A computer vision-based tunnel deformation monitoring method and system

Abstract

The application relates to the technical field of tunnel deformation monitoring, and discloses a tunnel deformation monitoring method and system based on computer vision, which comprises a determination module configured to determine the initial shooting frequency of a high-definition industrial camera based on basic parameters; a primary judgment module configured to judge whether deformation exists in the monitoring area based on analysis results; a secondary judgment module configured to judge whether the initial shooting frequency needs to be adjusted according to the initial deformation characteristic value; a processing module configured to analyze the deformation correlation of all the monitoring areas located in the edge range, determine the adjustment coefficient of the initial shooting frequency based on the deformation correlation, and obtain a final shooting frequency; and a grading module configured to determine the deformation grade of the tunnel to be monitored according to the final deformation characteristic value. The application realizes real-time monitoring of tunnel deformation, and has high monitoring precision.

Description

Technical Field

[0001] This invention relates to the field of tunnel deformation monitoring technology, and more specifically, to a tunnel deformation monitoring method and system based on computer vision. Background Technology

[0002] With the rapid development and continuous progress of modern transportation, tunnels, as a crucial infrastructure connecting various regions and ensuring smooth traffic flow, have increasingly attracted widespread attention regarding their safety performance and structural stability during actual use. To ensure that tunnels can serve the transportation system safely and sustainably for the long term, accurate and effective deformation monitoring is particularly important. However, traditional tunnel deformation monitoring methods often rely on manual operation, requiring professional technicians to regularly visit the site for on-site measurements and data collection. This monitoring method not only consumes a large amount of human resources and increases workload, but also, due to human factors and equipment limitations, often inevitably results in significant measurement errors, failing to meet the current high-standard, high-precision monitoring requirements.

[0003] Therefore, it is necessary to design a tunnel deformation monitoring method and system based on computer vision to solve the problems existing in the current technology. Summary of the Invention

[0004] In view of this, the present invention proposes a tunnel deformation monitoring method and system based on computer vision, which aims to meet the current high-standard and high-precision monitoring requirements.

[0005] In one aspect, the present invention proposes a tunnel deformation monitoring system based on computer vision, comprising:

[0006] The determination module is configured to determine the tunnel to be monitored, divide the tunnel to be monitored into regions to obtain several monitoring areas; set up a high-definition industrial camera in each monitoring area, collect the basic parameters of each monitoring area, and determine the initial shooting frequency of the high-definition industrial camera based on the basic parameters.

[0007] The primary judgment module is configured to acquire initial area image data of each monitoring area at the initial shooting frequency within a preset time interval, analyze the initial area image data, and determine whether there is deformation in the monitoring area based on the analysis result.

[0008] The secondary judgment module is configured to, when it is determined that there is deformation in the monitoring area, obtain the initial deformation feature value of the monitoring area, and determine whether to adjust the initial shooting frequency based on the initial deformation feature value;

[0009] The processing module is configured to, when it is determined that the initial shooting frequency needs to be adjusted, determine the edge range with the geometric center of the monitoring area as the center and k as the radius, analyze the deformation correlation of all monitoring areas within the edge range, determine the adjustment coefficient of the initial shooting frequency based on the deformation correlation, and obtain the final shooting frequency.

[0010] The grading module is configured to acquire final area image data of each monitoring area at the final shooting frequency within a preset time interval, and to parse all the final area image data to obtain the final deformation feature value of the tunnel to be monitored; and to determine the deformation level of the tunnel to be monitored based on the final deformation feature value.

[0011] Further, when collecting basic parameters for each monitoring area and determining the initial shooting frequency of the high-definition industrial camera based on the basic parameters, the process includes:

[0012] The basic parameters are analyzed to obtain the monitoring area, importance, and actual distance from the high-definition industrial camera to the monitoring area.

[0013] The basic shooting frequency of the high-definition industrial camera is determined based on the degree of importance.

[0014] The optimization coefficient of the basic shooting frequency is determined based on the monitored area and the actual distance;

[0015] The product of the optimization coefficient and the base shooting frequency is used as the initial shooting frequency.

[0016] Furthermore, when determining the basic shooting frequency of the high-definition industrial camera based on the aforementioned importance, the following steps are included:

[0017] The level of importance is categorized into high, medium, and low.

