured to broadcast data acquisition instructions based on a preset frequency and receive tunnel data containing location and time acquired by pre-installed rangefinders. Upon receiving the data acquisition instruction, any rangefinder acquires data based on independent application probabilities obtained by the rangefinder through local processing of the monitoring data.

[0040] A tunnel model updating module is used to update and display the tunnel model according to tunnel data including position and time;

[0041] The abnormal area display module is used to calculate the data upload frequency of each rangefinder within a preset time range based on the tunnel data containing position and time, mark the abnormal area in the tunnel model according to the data upload frequency, and enhance the display of the marked abnormal area.

[0042] Furthermore, the tunnel model creation module includes:

[0043] A basic model creation unit is used to obtain the BIM model of the subway tunnel as a basic model;

[0044] an acquisition parameter determination unit, configured to determine visual acquisition parameters based on the base model; the visual acquisition parameters are parameter groups, each parameter group including an acquisition position, an acquisition direction, and an acquisition wide angle; each parameter group corresponds to a acquisition area, and the union of the acquisition areas is not less than the inner surface of the base model;

[0045] a visual information recognition unit, configured to acquire visual information based on visual acquisition parameters and recognize the visual information;

[0046] The recognition result filling unit is used to fill the recognition result into the collection area corresponding to the visual collection parameter; the recognition result at least includes the object recognition result.

[0047] Furthermore, the tunnel data acquisition module includes:

[0048] an instruction broadcasting unit, configured to receive a time interval input by a management party, and broadcast a data acquisition instruction to all pre-installed rangefinders every time the time interval elapses;

[0049] A data receiving unit, used to receive tunnel data containing position and time fed back by the rangefinder;

[0050] When any rangefinder receives a data acquisition instruction, it determines whether to apply the data acquisition instruction based on the application probability; the process of determining the application probability is as follows:

[0051] Receive the lookback time pre-set by the management party, obtain monitoring data in real time, calculate the mean and standard deviation of the monitoring data within the lookback time, calculate the difference between the latest monitoring data and the mean, and determine the application probability by combining the difference and standard deviation.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The present invention installs a local recognition module on the rangefinder, which determines the data upload frequency and provides an intermittent data upload architecture, converting the real-time process into a timed process. While ensuring the validity of the data as much as possible, it greatly reduces resource consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention.

[0055] Figure 1 The overall flow chart of the subway tunnel deformation automatic monitoring method is shown.

[0056] Figure 2 The structural diagram of the subway tunnel deformation automatic monitoring system is shown. DETAILED DESCRIPTION

[0057] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] Figure 1 The following is a general flow chart of a method and system for automatically monitoring deformation of a subway tunnel. In an embodiment of the present invention, a method for automatically monitoring deformation of a subway tunnel is provided. The method includes:

[0059] Step S100: Acquire architectural data of a subway tunnel, acquire visual information of the subway tunnel, and create a tunnel model based on the architectural data and the visual information;

[0060] The subway tunnel itself is a type of building, and architectural data will be retained during the design phase. This application assumes that the management party has the authority to obtain architectural data (in fact, the execution subject of this method is a participant in the construction process, and the process of obtaining architectural data is to read data from its own database, so the authority itself is possessed). Based on the acquisition authority, the architectural data of the subway tunnel is obtained, and then the visual information of the subway tunnel is obtained. The visual information is generally obtained using an unmanned vehicle with mobile and shooting functions. The unmanned vehicle moves in the tunnel, continuously capturing images to obtain images of various locations in the subway tunnel. Combining the architectural data and visual information, a tunnel model can be created.

[0061] Step S200: broadcasting a data acquisition instruction based on a preset frequency, receiving tunnel data containing location and time acquired by pre-installed rangefinders; upon receiving the data acquisition instruction, any rangefinder acquires data based on an independent application probability; the application probability is obtained by the rangefinder after local processing of the monitoring data;

[0062] The data acquisition instruction is broadcast based on a preset frequency. Broadcasting means sending the data acquisition instruction to all rangefinders. When the rangefinder receives the data acquisition instruction, it determines whether to execute it based on the independent application probability. Assuming the application probability is 30%, when the data acquisition instruction is received, a random number is generated between 1 and 100. When the random number is between 0 and 30, it is considered to be executed, and other values are considered not to be executed. That is, there is a 30% probability of executing the data acquisition instruction and acquiring data once. When acquiring data, the rangefinder also acquires the position and time, and feeds the acquired data containing the position and time back to the execution subject of this method; the data acquired by the rangefinder is collectively referred to as tunnel data. Since the rangefinder acquires the distance, the tunnel data in this application generally refers to the distance.