[0018] When the importance level is high, the basic shooting frequency is determined to be the first shooting frequency;

[0019] When the importance level is medium, the base shooting frequency is determined to be the second shooting frequency, which is lower than the first shooting frequency.

[0020] When the importance level is low, the base shooting frequency is determined to be the third shooting frequency, which is lower than the second shooting frequency.

[0021] Furthermore, when determining the optimization coefficient of the basic shooting frequency based on the monitored area and the actual distance, the following steps are included:

[0022] The monitored area is compared with the monitored area threshold, and the actual distance is compared with the actual distance threshold. The optimization coefficient is determined based on the comparison results.

[0023] When the monitored area is greater than or equal to the monitored area threshold, and the actual distance is greater than or equal to the actual distance threshold, the optimization coefficient is determined to be the first optimization coefficient.

[0024] When the monitored area is greater than or equal to the monitored area threshold, and the actual distance is less than the actual distance threshold, the optimization coefficient is determined to be the second optimization coefficient.

[0025] When the monitored area is less than the monitored area threshold and the actual distance is greater than or equal to the actual distance threshold, the optimization coefficient is determined to be the third optimization coefficient.

[0026] When the monitored area is less than the monitored area threshold and the actual distance is less than the actual distance threshold, the optimization coefficient is determined to be the fourth optimization coefficient.

[0027] Further, when analyzing the initial region image data and determining whether deformation exists in the monitored region based on the analysis results, the process includes:

[0028] The initial region image data is analyzed to obtain the deformation parameters of the monitored region, including displacement, strain value and tilt angle.

[0029] The deformation parameters are compared with a preset deformation threshold. If any one or more of the deformation parameters exceed the deformation threshold, it is determined that the monitoring area is deformed.

[0030] Otherwise, it is determined that there is no deformation in the monitored area.

[0031] Further, when determining whether to adjust the initial shooting frequency based on the initial deformation feature value, the process includes:

[0032] The initial deformation feature value is compared with the deformation standard value, and the initial shooting frequency is adjusted based on the comparison result.

[0033] When the initial deformation feature value is greater than or equal to the deformation standard value, it is determined that the initial shooting frequency should be adjusted.

[0034] Otherwise, it is determined that the initial shooting frequency will not be adjusted.

[0035] Furthermore, when analyzing the deformation correlation of all monitored areas located within the said edge range, it includes:

[0036] The deformation characteristic values ​​of all monitoring areas within the edge range are compared to determine the deformation difference between adjacent monitoring areas.

[0037] Based on the deformation difference, calculate the deformation correlation coefficient between the monitoring areas;

[0038] The deformation correlation is determined based on the deformation correlation coefficient.

[0039] Further, when determining the adjustment coefficient of the initial shooting frequency based on the deformation correlation and obtaining the final shooting frequency, the process includes:

[0040] The deformation correlation is compared with a preset correlation threshold to determine the correlation strength level.

[0041] The adjustment coefficient for the initial shooting frequency is determined based on the correlation strength level.

[0042] The product of the adjustment coefficient and the initial shooting frequency is taken as the final shooting frequency.

[0043] Further, when determining the deformation level of the tunnel to be monitored based on the final deformation characteristic value, the process includes:

[0044] The final deformation characteristic value is compared with the first deformation value and the second deformation value, and the deformation level of the tunnel to be monitored is determined according to the comparison result; wherein, the first deformation value is less than the second deformation value;

[0045] When the final deformation characteristic value is less than or equal to the first deformation value, the deformation level of the tunnel to be monitored is determined to be the first level.

[0046] When the final deformation characteristic value is greater than the first deformation value and less than or equal to the second deformation value, the deformation level is determined to be the second level.

[0047] When the final deformation characteristic value is greater than the second deformation value, the deformation level is determined to be the third level.