[0063] Step S300: updating and displaying the tunnel model according to the tunnel data including the location and time;

[0064] After receiving the tunnel data including the position and time, the created tunnel model can be updated and the updated tunnel model can be displayed, so that the observer can observe the tunnel status intuitively and in real time.

[0065] Step S400: Calculating the data upload frequency of each rangefinder within a preset time range based on the tunnel data containing location and time, marking abnormal areas in the tunnel model based on the data upload frequency, and enhancing the display of the marked abnormal areas;

[0066] The present application also performs an additional processing on the tunnel data containing location and time. When each rangefinder uploads data, a source tag is inserted into the data. The source tag is used to identify which rangefinder uploaded the data. When the execution subject of the present method obtains the tunnel data containing location and time, the tunnel data is classified according to the source tag to obtain the data sequence corresponding to each rangefinder. The actual frequency can be calculated based on the data in the data sequence, which is called the data upload frequency. The working status of each rangefinder can be determined based on the data upload frequency (the data upload frequency is related to the actual measured data of the rangefinder, as described in step S200), and then the abnormal area is determined. After the abnormal area is determined, the abnormal area is enhanced and displayed to improve its intuitiveness.

[0067] The above process needs to be further differentiated and explained, specifically the difference between step S200 and step S400, which is explained as follows:

[0068] Suppose there is a rangefinder, which works independently, obtains monitoring data in real time, and performs independent local analysis to determine an application probability. The stronger the volatility of the monitoring data, the greater the application probability. The local working process of the rangefinder does not need to be reported to the execution subject of this method, and the execution subject of this method is unaware of it. However, the execution subject of this method will periodically send data acquisition instructions. When the rangefinder receives the data acquisition instruction, it determines whether to upload the data based on the application probability. If it is uploaded, the execution subject of this method will obtain the data sent by the rangefinder; therefore, the execution subject of this method can only intermittently obtain the data uploaded by the rangefinder. These data contain time information, and a frequency can be calculated based on the time information. The actual meaning of this frequency is that it is the application probability (approximate value) obtained by the execution subject of this method based on the acquired data. Probability reflects the volatility of the monitoring data. Therefore, based on the frequency, the fluctuation amplitude at the corresponding position of the monitored data can be determined, thereby determining the abnormal area. In this process, there is only a regular timed data upload process between the execution subject of this method and the rangefinder, and there is no real-time upload process (including the application of probability, which is not uploaded in real time), which greatly reduces the data transmission pressure. In fact, in some architectures, the rangefinder can actually upload local recognition results, that is, directly mark the abnormal area and upload it to the execution subject of this method. This method is certainly feasible, but it means that the data transmission module in the rangefinder needs to work in real time. Although it seems that the data to be transmitted is not much, the number of rangefinders is extremely large. When the number of rangefinders is extremely large, the working pressure of the processing equipment corresponding to the execution subject of this method will be very high.

[0069] Regarding step S100, the steps of obtaining architectural data of the subway tunnel, obtaining visual information of the subway tunnel, and creating a tunnel model based on the architectural data and visual information include:

[0070] Obtain the BIM model of the subway tunnel as the base model;

[0071] Determine visual acquisition parameters based on the basic model; the visual acquisition parameters are parameter groups, each parameter group includes an acquisition position, an acquisition direction, and an acquisition wide angle, each parameter group corresponds to a acquisition area, and the union of the acquisition areas is not less than the inner surface of the basic model;

[0072] Acquiring visual information based on visual acquisition parameters, and identifying the visual information;

[0073] Fill the recognition result into the acquisition area corresponding to the visual acquisition parameter; the recognition result at least includes the object recognition result.

[0074] The above content specifically defines the process of creating a tunnel model. The BIM model of the subway tunnel is known data. The BIM model of the subway tunnel is obtained as the basic model. Then, the visual acquisition parameters are determined based on the basic model. The actual meaning of the visual acquisition parameters is where to obtain the image in which direction and how wide the wide angle of the image is. The condition that needs to be met is that the acquired image needs to fully cover the inner surface of the tunnel. Visual information is obtained based on the visual acquisition parameters, and the visual information is recognized, and the recognition results are filled into the acquisition area corresponding to the visual acquisition parameters. This process can be understood as "decorating" each position in the basic model according to the visual information. The recognition results at least include object recognition results, that is, determining which objects are in the tunnel, and then inserting the objects into the corresponding positions in the basic model according to the scale of the basic model.

[0075] It is worth mentioning that if the recognition accuracy is high, the details of the basic model can be enriched to obtain a more refined tunnel model. The specific refinement process can refer to the existing three-dimensional modeling solution, and this application will not go into details.

[0076] Regarding step S200, the step of broadcasting a data acquisition instruction based on a preset frequency and receiving tunnel data containing position and time acquired by a pre-installed rangefinder includes:

[0077] receiving a time interval input by a management party, and broadcasting a data acquisition instruction to all pre-installed rangefinders every time the time interval elapses;

[0078] Receive tunnel data containing position and time fed back by the rangefinder;

[0079] The frequency of broadcasting data acquisition instructions is input by the manager, who is the tunnel manager. The manager receives the time interval input by the manager, and broadcasts the data acquisition instructions to all pre-installed distance meters every time the time interval passes, and receives the tunnel data containing the position and time fed back by each distance meter.

[0080] When any rangefinder receives a data acquisition instruction, it determines whether to apply the data acquisition instruction based on the application probability; the process of determining the application probability is as follows:

[0081] Receive the lookback time pre-set by the management party, obtain monitoring data in real time, calculate the mean and standard deviation of the monitoring data within the lookback time, calculate the difference between the latest monitoring data and the mean, and determine the application probability by combining the difference and standard deviation.

[0082] One way to determine the application probability is: Where, For application probability, and is the preset correction factor, is the absolute value of the difference, is the standard deviation; the specific description of the above determination method is as follows: the application probability is proportional to the absolute value and standard deviation of the difference, and the absolute value and standard deviation of the difference have different weights. Furthermore, the value range of the application probability is zero to one.

[0083] Regarding step S300, the step of updating and displaying the tunnel model according to the tunnel data including the location and time includes:

[0084] Select the latest tunnel data at each location based on time;

[0085] Query the coordinates of the distance meter corresponding to the latest tunnel data;

[0086] locating a correction point in the tunnel model based on the coordinates and the position of the tunnel data;

[0087] updating the coordinates of the correction points according to the tunnel data; wherein the updating process is a fitting process;

[0088] When the coordinates of a certain correction point are updated, the tunnel model is refreshed and displayed.

[0089] In one embodiment of the technical solution of the present invention, the tunnel model update and display process is specifically defined. The latest (latest) tunnel data at each different location is selected based on time. The source tag in the tunnel data determines which rangefinder it was sent from, and then the coordinates of the corresponding rangefinder are queried. The location of the tunnel data refers to a certain location in the subway tunnel. The acquired tunnel data is a distance, which refers to the distance between the coordinates of the rangefinder and a certain location in the subway tunnel. The point corresponding to the coordinates of the rangefinder and the point corresponding to a certain location in the subway tunnel are located in the tunnel model based on the scale of the tunnel model. The current distance is corrected based on the tunnel data. This process is called the update process. There is a lot of tunnel data, and multiple points in the tunnel model will be updated. Each update only updates a small area, which still needs to maintain its connection relationship with other points (to ensure that there are no breakpoints in the tunnel model). In other words, the update process is a fitting process.