[0048] Compared with existing technologies, the advantages of this invention are as follows: The computer vision-based tunnel deformation monitoring system provided by this invention can achieve real-time monitoring of tunnel deformation with high accuracy. By dividing the tunnel into regions and installing high-definition industrial cameras in each monitoring region, comprehensive monitoring of all areas of the tunnel can be achieved. Simultaneously, by determining the initial shooting frequency of the high-definition industrial cameras based on the basic parameters of the monitoring regions, unnecessary shooting can be reduced while ensuring monitoring accuracy, thus improving efficiency. When deformation is initially determined in the monitoring region, initial deformation feature values ​​are further obtained, and the initial shooting frequency is adjusted based on these values ​​to ensure accurate capture of the deformation. Furthermore, by analyzing the deformation correlation of all monitoring regions within the edge range, the adjustment coefficient of the initial shooting frequency can be determined more accurately, resulting in a more suitable final shooting frequency. Finally, the deformation level of the tunnel under monitoring is determined based on the final deformation feature values, providing an important basis for tunnel maintenance and management.

[0049] In another aspect, the present invention also proposes a tunnel deformation monitoring method based on computer vision, comprising the following steps:

[0050] The tunnel to be monitored is identified, and the tunnel to be monitored is divided into regions to obtain several monitoring areas;

[0051] A high-definition industrial camera is installed in each of the monitoring areas to collect basic parameters of each monitoring area, and the initial shooting frequency of the high-definition industrial camera is determined based on the basic parameters.

[0052] Within a preset time interval, initial area image data of each monitoring area is acquired at the initial shooting frequency, and the initial area image data is analyzed to determine whether there is deformation in the monitoring area based on the analysis results.

[0053] When it is determined that there is deformation in the monitoring area, the initial deformation feature value of the monitoring area is obtained, and it is determined whether to adjust the initial shooting frequency based on the initial deformation feature value;

[0054] When it is determined that the initial shooting frequency needs to be adjusted, the edge range is determined with the geometric center of the monitoring area as the center and k as the radius. The deformation correlation of all monitoring areas within the edge range is analyzed. Based on the deformation correlation, the adjustment coefficient of the initial shooting frequency is determined, and the final shooting frequency is obtained.

[0055] Within a preset time interval, final area image data of each monitoring area is acquired at the final shooting frequency, and all the final area image data are analyzed to obtain the final deformation characteristic value of the tunnel to be monitored.

[0056] The deformation level of the tunnel to be monitored is determined based on the final deformation characteristic value.

[0057] It is understandable that the above-mentioned computer vision-based tunnel deformation monitoring methods and systems have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0058] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0059] Figure 1 A structural block diagram of a computer vision-based tunnel deformation monitoring system provided in an embodiment of the present invention;

[0060] Figure 2 A flowchart of a computer vision-based tunnel deformation monitoring method provided in an embodiment of the present invention. Detailed Implementation

[0061] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0062] See Figure 1 As shown in some embodiments of this application, this embodiment provides a tunnel deformation monitoring system based on computer vision, including:

[0063] The determination module is configured to determine the tunnel to be monitored, divide the tunnel to be monitored into regions to obtain several monitoring areas; set up a high-definition industrial camera in each monitoring area, collect the basic parameters of each monitoring area, and determine the initial shooting frequency of the high-definition industrial camera based on the basic parameters.

[0064] The primary judgment module is configured to acquire initial area image data of each monitoring area at the initial shooting frequency within a preset time interval, analyze the initial area image data, and determine whether there is deformation in the monitoring area based on the analysis result.

[0065] The secondary judgment module is configured to, when it is determined that there is deformation in the monitoring area, obtain the initial deformation feature value of the monitoring area, and determine whether to adjust the initial shooting frequency based on the initial deformation feature value;

[0066] The processing module is configured to, when it is determined that the initial shooting frequency needs to be adjusted, determine the edge range with the geometric center of the monitoring area as the center and k as the radius, analyze the deformation correlation of all monitoring areas within the edge range, determine the adjustment coefficient of the initial shooting frequency based on the deformation correlation, and obtain the final shooting frequency.

[0067] The grading module is configured to acquire final area image data of each monitoring area at the final shooting frequency within a preset time interval, and to parse all the final area image data to obtain the final deformation feature value of the tunnel to be monitored; and to determine the deformation level of the tunnel to be monitored based on the final deformation feature value.

[0068] In this embodiment, k is preferably one-tenth of the tunnel length to ensure that the deformation of the tunnel can be captured comprehensively and accurately.