[0090] Regarding step S400, the steps of calculating the data upload frequency of each rangefinder within a preset time range based on the tunnel data containing the position and time, marking abnormal areas in the tunnel model based on the data upload frequency, and enhancing the display of the marked abnormal areas include:

[0091] The timeframe for receiving input from the management party;

[0092] The tunnel data is classified according to its source tag and sorted according to its time to obtain a data sequence; wherein the source tag is used to identify which distance meter uploaded the tunnel data and is inserted when the distance meter uploads the data;

[0093] Starting from the current time, retrieve the data in the data sequence within the time range in reverse order, and calculate the data upload frequency based on the number of data retrieved;

[0094] Comparing the data upload frequency with a preset frequency threshold, and when the data upload frequency reaches the preset frequency threshold, reading the position corresponding to the data within the time range;

[0095] The read positions are marked as abnormal positions, the abnormal positions are counted, the abnormal areas are obtained, and the marked abnormal areas are enhanced and displayed.

[0096] In one example of the technical solution of the present invention, the time range input by the receiving management party may be one hour, half a day or longer. For all the tunnel data received, the tunnel data is classified according to the source label of the tunnel data, and the tunnel data from the same source is analyzed separately.

[0097] For each type of tunnel data (data obtained by the same rangefinder), the data is sorted according to its time to obtain a data sequence. Starting from the current moment, the data in the data sequence within the time range is obtained in reverse order. The data upload frequency is obtained by dividing the amount of data by the time. When the data upload frequency reaches the preset frequency threshold, it means that the average application probability of the corresponding rangefinder within the time range is very high (the application probability is calculated in real time, and the time range is relatively long. Within the time range, the rangefinder may determine multiple application probabilities in real time), indicating that the data obtained by the rangefinder is highly volatile. At this time, the location of the tunnel data within the time range is marked as an abnormal location. Finally, the abnormal locations are counted to obtain the abnormal area, and the marked abnormal area is enhanced for display.

[0098] Among them, regarding the enhanced display process, the S value and V value in the HSV color value space can be determined according to the average application probability, where the H value is a preset value, such as 120, indicating red, and the S value and V value are used to adjust the degree of red. The S value and the V value are both proportional to the average application probability. When the average application probability is greater than the preset peak value, the S value and the V value are both 1, and the result of the enhanced display is pure red.

[0099] As a preferred embodiment of the technical solution of the present invention, the types of rangefinders are limited. The rangefinders include fixed rangefinders and mobile rangefinders. Any rangefinder is set on a base containing a steering component. The steering component is used to adjust the center orientation of the rangefinder in the tunnel cross-section. The practical significance of the above architecture is that when each rangefinder is measuring distance, its measuring direction is parallel to the tunnel cross-section. The specific angle is determined by the steering component. The base of the fixed rangefinder is fixed, and the base of the mobile rangefinder is movable. Regardless of the type of rangefinder and in which direction it measures distance, the acquired tunnel data records the position, time and source tag.

[0100] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0101] Images are acquired periodically by a visual inspection device installed on the roof of the subway, and the accuracy of the tunnel model is verified based on the images. The verification method is as follows:

[0102] When acquiring images, the acquisition time, acquisition position, and acquisition direction are recorded synchronously. A simulated image is captured from the tunnel model based on the acquisition position and acquisition direction. The model image and the acquired image are compared, and the accuracy of the tunnel model is verified based on the comparison results. The acquisition time is used as a label for accuracy.

[0103] In one example of the technical solution of the present invention, a visual inspection instrument is installed on the subway to periodically acquire images (generally images in the direction of the subway's forward movement). When acquiring images, the acquisition time, acquisition position and acquisition direction are synchronously recorded. According to the acquisition position and acquisition direction, the front view angle can be determined in the tunnel model to obtain a simulated image. Both the actual acquired image and the model image actually contain the outline of the tunnel ahead. By comparing this outline (specifically, the curvature at each point), the accuracy of the tunnel model can be verified. A verification process is only a verification process for the corresponding moment, so it is necessary to insert a time tag in the accuracy. It is worth mentioning that since the acquisition position is recorded, a location tag can also be inserted in the accuracy to facilitate subsequent problem tracing.