[0069] It is understood that the computer vision-based tunnel deformation monitoring system provided in this embodiment can achieve real-time monitoring of tunnel deformation with high accuracy. By dividing the tunnel into regions and installing high-definition industrial cameras in each monitoring region, comprehensive monitoring of all areas of the tunnel can be achieved. Simultaneously, determining the initial shooting frequency of the high-definition industrial cameras based on the basic parameters of the monitoring regions can reduce unnecessary shooting and improve efficiency while ensuring monitoring accuracy. When deformation is initially determined in a monitoring region, initial deformation feature values ​​are further obtained, and the initial shooting frequency is adjusted based on these values ​​to ensure accurate capture of the deformation. Furthermore, by analyzing the deformation correlation of all monitoring regions within the edge range, the adjustment coefficient of the initial shooting frequency can be determined more accurately, resulting in a more suitable final shooting frequency. Finally, the deformation level of the tunnel under monitoring is determined based on the final deformation feature values, providing an important basis for tunnel maintenance and management.

[0070] Specifically, when collecting basic parameters for each monitoring area and determining the initial shooting frequency of the high-definition industrial camera based on the basic parameters, the process includes:

[0071] The basic parameters are analyzed to obtain the monitoring area, importance, and actual distance from the high-definition industrial camera to the monitoring area.

[0072] The basic shooting frequency of the high-definition industrial camera is determined based on the degree of importance.

[0073] The optimization coefficient of the basic shooting frequency is determined based on the monitored area and the actual distance;

[0074] The product of the optimization coefficient and the base shooting frequency is used as the initial shooting frequency.

[0075] Understandably, the higher the importance of the monitored area, the higher the base shooting frequency should be set to ensure more frequent monitoring of that area. The larger the monitored area, or the greater the actual distance between the high-definition industrial camera and the monitored area, the higher the optimization coefficient should be set to increase the shooting frequency and capture more detailed changes. This design allows for flexible adjustment of the high-definition industrial camera's shooting frequency according to actual monitoring needs, ensuring both comprehensive monitoring and improved efficiency. Furthermore, optimizing the initial shooting frequency can extend the lifespan of the high-definition industrial camera to some extent and reduce monitoring costs.

[0076] Specifically, determining the base shooting frequency of the high-definition industrial camera based on the aforementioned importance includes:

[0077] The level of importance is categorized into high, medium, and low.

[0078] When the importance level is high, the basic shooting frequency is determined to be the first shooting frequency;

[0079] When the importance level is medium, the base shooting frequency is determined to be the second shooting frequency, which is lower than the first shooting frequency.

[0080] When the importance level is low, the base shooting frequency is determined to be the third shooting frequency, which is lower than the second shooting frequency.

[0081] Understandably, setting different base shooting frequencies for monitoring areas of varying importance allows for a more rational allocation of monitoring resources. For areas of higher importance, such as critical structural components of tunnels or areas with significant historical deformation, a higher shooting frequency enables timely detection and response to deformation, ensuring tunnel safety. Conversely, for less important areas, appropriately lowering the shooting frequency reduces data volume and processing burden while maintaining a certain level of monitoring accuracy, thus improving the overall efficiency of the system.

[0082] Specifically, when determining the optimization coefficient of the basic shooting frequency based on the monitored area and the actual distance, the following steps are included:

[0083] The monitored area is compared with the monitored area threshold, and the actual distance is compared with the actual distance threshold. The optimization coefficient is determined based on the comparison results.

[0084] When the monitored area is greater than or equal to the monitored area threshold, and the actual distance is greater than or equal to the actual distance threshold, the optimization coefficient is determined to be the first optimization coefficient.

[0085] When the monitored area is greater than or equal to the monitored area threshold, and the actual distance is less than the actual distance threshold, the optimization coefficient is determined to be the second optimization coefficient.

[0086] When the monitored area is less than the monitored area threshold and the actual distance is greater than or equal to the actual distance threshold, the optimization coefficient is determined to be the third optimization coefficient.

[0087] When the monitored area is less than the monitored area threshold and the actual distance is less than the actual distance threshold, the optimization coefficient is determined to be the fourth optimization coefficient.

[0088] Understandably, the first optimization coefficient is preferably set to its highest value to ensure that more detailed changes can be captured when the monitoring area is large and the distance is far. The second optimization coefficient is slightly lower than the first and is suitable for situations where the monitoring area is large but the distance is relatively short. In this case, although the distance is short, the large monitoring area still requires maintaining a high shooting frequency. The third optimization coefficient is suitable for situations where the monitoring area is small but the distance is far. In this case, due to the long distance, the shooting frequency needs to be increased to capture details, but because the monitoring area is small, the frequency can be relatively low. The fourth optimization coefficient is the lowest value and is suitable for situations where the monitoring area is small and the distance is close. In this case, detailed changes are easily captured, and the shooting frequency can be appropriately reduced. Through such optimization coefficient settings, the shooting frequency of the high-definition industrial camera can be adjusted more precisely, ensuring both the comprehensiveness and accuracy of monitoring and improving the overall efficiency of the system.

[0089] Specifically, when analyzing the initial region image data and determining whether the monitored region exhibits deformation based on the analysis results, the process includes:

[0090] The initial region image data is analyzed to obtain the deformation parameters of the monitored region, including displacement, strain value and tilt angle.

[0091] The deformation parameters are compared with a preset deformation threshold. If any one or more of the deformation parameters exceed the deformation threshold, it is determined that the monitoring area is deformed.

[0092] Otherwise, it is determined that there is no deformation in the monitored area.

[0093] Understandably, deformation parameters are obtained by analyzing the initial area image data using image processing tools. These tools can identify feature points in the image and calculate deformation parameters such as displacement, strain, and tilt angle based on changes in these feature points. These deformation parameters directly reflect the deformation status of the monitored area. If any one or more of the deformation parameters exceed the deformation threshold, it indicates a potential safety hazard in the monitored area, requiring timely further monitoring and analysis. This design enables rapid response to tunnel deformation, providing strong support for tunnel maintenance and management.

[0094] Specifically, determining whether to adjust the initial shooting frequency based on the initial deformation feature value includes:

[0095] The initial deformation feature value is compared with the deformation standard value, and the initial shooting frequency is adjusted based on the comparison result.

[0096] When the initial deformation feature value is greater than or equal to the deformation standard value, it is determined that the initial shooting frequency should be adjusted.

[0097] Otherwise, it is determined that the initial shooting frequency will not be adjusted.

[0098] In this embodiment, the initial deformation characteristic value is preferably one of the average or maximum value of the displacement, strain, and tilt angle.

[0099] Understandably, when the initial deformation characteristic value reaches or exceeds the preset deformation standard value, it means that the deformation in the monitoring area may be more serious, and more intensive monitoring is needed to capture deformation details. Therefore, the initial shooting frequency needs to be adjusted to improve the accuracy and timeliness of monitoring.

[0100] Specifically, analyzing the deformation correlation of all monitored areas within the edge range includes:

[0101] The deformation characteristic values ​​of all monitoring areas within the edge range are compared to determine the deformation difference between adjacent monitoring areas.

[0102] Based on the deformation difference, calculate the deformation correlation coefficient between the monitoring areas;

[0103] The deformation correlation is determined based on the deformation correlation coefficient.

[0104] Understandably, in calculating the deformation correlation coefficient, the deformation characteristic values ​​of all monitoring areas within the stated edge range are first obtained, and the deformation characteristic values ​​between any two adjacent monitoring areas are compared to obtain their deformation difference. The deformation difference can be expressed as the absolute value or weighted distance of parameters such as the difference in displacement, strain value, or tilt angle between adjacent areas. Then, based on the set of deformation difference values ​​for all adjacent areas, the deformation correlation coefficient is calculated using a normalized average or correlation function model. Specifically, the deformation correlation coefficient can be calculated using the following formula:

[0105]

[0106] Where ρ represents the deformation correlation coefficient, with a value range of [0,1]; Δi is the deformation difference degree of the i-th pair of adjacent monitoring areas; n is the total number of adjacent area pairs within the edge range; and θ is the preset maximum allowable deformation difference threshold.