[0104] Figure 2 The structure diagram of the automatic monitoring system for subway tunnel deformation is shown. In a preferred embodiment of the technical solution of the present invention, an automatic monitoring system for subway tunnel deformation is also provided. The system 10 includes:

[0105] a tunnel model creation module 11, configured to obtain architectural data of a subway tunnel, obtain visual information of the subway tunnel, and create a tunnel model based on the architectural data and the visual information;

[0106] The tunnel data acquisition module 12 is configured to broadcast data acquisition instructions based on a preset frequency and receive tunnel data containing location and time acquired by pre-installed rangefinders. Upon receiving the data acquisition instruction, any rangefinder acquires data based on an independent application probability obtained by the rangefinder after local processing of the monitoring data.

[0107] a tunnel model updating module 13, configured to update and display the tunnel model according to tunnel data including position and time;

[0108] The abnormal area display module 14 is used to calculate the data upload frequency of each rangefinder within a preset time range based on the tunnel data containing position and time, mark the abnormal area in the tunnel model according to the data upload frequency, and enhance the display of the marked abnormal area.

[0109] Furthermore, the tunnel model creation module 11 includes:

[0110] A basic model creation unit is used to obtain the BIM model of the subway tunnel as a basic model;

[0111] an acquisition parameter determination unit, configured to determine visual acquisition parameters based on the base model; the visual acquisition parameters are parameter groups, each parameter group including an acquisition position, an acquisition direction, and an acquisition wide angle; each parameter group corresponds to a acquisition area, and the union of the acquisition areas is not less than the inner surface of the base model;

[0112] a visual information recognition unit, configured to acquire visual information based on visual acquisition parameters and recognize the visual information;

[0113] The recognition result filling unit is used to fill the recognition result into the collection area corresponding to the visual collection parameter; the recognition result at least includes the object recognition result.

[0114] Specifically, the tunnel data acquisition module 12 includes:

[0115] an instruction broadcasting unit, configured to receive a time interval input by a management party, and broadcast a data acquisition instruction to all pre-installed rangefinders every time the time interval elapses;

[0116] A data receiving unit, used to receive tunnel data containing position and time fed back by the rangefinder;

[0117] When any rangefinder receives a data acquisition instruction, it determines whether to apply the data acquisition instruction based on the application probability; the process of determining the application probability is as follows:

[0118] Receive the lookback time pre-set by the management party, obtain monitoring data in real time, calculate the mean and standard deviation of the monitoring data within the lookback time, calculate the difference between the latest monitoring data and the mean, and determine the application probability by combining the difference and standard deviation.

[0119] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for automatically monitoring deformation of a subway tunnel, characterized in that: The method comprises: Obtain architectural data of the subway tunnel, obtain visual information of the subway tunnel, and create a tunnel model based on the architectural data and visual information; receiving a time interval input by a management party, and broadcasting a data acquisition instruction to all pre-installed rangefinders every time the time interval elapses; Receive tunnel data containing position and time fed back by the rangefinder; When any rangefinder receives a data acquisition instruction, it determines whether to apply the data acquisition instruction based on the application probability; the process of determining the application probability is as follows: Receive the lookback time pre-set by the management party, obtain monitoring data in real time, calculate the mean and standard deviation of the monitoring data within the lookback time, calculate the difference between the latest monitoring data and the mean, and combine the difference and standard deviation to determine the application probability; Update and display the tunnel model based on tunnel data including position and time; The data upload frequency of each rangefinder within a preset time range is calculated based on the tunnel data containing position and time. Abnormal areas are marked in the tunnel model according to the data upload frequency, and the marked abnormal areas are enhanced and displayed.

2. The subway tunnel deformation automatic monitoring method according to claim 1, characterized in that: The steps of obtaining architectural data of the subway tunnel, obtaining visual information of the subway tunnel, and creating a tunnel model based on the architectural data and the visual information include: Obtain the BIM model of the subway tunnel as the base model; Determine visual acquisition parameters based on the basic model; the visual acquisition parameters are parameter groups, each parameter group includes an acquisition position, an acquisition direction, and an acquisition wide angle, each parameter group corresponds to a acquisition area, and the union of the acquisition areas is not less than the inner surface of the basic model; Acquiring visual information based on visual acquisition parameters, and identifying the visual information; Fill the recognition result into the acquisition area corresponding to the visual acquisition parameter; the recognition result at least includes the object recognition result.