[0107] Understandably, when determining deformation correlation based on the deformation correlation coefficient, if the coefficient ρ is close to 1, it indicates a high degree of consistency in deformation among monitoring areas within the edge range, suggesting a strong deformation correlation. This may mean that these areas are jointly affected by some external factor or internal stress, leading to similar deformation patterns. In this case, the system can further analyze the causes of these common deformations, providing more specific guidance for tunnel maintenance and management. Conversely, if the deformation correlation coefficient ρ is low, close to 0, it indicates a significant difference in deformation among monitoring areas within the edge range, suggesting a weak deformation correlation. This may mean that the deformation in these areas is caused by different factors or stresses, or that the deformation is spatially unevenly distributed. In this case, the system needs to conduct in-depth analysis of each monitoring area separately to determine the specific causes of deformation and take corresponding maintenance measures.

[0108] In this embodiment, each deformation correlation coefficient corresponds to a deformation correlation. For example, when the deformation correlation coefficient is 0.3, the deformation correlation is determined to be 30; when the deformation correlation coefficient is 0.5, the deformation correlation is determined to be 50.

[0109] Specifically, when determining the adjustment coefficient of the initial shooting frequency based on the deformation correlation and obtaining the final shooting frequency, the process includes:

[0110] The deformation correlation is compared with a preset correlation threshold to determine the correlation strength level.

[0111] The adjustment coefficient for the initial shooting frequency is determined based on the correlation strength level.

[0112] The product of the adjustment coefficient and the initial shooting frequency is taken as the final shooting frequency.

[0113] Understandably, stronger deformation correlation means more similar deformation patterns across monitoring areas, potentially requiring more intensive monitoring to capture underlying deformation details; therefore, the adjustment coefficient can be set relatively high. Conversely, weaker deformation correlation allows for a relatively lower adjustment coefficient, as deformation patterns in each monitoring area may be relatively independent, necessitating less intensive monitoring. This adjustment coefficient setting allows for a more rational allocation of monitoring resources, ensuring both comprehensiveness and accuracy while avoiding unnecessary resource waste.

[0114] Specifically, determining the deformation level of the tunnel to be monitored based on the final deformation characteristic value includes:

[0115] The final deformation characteristic value is compared with the first deformation value and the second deformation value, and the deformation level of the tunnel to be monitored is determined according to the comparison result; wherein, the first deformation value is less than the second deformation value;

[0116] When the final deformation characteristic value is less than or equal to the first deformation value, the deformation level of the tunnel to be monitored is determined to be the first level.

[0117] When the final deformation characteristic value is greater than the first deformation value and less than or equal to the second deformation value, the deformation level is determined to be the second level.

[0118] When the final deformation characteristic value is greater than the second deformation value, the deformation level is determined to be the third level.

[0119] In this embodiment, the final deformation feature value is obtained by parsing all the final region image data and then performing a weighted summation of all the obtained deformation feature values.

[0120] In this embodiment, the deformation characteristic value is preferably a comprehensive evaluation value obtained by weighted summation after normalizing the displacement, strain, and tilt angle. This comprehensive evaluation value can fully reflect the deformation of the tunnel.

[0121] Understandably, Level 1 indicates minor tunnel deformation, possibly involving small deformations or displacements that do not significantly impact the tunnel's structure and safety. In this case, routine maintenance and monitoring measures can be recommended to ensure normal tunnel operation. Level 2 indicates moderate tunnel deformation, which may have already affected the tunnel's structure but has not yet reached a dangerous level. For this situation, enhanced monitoring and analysis of the tunnel are necessary, along with timely maintenance measures to prevent further deformation. Level 3 is the most severe, indicating severe tunnel deformation that may pose a serious threat to the tunnel's structure and safety. In this case, immediate emergency measures are required, including a comprehensive inspection and repair of the tunnel to ensure its safe and stable operation. This deformation level classification provides more specific and targeted guidance for tunnel maintenance and management, ensuring tunnel safety and reliability.

[0122] See Figure 2 As shown in some embodiments of this application, this embodiment provides a tunnel deformation monitoring method based on computer vision, including the following steps:

[0123] S100: Determine the tunnel to be monitored, divide the tunnel to be monitored into regions, and obtain several monitoring areas;

[0124] S200: A high-definition industrial camera is installed in each of the monitoring areas to collect basic parameters of each monitoring area, and the initial shooting frequency of the high-definition industrial camera is determined based on the basic parameters.

[0125] S300: Within a preset time interval, initial area image data of each monitoring area is acquired at the initial shooting frequency, and the initial area image data is analyzed to determine whether there is deformation in the monitoring area based on the analysis results;

[0126] S400: When it is determined that there is deformation in the monitoring area, the initial deformation feature value of the monitoring area is obtained, and it is determined whether to adjust the initial shooting frequency based on the initial deformation feature value;

[0127] S500: When it is determined that the initial shooting frequency needs to be adjusted, the edge range is determined with the geometric center of the monitoring area as the center and k as the radius. The deformation correlation of all monitoring areas within the edge range is analyzed. Based on the deformation correlation, the adjustment coefficient of the initial shooting frequency is determined, and the final shooting frequency is obtained.

[0128] S600: Within a preset time interval, acquire final area image data of each monitoring area at the final shooting frequency, and analyze all the final area image data to obtain the final deformation feature value of the tunnel to be monitored.

[0129] S700: Determine the deformation level of the tunnel to be monitored based on the final deformation characteristic value.

[0130] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0131] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0132] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0133] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A tunnel deformation monitoring system based on computer vision, characterized in that, include: The determination module is configured to determine the tunnel to be monitored, divide the tunnel to be monitored into regions, and obtain several monitoring regions; A high-definition industrial camera is installed in each of the monitoring areas to collect basic parameters of each monitoring area, and the initial shooting frequency of the high-definition industrial camera is determined based on the basic parameters. The primary judgment module is configured to acquire initial area image data of each monitoring area at the initial shooting frequency within a preset time interval, analyze the initial area image data, and determine whether there is deformation in the monitoring area based on the analysis result. The secondary judgment module is configured to, when it is determined that there is deformation in the monitoring area, obtain the initial deformation feature value of the monitoring area, and determine whether to adjust the initial shooting frequency based on the initial deformation feature value; The processing module is configured to, when it is determined that the initial shooting frequency needs to be adjusted, determine the edge range with the geometric center of the monitoring area as the center and k as the radius, analyze the deformation correlation of all monitoring areas within the edge range, determine the adjustment coefficient of the initial shooting frequency based on the deformation correlation, and obtain the final shooting frequency. The grading module is configured to acquire final area image data of each monitoring area at the final shooting frequency within a preset time interval, and to analyze all the final area image data to obtain the final deformation feature value of the tunnel to be monitored. The deformation level of the tunnel to be monitored is determined based on the final deformation characteristic value; When analyzing the deformation correlation of all monitored areas within the said edge range, the following is included: The deformation characteristic values ​​of all monitoring areas within the edge range are compared to determine the deformation difference between adjacent monitoring areas. Based on the deformation difference, calculate the deformation correlation coefficient between the monitoring areas; The deformation correlation is determined based on the deformation correlation coefficient; When determining the adjustment coefficient of the initial shooting frequency based on the deformation correlation and obtaining the final shooting frequency, the process includes: The deformation correlation is compared with a preset correlation threshold to determine the correlation strength level. The adjustment coefficient for the initial shooting frequency is determined based on the correlation strength level. The product of the adjustment coefficient and the initial shooting frequency is taken as the final shooting frequency.

2. The tunnel deformation monitoring system based on computer vision according to claim 1, characterized in that, When collecting basic parameters for each monitoring area and determining the initial shooting frequency of the high-definition industrial camera based on the basic parameters, the process includes: The basic parameters are analyzed to obtain the monitoring area, importance, and actual distance from the high-definition industrial camera to the monitoring area. The basic shooting frequency of the high-definition industrial camera is determined based on the degree of importance. The optimization coefficient of the basic shooting frequency is determined based on the monitored area and the actual distance; The product of the optimization coefficient and the base shooting frequency is used as the initial shooting frequency.

3. The tunnel deformation monitoring system based on computer vision according to claim 2, characterized in that, When determining the base shooting frequency of the high-definition industrial camera based on the aforementioned importance, the following are included: The level of importance is categorized into high, medium, and low. When the importance level is high, the basic shooting frequency is determined to be the first shooting frequency; When the importance level is medium, the base shooting frequency is determined to be the second shooting frequency, which is lower than the first shooting frequency. When the importance level is low, the base shooting frequency is determined to be the third shooting frequency, which is lower than the second shooting frequency.