3. The subway tunnel deformation automatic monitoring method according to claim 1, characterized in that: The step of updating and displaying the tunnel model according to the tunnel data including the position and time includes: Select the latest tunnel data at each location based on time; Query the coordinates of the distance meter corresponding to the latest tunnel data; locating a correction point in the tunnel model based on the coordinates and the position of the tunnel data; updating the coordinates of the correction points according to the tunnel data; wherein the updating process is a fitting process; When the coordinates of a certain correction point are updated, the tunnel model is refreshed and displayed.

4. The subway tunnel deformation automatic monitoring method according to claim 3, characterized in that: The steps of calculating the data upload frequency of each rangefinder within a preset time range based on the tunnel data containing the position and time, marking abnormal areas in the tunnel model based on the data upload frequency, and enhancing the display of the marked abnormal areas include: The timeframe for receiving input from the management party; The tunnel data is classified according to its source tag and sorted according to its time to obtain a data sequence; wherein the source tag is used to identify which distance meter uploaded the tunnel data and is inserted when the distance meter uploads the data; Starting from the current time, retrieve the data in the data sequence within the time range in reverse order, and calculate the data upload frequency based on the number of data retrieved; Comparing the data upload frequency with a preset frequency threshold, and when the data upload frequency reaches the preset frequency threshold, reading the position corresponding to the data within the time range; The read positions are marked as abnormal positions, the abnormal positions are counted, the abnormal areas are obtained, and the marked abnormal areas are enhanced and displayed.

5. The subway tunnel deformation automatic monitoring method according to claim 1, characterized in that: The distance meter includes a fixed distance meter and a mobile distance meter. Any distance meter is arranged on a base containing a steering component, and the steering component is used to adjust the center direction of the distance meter in the tunnel section.

6. The subway tunnel deformation automatic monitoring method according to claim 1, characterized in that: The method further comprises: Images are acquired periodically by a visual inspection device installed on the roof of the subway, and the accuracy of the tunnel model is verified based on the images. The verification method is as follows: When acquiring images, the acquisition time, acquisition position, and acquisition direction are recorded synchronously. A simulated image is captured from the tunnel model based on the acquisition position and acquisition direction. The model image and the acquired image are compared, and the accuracy of the tunnel model is verified based on the comparison results. The acquisition time is used as a label for accuracy.

7. An automatic monitoring system for subway tunnel deformation, characterized in that: The system comprises: A tunnel model creation module is used to obtain architectural data of the subway tunnel, obtain visual information of the subway tunnel, and create a tunnel model based on the architectural data and visual information; an instruction broadcasting unit, configured to receive a time interval input by a management party, and broadcast a data acquisition instruction to all pre-installed rangefinders every time the time interval elapses; A data receiving unit, used to receive tunnel data containing position and time fed back by the rangefinder; When any rangefinder receives a data acquisition instruction, it determines whether to apply the data acquisition instruction based on the application probability; the process of determining the application probability is as follows: Receive the lookback time pre-set by the management party, obtain monitoring data in real time, calculate the mean and standard deviation of the monitoring data within the lookback time, calculate the difference between the latest monitoring data and the mean, and combine the difference and standard deviation to determine the application probability; A tunnel model updating module is used to update and display the tunnel model according to tunnel data including position and time; The abnormal area display module is used to calculate the data upload frequency of each rangefinder within a preset time range based on the tunnel data containing position and time, mark the abnormal area in the tunnel model according to the data upload frequency, and enhance the display of the marked abnormal area.

8. The subway tunnel deformation automatic monitoring system according to claim 7, characterized in that: The tunnel model creation module includes: A basic model creation unit is used to obtain the BIM model of the subway tunnel as a basic model; an acquisition parameter determination unit, configured to determine visual acquisition parameters based on the base model; the visual acquisition parameters are parameter groups, each parameter group including an acquisition position, an acquisition direction, and an acquisition wide angle; each parameter group corresponds to a acquisition area, and the union of the acquisition areas is not less than the inner surface of the base model; a visual information recognition unit, configured to acquire visual information based on visual acquisition parameters and recognize the visual information; The recognition result filling unit is used to fill the recognition result into the collection area corresponding to the visual collection parameter; the recognition result at least includes the object recognition result.

Images