4. The tunnel deformation monitoring system based on computer vision according to claim 3, characterized in that, When determining the optimization coefficient of the basic shooting frequency based on the monitored area and the actual distance, the following are included: The monitored area is compared with the monitored area threshold, and the actual distance is compared with the actual distance threshold. The optimization coefficient is determined based on the comparison results. When the monitored area is greater than or equal to the monitored area threshold, and the actual distance is greater than or equal to the actual distance threshold, the optimization coefficient is determined to be the first optimization coefficient. When the monitored area is greater than or equal to the monitored area threshold, and the actual distance is less than the actual distance threshold, the optimization coefficient is determined to be the second optimization coefficient. When the monitored area is less than the monitored area threshold and the actual distance is greater than or equal to the actual distance threshold, the optimization coefficient is determined to be the third optimization coefficient. When the monitored area is less than the monitored area threshold and the actual distance is less than the actual distance threshold, the optimization coefficient is determined to be the fourth optimization coefficient.

5. The tunnel deformation monitoring system based on computer vision according to claim 4, characterized in that, When analyzing the initial region image data and determining whether deformation exists in the monitored region based on the analysis results, the process includes: The initial region image data is analyzed to obtain the deformation parameters of the monitored region, including displacement, strain value and tilt angle. The deformation parameters are compared with a preset deformation threshold. If any one or more of the deformation parameters exceed the deformation threshold, it is determined that the monitoring area is deformed. Otherwise, it is determined that there is no deformation in the monitored area.

6. The tunnel deformation monitoring system based on computer vision according to claim 5, characterized in that, When determining whether to adjust the initial shooting frequency based on the initial deformation feature value, the following are included: The initial deformation feature value is compared with the deformation standard value, and the initial shooting frequency is adjusted based on the comparison result. When the initial deformation feature value is greater than or equal to the deformation standard value, it is determined that the initial shooting frequency should be adjusted. Otherwise, it is determined that the initial shooting frequency will not be adjusted.

7. The tunnel deformation monitoring system based on computer vision according to claim 6, characterized in that, Determining the deformation level of the tunnel to be monitored based on the final deformation characteristic value includes: The final deformation characteristic value is compared with the first deformation value and the second deformation value, and the deformation level of the tunnel to be monitored is determined according to the comparison result; wherein, the first deformation value is less than the second deformation value; When the final deformation characteristic value is less than or equal to the first deformation value, the deformation level of the tunnel to be monitored is determined to be the first level. When the final deformation characteristic value is greater than the first deformation value and less than or equal to the second deformation value, the deformation level is determined to be the second level. When the final deformation characteristic value is greater than the second deformation value, the deformation level is determined to be the third level.

8. A computer vision-based tunnel deformation monitoring method, applied to the computer vision-based tunnel deformation monitoring system as described in any one of claims 1-7, characterized in that, include: The tunnel to be monitored is identified, and the tunnel to be monitored is divided into regions to obtain several monitoring areas; A high-definition industrial camera is installed in each of the monitoring areas to collect basic parameters of each monitoring area, and the initial shooting frequency of the high-definition industrial camera is determined based on the basic parameters. Within a preset time interval, initial area image data of each monitoring area is acquired at the initial shooting frequency, and the initial area image data is analyzed to determine whether there is deformation in the monitoring area based on the analysis results. When it is determined that there is deformation in the monitoring area, the initial deformation feature value of the monitoring area is obtained, and it is determined whether to adjust the initial shooting frequency based on the initial deformation feature value; When it is determined that the initial shooting frequency needs to be adjusted, the edge range is determined with the geometric center of the monitoring area as the center and k as the radius. The deformation correlation of all monitoring areas within the edge range is analyzed. Based on the deformation correlation, the adjustment coefficient of the initial shooting frequency is determined, and the final shooting frequency is obtained. Within a preset time interval, final area image data of each monitoring area is acquired at the final shooting frequency, and all the final area image data are analyzed to obtain the final deformation characteristic value of the tunnel to be monitored. The deformation level of the tunnel to be monitored is determined based on the final deformation characteristic value.

Images